Can Sci-Tech Finance Policy Boost Corporate ESG Performance? Evidence from the Pilot Experiment of Promoting the Integration of Technology and Finance in China

Abstract

Based on the quasi-natural experiment of “the pilot policy of combining science and technology with finance” (Sci-Tech Finance pilot policy) carried out in China in recent years, this paper constructs a multi-stage difference-in-differences model to explore its impact on corporate ESG performance and the influence mechanisms. The main research findings of this paper are as follows: (1) The Sci-Tech Finance pilot policy significantly enhances corporate ESG performance, a finding that remains consistent after conducting parallel trends testing, propensity score matching, and placebo tests. (2) The policy promotes the corporate ESG performance through three intermediary channels, namely alleviating financial constraints, improving total factor productivity, and enhancing green technology innovation. Notably, the first two intermediary channels exhibit the most prominent effects. (3) The impact of the pilot policy on the corporate ESG performance exhibits heterogeneity at both the regional and corporate levels; it demonstrates a more pronounced impact on corporates located in the Eastern Region, within high digital economic zones, and among high-tech, capital-intensive, heavily polluting, and state-owned corporates. (4) The policy has apparent spatial spillover effects on corporate ESG performance, accounting for about 8% of the direct effect in the pilot areas. This study enriches the literature on the impacts of Sci-Tech Finance on corporate behaviors, providing insights for government regulatory authorities to leverage Sci-Tech Finance policies to promote corporate ESG performance and sustainable development.

1. Introduction

The ESG concept, as an investment philosophy and corporate evaluation approach in the sustainable development framework, has garnered increasing attention in recent years. It not only pays attention to a company’s internal governance and operational performance but also emphasizes its environmental governance and social responsibility performance. In the context of the era of green and sustainable development, ESG standards are gradually becoming important guidelines for the development goals of enterprises in various countries, and improving corporates’ ESG performance is also regarded as the micro-foundation for promoting social sustainable development (Gillan et al., 2021) [1]. With China’s green transformation process and the enhancement of corporate social responsibility awareness (Tan et al., 2024) [2], an increasing number of Chinese companies are responding to and practicing the concept of ESG development, actively integrating into the international ESG standard system. According to data provided by the International Institute of Green Finance, Central University of Finance and Economics, the proportion of China’s A-share listed companies issuing independent ESG reports increased from 24.64% to 33.93% between 2021 and 2023.Although the coverage of corporate ESG disclosures in China is continuously expanding, the overall ESG ratings are still relatively low compared to international standards, and the investment, sustainability, and evaluation scores of Chinese enterprises in the ESG field still need to be improved.

The ESG activities of enterprises cannot be separated from support from the financial market. Consistent external financing and diversified financial services are pivotal in advancing non-profit endeavors such as corporate social responsibility and environmental governance. Furthermore, technological advancements play a pivotal role in corporate ESG practices, facilitating the resolution of environmental, social, and governance challenges, and enabling corporations to achieve their ESG goals. As modern financial service scenarios and outreach expand, financial support dedicated to technological innovation diversifies, drawing increasing attention to the concept of “Sci-Tech Finance” (Sheng et al., 2021 [3]; Lu et al., 2022 [4]). This focuses on the integration of financial services into technology-supported sectors, aiming to provide an efficient financial environment for technological innovation and development (Li et al., 2019 [5]). In October 2011, the Chinese government selected 41 cities, including Beijing and Tianjin, etc., to carry out the policy of “promoting the integration of science and technology and finance”(referred to as the “Sci-Tech Finance Pilot”). Subsequently, in 2016, China launched the second batch of the Sci-Tech Finance pilot policy, selecting an additional nine cities, including Shenyang, Zhengzhou, Ningbo, Xiamen, Jinan, and Nanchang, among others. The policy encompasses innovating the Sci-Tech Finance market system, establishing and enhancing a diversified technology investment and financing system, and accelerating the development of Sci-Tech insurance and the Sci-Tech finance service system, all with the aim of providing robust support for regional innovation and high-quality development.

The development of China’s Sci-Tech Finance undoubtedly plays a pivotal role in advancing technological innovation practices within enterprises and promoting their sustainable high-quality development. As ESG performance has become a key indicator for measuring the sustainability of corporate development in the new era, the focus of this paper lies in the following question: has the implementation of China’s Sci-Tech Finance pilot policy contributed to the enhancement of corporates ESG performance? The answer to this question not only provides insights into the impact of Sci-Tech Finance policy on the green development and ESG behaviors of enterprises but also offers crucial empirical enlightenment for facilitating the integration of Sci-Tech Finance into corporate ESG governance practices.

The main content and contributions of this paper are as follows. Firstly, by incorporating China’s city-level Sci-Tech Finance pilot policy and micro-level corporate ESG activities into a unified framework, this paper explores the impact mechanism and actual effects of Sci-Tech Finance policy on corporate ESG performance, offering new evidence for and insight into the influence of Sci-Tech Finance on corporate behaviors. Secondly, this paper elucidates the channels through which the Sci-Tech Finance pilot policy affects ESG performance, mainly from the perspectives of alleviating the financing constraints of enterprises, fostering green innovation, and enhancing corporate total factor productivity, and clarifies the primary intermediary paths involved. Thirdly, this paper discusses the positive impact, heterogeneous influence, and spatial spillover effect of Sci-Tech Finance pilot policy on enhancing ESG performance, which provides support for the interactive integration and policy synergy practice of Sci-Tech Finance and corporate ESG governance.

The following paper is arranged as follows: Part 2 reviews the literature; Part 3 presents the theoretical foundation and research hypotheses of the whole paper; Part 4 explains the empirical design and data sources; Part 5 and Part 6 are dedicated to empirical analysis and result presentation; Part 7 comes up with the conclusion of the paper. Figure 1 provides a visual presentation and explanation of the framework of this paper.

2. Literature Review

2.1. Influencing Factors of Corporate ESG

The ESG concept focuses on three dimensions: environment (E), social responsibility (S), and governance (G), emphasizing that companies should not only pursue economic benefits but also actively fulfill their obligations in terms of environmental protection and social responsibility practice. With the growing emphasis on sustainable development, corporate ESG performance has become a focal point for both investors and academic researchers (Gillan et al., 2021 [1]; Dorfleitner et al., 2015 [6]).

Many researchers have explored the factors influencing corporate ESG activities, with scholars examining both internal corporate behaviors and external factors. From the perspective of internal factors, Li and Pang (2023) [7] highlighted that green technological innovation significantly promotes corporate ESG performance. Fang et al. (2023) [8] suggested that digital transformation can significantly improve ESG performance by reducing agency costs and enhancing corporate reputation. Moreover, organizational mechanisms within firms are closely related to their ESG performance; the independence of the board, management transparency, and the protection of shareholder rights have significant effects on corporate ESG behavior (Rossi et al., 2021 [9]; Barko et al., 2022 [10]). Some studies investigated the influence of executive characteristics on corporate ESG performance, highlighting significant impacts from factors like gender and age (Liu et al., 2024 [11]; Al-Shaer et al., 2024 [12]).

From the perspective of external factors, social supervision plays a crucial role in influencing corporate ESG performance. Zhang et al. (2023) [13] argued that analyst attention can enhance corporate ESG performance. Dyck et al. (2019) [14] found that institutional investors, driven by both financial and social returns, can incentivize firms to improve their ESG practices. In the realm of social auditing, auditors’ reviews of corporate sustainability reports contribute to improved ESG performance (Del and Rigamonti, 2020 [15]). Additionally, ecological environments have been found to influence corporate ESG behavior (Zhao et al., 2024 [16]).

Financial activities are an essential component of corporate development, and the impact of financial markets and financial environments on corporate ESG has been widely discussed in the literature. Studies indicate that an efficient financial environment provides strong support for external financing, facilitating investments in and implementation of ESG initiatives (Ng et al., 2020 [17], Guo et al., 2024 [18]). Zhang (2023) [19] highlighted that the development of green finance market encourages enterprises to undertake green projects, thus incentivizing enterprises to engage in ESG activities. Ma et al. (2024) [20] argued that green credit policies, through optimizing credit resource allocation, raising managerial awareness of environmental issues, and increasing R&D investment, contribute to the enhancement of corporate ESG performance. Furthermore, the development of financial markets and continuous innovation in financial services have fueled the growth of sustainable investment products, represented by ESG funds (Busch et al., 2016) [21], which further incentivize corporate engagement in ESG activities.

2.2. The Impact Effects of Sci-Tech Finance Policy

Enhancing Sci-Tech Finance is a pivotal initiative undertaken by the Chinese government to advance the construction of a modern financial system and support enterprise innovation and high-quality development (Jiang et al., 2022 [22]; Lu et al., 2024 [23]). Following the introduction of corresponding policy, several studies have explored the economic and social effects of the Sci-Tech Finance activities and related policies.

Existing research indicates that Sci-Tech Finance contributes to creating an innovative market environment and facilitates the inflow of external capital into innovative sectors (Manta, 2024) [24]. High-tech industries benefit more significantly from the Sci-Tech Finance policy (Liu and Zhang, 2023) [25]. Additionally, literature studies suggest that the advancement of the Sci-Tech Finance policy enhances industrial total factor productivity (Li et al., 2024) [5] and promotes the efficiency of regional financial activities and technological progress (Su, 2023) [26]. Furthermore, some studies point out that the implementation of the Sci-Tech Finance policy has an inhibitory effect on regional carbon emissions, thereby promoting the development of low-carbon economy (Li et al., 2022) [27].

At the micro level, the Sci-Tech Finance policy has far-reaching impacts on corporate activities. Sci-Tech Finance provides full-life-cycle financial services for technology innovation-oriented enterprises and projects, which not only accelerates the transformation of technological achievements into productivity, promoting technological upgrading within enterprises, but also alleviates the issues of high financing costs and difficulties in R&D investment during innovative activities, thereby reducing corporate financial risks (Liu and Zhang, 2023) [25]. With the advancement of the Sci-Tech Finance policy, Sci-Tech Finance market and services have continued to develop, significantly promoting collaborative innovation activities among enterprises, universities, and research institutions (Li et al., 2022) [27]. Existing research indicates that the Sci-Tech Finance policy significantly promotes the entry of enterprises into strategic emerging industries, technology-intensive industries, and high-tech industries while inhibiting their entry into other sectors (Liu and Liu, 2024) [28]. Furthermore, the Sci-Tech Finance policy has effectively enhanced the enterprises’ environmental information disclosure (Zhao, 2024) [29]. Additional literature studies point out that the advancement of the Sci-Tech Finance policy has significantly improved corporate technological innovation and total factor productivity (Zhong and Jin, 2024 [30]). Scholars have also noted that the implementation of China’s first batch of pilot policies in 2011 demonstrated a clear promotional effect on enhancing the market value of enterprises in the pilot cities (Li and Liu, 2022) [31]. However, there is a lack of direct research on the impact of the Sci-Tech Finance pilot policy on corporate ESG activities in the existing literature

2.3. Sci-Tech Finance Policy and Corporate ESG Activities

The existing literature has explored topics related to the influence of financial support, innovative behavior, and technological advancements on corporate ESG activities, providing certain support for clarifying the impact of the Sci-Tech Finance policy on corporate ESG activities.

The Sci-Tech Finance pilot policy, accompanied by technological credit support, the establishment of a multi-tiered capital market system, and advancements in fintech, has effectively promoted the optimization of the financial service environment (Gao et al., 2024) [32]. The significant impact of financial market development and financial environment optimization on corporate ESG behavior is supported by numerous studies.

Ng et al. (2022) [17] conducted a study using data from Asian countries from 2013 to 2017, revealing a significant positive correlation between financial development and corporate ESG activities. Li and Pang (2023) [7] utilized data from Chinese A-share listed companies from 2011 to 2020 and found that the development and deepening of regional digital finance promote corporate ESG performance. Existing research indicates that the development of fin-tech can enhance corporate ESG performance by fostering green innovation, alleviating financial mismatches, and reducing environmental uncertainty for enterprises (Trottat et al., 2024 [33]; Wang et al., 2022 [34]). In particular, the development of green fintech can effectively address some shortcomings in the current sustainable finance framework, such as ESG information disclosure and ESG ratings (Macchiavello and Siri, 2022 [35]), exerting a significant impact on corporate ESG activities. Scholars have also studied the impact of technology loans on corporate ESG activities and found a significant positive correlation between them (Qian et al., 2023 [36]).

Technological innovation, being the direct service target of the Sci-Tech Finance pilot policy, has attracted attention in many studies regarding its impact on corporate ESG activities. Technological innovation is closely related to a corporate’s behaviors (Baek and Lee, 2024 [37]), especially with the increasing attention of governments, investors, and enterprises on innovation activities. The existing literature indicates that technological innovation has significantly contributed to the improvement of corporate environmental performance. For example, the development and design of green products and the improvement of clean technology capabilities can reduce the enterprises’ energy consumption in the production process and help establish the social image of environmental protection of enterprises (Gillan et al., 2021 [1]). Nguyen and Almodbvar (2018) [38] pointed out that green innovation activities can effectively improve the corporate’s environmental management level and financial performance. Technological innovation activities and corporate social responsibility are also correlated. Zhou et al. (2019) [39] found that there is a significant positive correlation between CSR and technological innovation investment. Firms with substantial investment in technological innovation tend to be those with a strong sense of CSR. Furthermore, a positive feedback relationship exists between corporate innovation and corporate governance. On the one hand, enterprises can use new technologies to manage enterprises better and improve the transparency and efficiency of internal decision-making. On the other hand, corporate innovation activities will also react to the improvement of corporate governance structure and governance capacity (Arifin et al., 2022) [40]. Some literature empirically discusses the impact of innovation activities on corporate ESG performance and indicates that scientific and technological innovation plays a significant role in promoting ESG behaviors (Mu et al., 2023 [41]).

The above literature provides indirect support for elucidating the intrinsic correlation between Sci-Tech Finance and corporate ESG. Nevertheless, existing research lacks direct exploration and in-depth analysis regarding the impact of technology finance policies on corporate ESG performance. What are the implications of Sci-Tech Finance policy for corporate ESG behaviors? The question of how to boost corporate ESG performance with the power of Sci-Tech Finance policy is a research topic of great significance. Furthermore, given the pivotal role of Sci-Tech Finance in fostering high-quality economic development in contemporary nations, examining the impact of China’s Sci-Tech Finance pilot policy on corporate ESG activities offers a significant empirical basis for other countries to invigorate Sci-Tech Finance and harness it to advance corporate sustainable development.

3. Policy Background and Theoretical Discussion

3.1. Background and Overview of Sci-Tech Finance Pilot Policy

To address the longstanding issue of inadequate financial support for enterprise innovation, the Chinese government launched the policy of promoting the integration of science, technology, and finance in 2011 at the city level (referred to as the “Sci-Tech Finance Pilot policy”). Beijing, Shanghai, Guangdong, and 12 other regions were designated as the first batch of pilot areas, covering 43 cities. Corresponding policies encompass a series of measures, including innovating financial investment methods in science and technology, guiding financial capital to participate in national science and technology special projects and support plans through means such as grants and risk compensation; promoting intellectual property pledge loans for technology-based enterprises; supporting technology-based enterprises to enter multi-tiered capital markets; and enhancing the system of technology insurance products, among others. In 2016, an additional nine cities, namely Shenyang, Zhengzhou, Ningbo, Xiamen, Jinan, Nanchang, Guiyang, Yinchuan, and Baotou, were selected. The second batch of pilot policies further encouraged financial institutions to establish specialized risk control mechanisms for Sci-Tech Finance, optimized financing channels for technology enterprises, explored innovative models of investment and loan collaboration, and optimized the ecosystem for Sci-Tech Finance. The objective of Sci-Tech Finance policy is to effectively integrate technological and financial resources, accelerating the creation of an investment and financing system centered around technological innovation, achievement transformation, and application promotion.

3.2. The Internal Logic of Sci-Tech Finance Policy Influencing Corporate ESG

As a deep extension of financial services to the field of technology support, Sci-Tech Finance is self-evidently significant for promoting corporate innovation behaviors and long-term development vitality. From the perspectives of environmental (E), social (S), and governance (G), the following elaborates on the internal logic through which the Sci-Tech Finance pilot policy influences corporate ESG behaviors.

Firstly, the Sci-Tech Finance pilot policy plays a significant role in guiding and incentivizing enterprises to upgrade energy technologies and enhance environmental management. The improvement of corporates’ environmental management is inseparable from capital investment and output performance in green production, resource recycling, and environmental governance. The Sci-Tech Finance policy aims to continuously stimulate and promote enterprises’ vitality and output in technological innovation. It provides crucial financial services and credit support for innovation activities in green production and green management, which is conducive to enhancing enterprises’ green productivity and environmental governance capabilities.

Secondly, the Sci-Tech Finance pilot policy contributes to promoting corporate social responsibility. According to corporate reputation theory, fulfilling social responsibilities is a crucial driver in shaping a company’s reputation, conferring upon enterprises’ competitive advantages (Mai et al., 2021) [42]. The Sci-Tech Finance pilot policy significantly stimulates collaborative innovation and cooperation among microeconomic entities, including enterprises, universities, and research institutions. This facilitates the development of more diversified innovation models and collaboration mechanisms by enterprises, enabling them to engage more effectively in CSR practices. Furthermore, Sci-Tech Finance pilot policy guides and promotes the application of new information management technologies in corporate management and decision-making process. This enhances enterprises’ decision-making capabilities and efficiency (Zhong and Jin, 2024 [30]), enabling them to more effectively predict and respond to social image issues such as product recalls, quality lawsuits, and public opinion crises, thereby controlling social risks and elevating social responsibility ability.

Lastly, the Sci-Tech Finance pilot policy provides impetus for enhancing corporate governance by fostering innovations and applications in digital information management technology. On the one hand, the policy encourages and guides enterprises to expedite advancements in management technology and information technology upgrades, thereby enhancing internal management and risk control capabilities within enterprises and fostering alignment between the interests of shareholders and other stakeholders. On the other hand, the policy drives the output and application of regional innovative technologies, encompassing not only production technology innovations but also innovations in management, operating models, and business models. Regardless of whether these technologies originate externally or internally within the enterprise, their effective integration and application contribute to strengthening the corporation’s comprehensive governance capabilities in areas such as internal management, production, operations, and other facets.

Based on the above discussion, research Hypothesis H1 is put forward: the Sci-Tech Finance pilot policy contributes to enhancing enterprises’ ESG performance.

3.3. Theoretical Path of Sci-Tech Finance Policy Influencing Corporate ESG

The following provides a theoretical elaboration on the impact pathways through which the Sci-Tech Finance pilot policy enhances corporate ESG performance, primarily encompassing three pathways: reducing financing constraints for ESG activities, enhancing corporate green innovation, and boosting corporate total factor productivity.

3.3.1. The Intermediary Path for Alleviating Financing Constraints

The theory of financing constraints emphasizes that financing sustainability is a crucial factor influencing corporate development (Pfeffer and Salancik, 2015) [43]. Under the ESG framework, companies face significant financing constraints when engaging in ESG activities (Hall and Lerner, 2010) [44]. Specifically, ESG activities involve long-term capital investments in environmental management and social services, which not only increase operational costs but also carry substantial uncertainty regarding short-term profit returns (Haessler, 2020) [45]. This, in turn, dampens the enthusiasm of both companies and investors for ESG activities. Moreover, to meet market investors’ expectations, some companies resort to “greenwashing” behaviors to improve their ESG image and obtain higher market premiums (Baldi et al., 2022) [46]. Such practices hinder the accurate assessment of the true quality of ESG projects by investors, thereby reducing their willingness to invest in ESG initiatives.

The Sci-Tech Finance pilot policy can alleviate the financing constraints and information asymmetry faced by companies in ESG activities. First, the implementation of the policy strengthens the “financial support + big data” model provided by financial institutions, facilitating data-sharing between enterprises and financial institutions, improving the financing efficiency for ESG activities. Second, improvements in corporate ESG performance are accompanied by continuous advancements in environmental management, product quality, and social responsibility, all of which are inseparable from corporate innovation awareness and activities. The Sci-Tech Finance policy aims to encourage technological innovation, which guides financial institutions to integrate ESG performance into their corporate assessment systems. This, in turn, disseminates ESG principles in lending and venture capital activities, increasing the attention of financial markets on corporate ESG activities (Hübel & Scholz, 2020) [47]. As a result, the pilot policy broadens ESG financing channels and reduces financing costs. Finally, the implementation of Sci-Tech Finance policy improves the multi-tiered capital market system, optimizes regional financial environments and services, and alleviates the financing difficulties faced by small and medium-sized enterprises in production and operation processes. Therefore, these enterprises have more available funds to engage in ESG activities.

Accordingly, Hypothesis H2 is proposed: the Sci-Tech Finance pilot policy promotes corporate ESG performance by alleviating the financing constraints of corporates.

3.3.2. The Intermediary Path for Advancing Green Innovation

Innovation theory underscores the idea that innovation is a key driver driving the long-term development of enterprises (Schumpeter, 1934) [48]. Within the ESG framework, green innovation has emerged as a distinctive feature of enterprise development. Green innovation not only aids enterprises in achieving energy conservation, emission reduction, and resource recycling in environmental terms but also enhances their economic performance. Moreover, green innovation is a vital expression of corporate social responsibility, prompting firms to continuously provide environmentally sustainable products and services.

The Sci-Tech Finance pilot policy plays a pivotal role in facilitating corporate green innovation. Firstly, as guiding principles leading the development of Sci-Tech Finance, the pilot policy enhances the environmental atmosphere for green innovation activities in a region, thereby supporting green technological advancements in enterprises. Secondly, the Sci-Tech Finance pilot policy focuses on providing multi-faceted and convenient financial services for innovation activities, which enhances the efficiency of green innovation in enterprises. The numerous small and medium-sized enterprises in particular often face the “Macmillan Gap” (Cull and Xu, 2000) [49] in green innovation practices due to the preferential allocation of financial resources in the traditional financial environment. Sci-Tech Finance pilot policy effectively addresses these issues by providing credit and financing support for diverse green innovation activities, thereby enhancing enterprises’ investment and output in green innovation. Lastly, the development of Sci-Tech Finance relies on the continuous infiltration and integration of big data technology into financial service operations. The implementation of Sci-Tech Finance policy promotes the application of big data technology in financial activities, improving the efficiency of financial supervision and support for green innovation endeavors.

In view of the above discussion, Hypothesis H3 is proposed: the Sci-Tech Finance pilot policy promotes corporate ESG performance by encouraging corporates’ green innovation activities.

3.3.3. The Intermediary Path for Improving TFP

Within the neoclassical economic framework, productivity serves as a crucial driver for corporate profitability and competitiveness. Total Factor Productivity (TFP), as a crucial indicator for measuring efficiency improvements in production management, operational strategies, investment and financing activities, and other aspects, not only directly correlates with enterprises’ economic benefits but also profoundly influences enterprises’ environmental management, internal governance, and social responsibility fulfillment. Enhancing TFP contributes to boosting corporate ESG performance.

The Sci-Tech Finance pilot policy is dedicated to stimulating corporate innovation vitality. This policy not only helps stimulate innovation in production management, operational activities, and other areas but also guides enterprises to prioritize productivity advancement as a key mission for development, thereby fostering technological advancements and enhancing factor productivity. From the perspective of Stakeholder Theory, the sustained engagement in innovation activities not only augments corporate operating performance and market share but also influences the interests of shareholders, investors, and consumers, encompassing the fulfillment of shareholders’ expectations for financial returns, the satisfaction of consumers’ demands for product quality, and the reinforcement of a company’s long-term market competitiveness. Furthermore, the Sci-Tech Finance policy facilitates the dissemination and adoption of new technologies within regions. By leveraging, assimilating, and implementing these technologies, enterprises can elevate TFP. This process not only aids in reducing corporate environmental governance costs but also enhances corporate reputation and transparency, thereby bolstering investors’ trust and support.

Lastly, under the guidance of Sci-Tech Finance policy, credit funds and various social resources are directed towards enterprises engaging in technological innovation, which also exhibit a stronger inclination towards technological innovation and social responsibility awareness. Through this screening effect, financial resources provide more pronounced support to enterprises with “willingness for innovation and ESG behavior motivation”, aiding in further advancing the technological level and TFP of corresponding enterprises, thus laying a foundation for the long-term sustainable development under ESG goals.

Therefore, Hypothesis H4 is put forward: the Sci-Tech Finance pilot policy promotes corporate ESG performance by improving the corporate TFP.

Based on the theoretical analysis in this section, we present the mechanism of the impact of Sci-Tech Finance policy on corporate ESG through Figure 2.

4. Methodology and Data

4.1. Model Design

Two batches of pilot cities were designated as the pilot cities for Sci-Tech Finance in China in 2011 and 2016, respectively. Therefore, this paper utilizes the Staggered DID model as the benchmark model and constructs a dummy variable as a proxy for the Sci-Tech Finance pilot policy, based on the quasi-natural experiment of Sci-Tech Finance pilot, in an attempt to analyze the impact of promoting Sci-Tech Finance on the ESG behaviors of enterprises. The specific setting of the model is shown in Equation (1)

$${ESG}_{i,t}={\alpha }_0+{\alpha }_1{did}_{i,t}+\beta {Controls}_{i,t}+{\gamma }_i+{\mu }_t+{\epsilon }_{i,t}$$

(1)

${ESG}_{i,t}$ is the ESG score of corporate i at time t; ${did}_{i,t}$ is the dummy variable to measure whether the city has implemented the pilot policy of Sci-Tech Finance at time t where corporate i is located. ${did}_{i,t}=$ 1 means that the region has carried out the pilot policy of Sci-Tech Finance where corporate i is situated at time t; ${did}_{i,t}=0$ means that the city where the corresponding corporate situated at time t has not started the pilot policy. The coefficient of ${did}_{i,t}$, ${\alpha }_1,$ functions as a metric for assessing the impact of the pilot policy of Sci-Tech finance on corporate ESG performance. ${Controls}_{i,t}$ denote other control variables, while ${\gamma }_i$ and ${\mu }_t$ represent the individual effects and time effects, respectively. ${\epsilon }_{i,t}$ stands for a stationary error term with a mean of zero.

4.2. Definition and Description of Variables

4.2.1. Explained Variables

According to Li et al. (2024) [50], the ESG score of listed companies published by the Sino-Securities ESG Rating system is selected as the explained variable (denoted as ESG). The database evaluates corporate ESG performance across three dimensions: environmental protection, social responsibility, and corporate governance, utilizing a total of 16 s-level indicators and 44 third-level indicators. Finally, the ESG performance of listed companies is rated as AAA, AA, A, BBB, BB, B, CCC, CC, and C from high to low, and the corresponding levels are assigned from 9 to 1 to obtain the average corporate ESG score. The higher the ESG score is, the better ESG performance the company has.

4.2.2. Core Explanatory Variables

The study treats the implementation of the Sci-Tech Finance pilot policy as a quasi-natural experiment and quantifies it using the core explanatory variable ${did}_{i,t}$ = ${treat}_i$*${post}_t$, which is specifically defined as the interaction between the timing of the pilot policy and the city where the corporate is located. Specifically, ${treat}_i$ is a dummy variable to measure whether the city where corporate i is located is on the pilot policy’ implementation list; if it is true, ${treat}_i$ is assigned a value of 1, and otherwise, it is 0. The dummy variable ${post}_t$ indicates the year of approval of Sci-Tech Finance pilot city where the corporate i is located; it is assigned the value of 1 when the city is on the pilot list and 0 otherwise.

4.2.3. Control Variables

In reference to related studies (Dicks and Fulghieri (2021) [51], Gao et al. (2023) [52] and Coad et al. (2018) [53]), the control variables include the following:

(1) The asset–liability ratio (Lev): this reflects the degree of dependence on debt financing as the ratio of total annual liabilities to total assets. A high asset–liability ratio may indicate that enterprises are facing greater financial pressure, which may force enterprises to sacrifice long-term sustainable development goals in the short term, thus affecting ESG investment.

(2) Corporate age (Age): This indicates the maturity and experience of enterprises in the market. Older enterprises may have established a relatively stable operation mode and corporate culture, including corporate ESG construction. However, due to their established operational model, they may encounter challenges in adapting to new ESG standards and practices. The index of corporate age is represented by the logarithm of the year of observation minus the year of establishment of the enterprise.

(3) Return on shareholders (ROE): This represents the ratio of a corporation’s net profit to its shareholders’ equity. A high ROE may suggest that the corporation possesses high operational efficiency and the capacity to invest in social responsibility and environmental protection projects, thereby enhancing its ESG performance.

(4) Financial leverage (FL): This indicates the degree to which enterprises utilize debt financing, calculated as the ratio of total debt to shareholders’ equity. Highly leveraged enterprises require more resources for debt repayment and are more inclined to prioritize short-term financial indicators over long-term sustainable development.

(5) Corporate growth (Growth): This is measured by the growth rate of operating income, specifically the ratio of the increase in operating income to the total operating income of the previous year. Fast-growing companies may be more inclined to innovate and expand their ESG practices. Additionally, growth may provide companies with the resources and capabilities necessary to better manage their environmental and social impacts as they expand.

(6) Cash flow (lnCash): This is measured by the logarithm of the total amount of cash income and expenditure generated by operating activities, investment activities, and financing activities every year. Sufficient cash flow enables enterprises to make ESG investments.

(7) Proportion of independent directors (lnd): This reflects the independence of voice from management within the board of directors, calculated as the logarithm of the ratio of the number of independent directors to the total number of directors. A high proportion of independent directors may contribute to enhancing the transparency and fairness of corporate decision-making, thus promoting corporate actions related to environmental and social responsibility.

(8) Board size (lnBoard): This is measured by taking the logarithm of the number of board members of the company. The size of the board may affect the efficiency and diversity of its decision-making and help enterprises make more comprehensive decisions on ESG issues.

4.2.4. Mediating Variables

The previous theoretical analysis points out that the implementation of Sci-Tech Finance policy mainly influences corporate ESG through three channels: alleviating corporate financing constraints, promoting green innovation, and improving corporate total factor productivity. Thus, in the subsequent discussion, the following indicators are selected as mediating variables for empirical testing and analysis.

(1) Absolute value of SA index: This index serves to measure the degree of financial constraints of enterprises. This paper refers to the absolute value of the SA index proposed by Hadlock and Pierce (2010) [54] as the index to measure the degree of corporate financial constraints, $SA=-0.737$ × Size + 0.043 × ${\mathrm{S}\mathrm{i}\mathrm{z}\mathrm{e}}^2$ − 0.04 × Age, where size is the natural logarithm of the enterprises’ total asset scale and age is the operating year of the enterprise, equal to the observation year (current accounting period) minus the establishment year of the enterprise.

(2) The logarithm of the number of enterprises’ green patents (GP): This index is used to measure enterprises’ green innovation capability (Hall and Lerner, 2010) [44]. Compared to the number of green patent applications, the number of granted green patents has stronger legal validity and can more accurately reflect the level of green innovation within enterprises. In this paper, we use the logarithm of the number of granted green patents by enterprises as a metric for measuring their green innovation capability.

(3) Total factor productivity (TFP). Considering the measurement method of total factor productivity by Levinsohn and Petrin (2003) [55], the LP semi-parameter estimation method is taken to calculate TFP. The production function is set as follows: $Y_{i,t}$ = $A_{i,t}{K}_{i,t}{L}_{i,t}{M}_{i,t}$, where $Y_{i,t}$ is the output of corporate i at time t, $K_{i,t}$ is the capital input, $L_{i,t}$ is the labor input, $M_{i,t}$ is the intermediate input, and $A_{i,t}$ is the total factor productivity (TFP) of the enterprises.

4.3. Data Description

4.3.1. Data Source

This paper analyzes the data of A-share listed companies in their cities from 2009 to 2022 as the research sample and conducts the following processing on the data: (1) excluding the data of the financial and insurance industry; (2) excluding ST, *ST and PT companies; (3) winsorizing the continuous data before and after 4%; (4) calculating the logarithm of some control variables; (5) removing the sample data with missing important variables, and handling the annual panel data of a total of 13,931 enterprises. The ESG rating data of enterprises are from the Wind database, and the data on corporate green patent comes from the China Research Data Services Platform (CNRDS), while the rest of the empirical data indicators are mainly obtained from the CSMAR database, the China City Statistical Yearbook, and the Digital Finance Research Center of Peking University.

4.3.2. Descriptive Statistics

After the processing of missing values, abnormal items, and 4% truncation, 13,931 corporates sample observations were finally obtained.

Table 1. Descriptive statistics.

Variables	Obs	Mean	Median	SD	Min	Max
ESG	13.931	3.992	4.000	1.195	1.000	6.000
did	13.931	0.630	1.000	0.483	0.000	1.000
Lev	13.931	0.430	0.420	0.203	0.098	0.816
lnCash	13.931	19.173	19.176	1.584	16.013	22.286
FL	13.931	1.128	1.036	0.642	0.000	3.064
ROE	13.931	0.080	0.081	0.094	−0.180	0.263
Growth	13.931	0.236	0.103	0.488	−0.376	1.834
lnBoard	13.931	2.133	2.197	0.174	1.792	2.485
Age	13.931	2.295	2.485	0.758	0.693	3.258
Ind	13.931	3.138	3.000	0.470	2.000	4.000

shows the brief descriptive statistics of the sample information. The average value of corporate ESG scores is 3.992, with the standard deviation being 1.195, reflecting the differences in ESG performance among the sampled enterprises. The average value of the variable did is 0.63, with the corresponding standard deviation being 0.483, which means that 63% of the sample enterprises are affected by the Sci-Tech Finance pilot policy, and the values and fluctuations of the remaining variables are within the description range of the existing literature.

5. Empirical Results

5.1. Benchmark Regression

The multi-stage DID model is adopted to examine the impact of the Sci-Tech Finance pilot policy on the corporate ESG performance, with the results presented in

Table 2. Benchmark regression table of variables.

	(1)	(2)	(3)	(4)
	ESG	ESG	ESG	ESG
did	0.142 ***	0.111 ***	0.039 *	0.148 ***
	(0.053)	(0.034)	(0.042)	(0.052)
Lev		0.698 ***	0.696 ***	0.755 ***
		(0.101)	(0.098)	(0.103)
lnCash		0.001	0.148 ***	0.008
		(0.007)	(0.011)	(0.007)
FL		−0.017	0.007	−0.010
		(0.014)	(0.021)	(0.014)
ROE		0.476 ***	2.061 ***	0.438 ***
		(0.133)	(0.174)	(0.133)
Growth		0.024	0.103 ***	0.045 **
		(0.023)	(0.032)	(0.023)
lnBoard		0.315 ***	0.263 **	0.342 ***
		(0.114)	(0.124)	(0.114)
Age		0.166 ***	0.098 ***	0.177 ***
		(0.033)	(0.024)	(0.054)
Ind		0.183 ***	0.295 ***	0.176 ***
		(0.038)	(0.044)	(0.038)
_cons	3.899 ***	3.879 ***	1.094 ***	3.806 ***
	(0.033)	(0.261)	(0.269)	(0.281)
Firm effect	Yes	Yes	NO	Yes
Year effect	Yes	NO	Yes	Yes
R2	0.679	0.679	0.122	0.685
N	13,931	13,931	13,931	13,931

Note: clustering standard errors at the firm level are shown in parentheses, *** p < 0.01, ** p < 0.05, and * p < 0.1.

. Column (1) shows the results without control variables. The regression coefficient of the pilot policy of Sci-Tech Finance and the corporate ESG performance is 0.142, which is significantly positive at the level of 1%, indicating that the pilot policy of Sci-Tech Finance has a statistically significant positive impact on the corporate ESG performance. Columns (2) and (3) show the regression results with control variables, which involve individual-fixed effects and time-fixed effects, respectively. The results are statistically significant at the 1% and 10% levels, respectively, indicating that the Sci-Tech Finance pilot policy can enhance corporate ESG performance. Column (4) displays the regression analysis data with control variables; individual and time-fixed effects are considered. The regression coefficient of the pilot policy of Sci-Tech Finance on the corporate ESG performance is 0.148, which is fully significant at the level of 1%, demonstrating that the effect of the Sci-Tech Finance pilot policy on corporate ESG performance is still obvious. The aforementioned results, derived from model formula 1, demonstrate that the implementation of the Sci-Tech Finance pilot policy can promote corporate ESG performance, thereby supporting Hypothesis 1.

5.2. Parallel Trend Test

In the DID empirical analysis, it is necessary to conduct the parallel trend test between the experimental group and the control group to identify better the causal relationship between the Sci-Tech Finance pilot policy and the corporate ESG performance. The null hypothesis of the parallel trend test is that if there is no special treatment, the average results of the experimental group and the non-treatment group (control group) will exhibit the same development trend. The alternative hypothesis is that the ESG performance of the experimental group and the control group will follow the same trend before the implementation of the Sci-Tech Finance pilot policy, with all other conditions remaining constant. However, subsequent to the policy implementation, the experimental group exhibited a notable change in trend compared to the control group. To test the parallel trend, the following dynamic model is constructed:

$${ESG}_{i,t}=C+{\sum }_{k=-6,k\ne -1}^6{\alpha }_k{did}_{i,t+k}+\beta {Controls}_{i,t}+{\gamma }_i+{\mu }_t+{\epsilon }_{i,t}$$

(2)

In Equation (2), the value of k is set to range from −6 to 6. k < 0 represents the years before the implementation of the Sci-Tech Finance pilot policy in some regions; otherwise, it represents the years after the policy implementation. Considering the inherent lag in policy transmission, this study takes the year immediately preceding the implementation of the Sci-Tech Finance pilot policy as the base year. The coefficient ${\alpha }_k$ reflects the corporate ESG performance before and after the implementation of the Sci-Tech Finance pilot policy, while controlling for other variables. The results of the parallel trend test in Figure 3 indicate that the estimated coefficients ${\alpha }_k$ are close to zero at multiple time points before the pilot policy implementation, and the 90% confidence intervals of the coefficients all cover the zero line, suggesting that there is no significant trend difference between the treatment group and the control group at these time points. After the policy implementation, the coefficients ${\alpha }_k$ remains within the positive range, and their confidence intervals exclude the 0-axis, indicating a significant difference from the coefficient estimates prior to the policy implementation. This demonstrates that there was no apparent trend difference between the experimental group and the control group before the implementation of the policy, and the specific difference emerged only after the policy was put into effect. In other words, Figure 2 shows that the results have passed the parallel trend test.

5.3. PSM-DID

With reference to the practice of Yang et al. (2023) [56], this paper uses the propensity score matching (PSM) method to screen the two groups of samples. The Logit model and kernel density are combined to match the two types of enterprises, and all the control variables in Equation (1) are selected as covariates. Specifically, the Logit model is first adopted to estimate the propensity score of each individual, and the score indicates the probability that the individual accepts the treatment. Subsequently, the kernel density method was employed for matching, with weights assigned based on the similarity between individuals in the treatment and control groups. The weights were based on the similarity of the propensity score.

Figure 4 illustrates the trend of density function values, revealing a larger discrepancy in score values before matching, which indicates the differences in characteristic variables between the two groups of samples prior to matching. After matching, the propensity score values are relatively close, with the distribution of the selected control variables similar between the experimental group and the control group, ensuring the two groups’ comparability.

$${{ESG}^{PSM}}_{i,t}={\alpha }_0+{\alpha }_1{\mathrm{d}\mathrm{i}\mathrm{d}}_{\mathrm{i},\mathrm{t}}+\mathsf{\beta }\sum {Controls}_{i,t}+{\gamma }_i+{\mu }_t+{\epsilon }_{i,t}$$

(3)

${{ESG}^{PSM}}_{i,t}$ is the enterprise’ ESG score obtained after propensity matching.

Table 3. PSM-DID model results.

Items	${\mathit{E}\mathit{S}\mathit{G}}^{\mathit{P}\mathit{S}\mathit{M}}$
did	0.009 ***
	(0.003)
Firm effect	YES
Year effect	YES
R2	0.497
N	13,931

Note: Due to space limitations, the results of the other control variables are not listed. Clustering standard errors at the firm level are shown in parentheses, *** p < 0.01.

shows the regression results under the PSM-DID method. The coefficient of did in Table 3 is significantly positive, indicating that the implementation of policy has a positive effect on improving the corporate ESG performance.

5.4. Placebo Test

The placebo test is commonly used to verify the robustness of model results and the plausibility of causality by introducing a dummy “treatment” to observe its effect on the outcome variable. If the regression coefficient of the placebo variable is not significant, it indicates that the treatment effect in the actual model is more likely to be causal rather than caused by other external factors. Subsequent text further validates the causal relationship between the Sci-Tech Finance pilot policy and the development of corporate ESG through a series of “placebo tests”.

5.4.1. Fictitious Time Groups

Policy implementation has a lagged effect on enterprises and may be influenced by other factors when assessing the impact of Sci-Tech Finance pilot policy on corporate ESG performance. Therefore, this study advances the policy timeline by 1 to 3 years, considering these periods as hypothetical implementation stages. The placebo test replaces the actual treatment variable with these fictitious periods to assess their impact on the dependent variable, while insignificant results from the fictional policy would indicate a robust model and credible treatment effect.

Columns (1), (2), and (3) of

Table 4. Results of the fictitious time treatment group.

	(1)	(2)	(3)
Items	ESG	ESG	ESG
did (−1)	0.103
	(1.225)
did (−2)		0.051
		(0.623)
did (−3)			−0.098
			(1.305)
Firm effect	Yes	Yes	Yes
Year effect	Yes	Yes	Yes
R2	0.716	0.777	0.708
N	6716	5653	4735

Note: clustering standard errors at the firm level are shown in parentheses.

show the regression results with the policy lagged for 1, 2, and 3 years, respectively. The corporate ESG performance in corresponding regression designs does not exhibit significant variations, suggesting that the treatment effect manifests itself only after the policy’s implementation, rather than being a consequence of randomness or anticipation. The placebo test further strengthens the credibility of the main regression results.

5.4.2. Fictitious Experimental Group

The inclusion of a fictitious experimental group in placebo tests can assess the robustness of the model by simulating a non-existent treatment effect, aiding in distinguishing between genuine policy impacts and spurious effects that may arise due to random variations. As follows, we randomly assign individual samples to the experimental group for the Sci-Tech Finance policy and replicate the process 2000 times. Based on this, we plot the distribution of the estimated coefficients of did under random matching.

In Figure 5, in the placebo test results, the horizontal axis is the regression coefficient, and the vertical axis is the p-value. The p values are highly concentrated near zero, indicating that the effects of the pseudo-policy variables are not random fluctuations, which may suggest systematic problems or omitted variables in the model. In addition, the kernel density estimation curve shows a normal distribution, with the estimated beta coefficient generally distributed around 0. All deviate from the regression coefficient of 0.148, as shown in Column (2) of Table 3, indicating that the actual impact of most pseudo-policy variables on the dependent variable is minimal or even nonexistent. These results exclude the possibility that the estimated results in the benchmark regression are affected by unobservable factors and confirm that implementing the Sci-Tech Finance pilot policy has a significant role in promoting the corporate ESG performance.

5.5. Heterogeneous Treatment Effects in Staggered DID

The Sci-Tech Finance pilot policy exhibits characteristics of staggered or progressive DID due to differences in treatment timing, as samples that are treated earlier serve as control groups for those treated later, potentially leading to estimation bias. To address this, this paper conducts robustness checks by referring to methods from existing research. Following the methodology of Callaway and Sant’Anna (2021) [57], the average treatment effect for each group-period was first calculated, and then a weighted aggregation across both group and period dimensions was performed to obtain the final policy treatment effect (denoted as ATT in

Table 5. Results of group-period average treatment effect.

	(1)	(2)
Items	ESG	ESG
ATT	0.193 **	0.254 ***
	(0.079)	(0.087)
Controls	NO	YES
Firm effect	YES	YES
Year effect	YES	YES
N	6183	4766

Note: clustering standard errors at the firm level are shown in parentheses, *** p < 0.01 and ** p < 0.05.

). The analysis results indicate that the Sci-Tech Finance pilot policy significantly enhanced corporate ESG performance, demonstrating that the baseline regression result in Formula (1) is minimally affected by the issue of in treatment effects heterogeneity.

5.6. Instrumental Variables

Although external policy dummy variables can mitigate endogeneity issues, there remains a concern that regions with advanced ESG development may have a higher probability of being selected as pilot areas for Sci-Tech Finance, thereby potentially introducing endogeneity problems. To address this, we chose the product term of the number of urban telephone subscribers in 2004 and the degree of urban digital financial development as an instrumental variable (denoted as iv) for Sci-Tech Finance pilot policy. The reason for selecting it as an instrumental variable is that better information infrastructure and higher digital financial development are conducive to promoting regional innovative financing and the development of Sci-Tech Finance (Gomber et al., 2017 [58]; Jiang et al., 2022 [59]; Dereje, 2024 [60]). This instrumental variable has a theoretical correlation with Sci-Tech Finance activities. Furthermore, the number of urban fixed-line telephone subscribers in 2004, as historical data, is not directly related to the pilot policy in this paper and is exogenous to the Sci-Tech Finance system, thus satisfying the requirement of exogeneity.

We conducted instrumental variable estimation using two-stage least squares, and the results are presented in

Table 6. Results of instrumental variables estimation.

	(1)	(2)
Items	First-Stage	Second-Stage
iv	0.000 ***	0.515 *
	(0.000)	(0.278)
Controls	YES	YES
Firm effect	YES	YES
Year effect	YES	YES
N	10,914	10,914

Note: In the construction of variable iv, the number of urban telephone subscribers and the digital finance development index are sourced from the China City Statistical Yearbook and the Peking University Inclusive Finance Database, respectively. Clustering standard errors at the firm level are shown in parentheses, *** p < 0.01 and * p < 0.1.

. The results further validate that in the regression results of column (1), the coefficient of iv is positive and significant at the 1% level, indicating a strong correlation between the instrumental variable and the policy pilot variable. In the regression of column (2), the coefficient of iv is positive and significant at the 10% level, suggesting that after further mitigating potential endogeneity issues, the Sci-Tech Finance pilot policy retains a significant positive effect on corporate ESG performance.

5.7. Other Robustness Tests

5.7.1. Entropy Balanced Matching

The entropy balance matching method by Hainmueller (2012) [61] has the core aim of optimizing the objective function of a specific weight to adjust the sample matching, creating a statistical equilibrium between the experimental group and control group on the selected covariate. This balance is the purpose of the simulation of the randomized control trial conditions, thus reducing the estimation bias caused by the covariate imbalance to ensure the randomness of the experiment and exogeneity (Best and Sinha, 2021) [62]. Specifically, this paper first selects all the control variables in Equation (1) as the covariates of the experimental group and control group. Secondly, the weight of each sample is determined by optimizing an objective function containing entropy maximization. Then, the first moment of the covariates is considered for matching so that the adjusted weight makes the distribution of the selected covariates in the experimental group and the control group as similar as possible. Finally, the balanced sample data are substituted into Model (1) for regression estimation, and the analysis results are shown in Column (1) of

Table 7. Other robustness test results.

	(1)	(2)	(3)	(4)	(5)	(6)	(7)
Items	ESG	ESG	ESG	ESG	ESG	ESG	ESG
did	0.187 ***	0.178 ***	0.141 **	0.158 **	0.100 **	0.149 ***	0.132 **
	(0.045)	(0.063)	(0.057)	(0.061)	(0.040)	(0.056)	(0.055)
did2							−0.344 ***
							(0.117)
did3							−0.007
							(0.064)
Firm effect	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Year effect	Yes	Yes	Yes	Yes	Yes	Yes	Yes
R2	0.684	0.682	0.735	0.686	0.032	0.686	0.672
N	13,553	10,209	7577	11,348	7187	13,530	13,010

Note: clustering standard errors at the firm level are shown in parentheses, *** p < 0.01 and ** p < 0.05.

.

The results show that the regression coefficient is significant at the 1% level. This demonstrates that after considering the problem of sample selection bias, this policy can significantly improve corporate ESG performance The robustness of the core conclusion is also verified in this paper.

5.7.2. Excluding Centrally Administered Municipalities

Given that municipalities usually have special administrative status and higher policy autonomy (Hou and Guo, 2023) [63], as well as large economic scales and high degrees of corporates concentration, their market environment is relatively unique, which may lead to significant differences between their policy implementation methods and those of other cities, thus affecting the robustness of regression results. Therefore, the data of “Beijing”, “Shanghai”, “Tianjin”, and “Chongqing” are excluded to reduce the interference caused by the differences in systems and market structure in different regions, so as to better measure the real impact of Sci-Tech Finance policy on ordinary urban enterprises. Except for the four municipalities directly under the central government administration, regression analysis results are shown in Column (2) of Table 7, with the regression coefficient being significantly positive at the 1% level, thereby demonstrating the robustness of the main regression.

5.7.3. Time Window Shortening

Shortening the sample time window can improve the ability to identify the impact of the policy and reduce the external interference error caused by the period in order to better evaluate the short-term effect of the policy. China implemented the Sci-Tech Finance pilot policy in 2011 and 2016, respectively. Subsequently, the sample time window was shortened to the two years preceding and following the policy implementation, specifically 2009–2013 and 2014–2018, for which the regression analysis was conducted. Column (3) of Table 7 shows the results after adding variable control. The regression coefficients of did are all significantly correlated at the level of 5%, which verifies the reliability of the conclusion of this paper.

5.7.4. Sample Censoring

This paper also conducts a robustness analysis by incorporating truncated sample processing. Columns (4) and (5) of Table 7 show the results of sample censoring at the levels of 1% and 5%, respectively, with all regression coefficients being significantly positively correlated at the 5% level, thereby validating the reliability of the main regression results.

5.7.5. Include Robust Standard Errors at the City Level

To further ensure the robustness of the analysis results, this study conducts another regression with robust standard errors clustered at the city level. In the context of this study, ESG performance of firms within the same city may exhibit certain similarities or dependencies, which could induce correlation among model error terms and lead to an underestimation of traditional standard errors. By clustering the robust standard errors at the city level, this approach effectively controls for these issues, allowing the regression model to better address heteroscedasticity and serial correlation. The empirical results, as shown in Columns (6) of Table 7, reveal that the signs and significance of the coefficient of did under the city-level robust standard error specification are largely consistent with the baseline regression results, further affirming the reliability of the study’s conclusions.

5.7.6. Isolate the Effects of Other Policies

The impact of Sci-Tech Finance pilot policy on corporate ESG performance may be influenced by other policies. To address this, the study attempts to isolate the effects of other policies on the estimation results. Green credit policies and innovation-driven city pilot programs, which were introduced by the Chinese government in 2007 and 2008, respectively, could potentially interfere with the analysis. Therefore, we include interaction terms between dummy variables for these two policies and time trends, denoted as did2 and did3, to examine whether these policies affect the analysis based on the benchmark model (Equation (1)). The estimation results, shown in column (7) of Table 7, indicate that, after controlling other policies, the coefficient of the Sci-Tech Finance pilot policy, did, remains positively significant, further validating the robustness of our findings.

6. Further Analysis

6.1. Intermediary Mechanism Test

To explore the specific impact mechanism of Sci-Tech Finance pilot policy on corporate ESG, the mediating effect model is adopted for the mechanism test, and the formula is presented in Equations (1), (4) and (5).

$$M_{i,t}={\phi }_0+{\phi }_1{did}_{i,t}+{\phi }_2\sum {Controls}_{i,t}+{\gamma }_i+{\mu }_t+{\epsilon }_{i,t}$$

(4)

$${ESG}_{i,t}=k_0+k_1{did}_{i,t}+k_2{M}_{i,t}+k_3\sum {Controls}_{i,t}+{\gamma }_i+{\mu }_t+{\epsilon }_{i,t}$$

(5)

$M_{i,t}$ is the intermediary variable, including financial constraints, the total amount of green patents granted and total factor productivity of enterprises. Coefficients ${\phi }_{1 }, k_2$ represent the mediating effect of $M_{i,t}$. When the coefficients ${\phi }_{1 },k_2$ are both significant, it indicates that the Sci-Tech Finance pilot policy can impact corporate ESG performance through corresponding channels.

In reference to the three-step method of the intermediate mechanism by Rijnhart et al. (2021) [64], after controlling the mediating variable M, in the study of the relationship between the pilot policy (did) and corporate ESG, the coefficient ${\phi }_1$ represents the effect of did on the mediating variable, and the coefficient $k_2$ represents the effect of the mediating variable on corporate ESG performance. These two constitute the indirect effect of the relationship between the variables in the table. The coefficient $k_1$ denotes the effect of did on ESG after controlling for the mediating variables, namely the direct effect. Then, the total effect of variables should be equal to the direct effect plus the indirect effect, that is, the total effect = ${\phi }_1$ × $k_2$ + $k_1$, and the proportion of some intermediate effects is ${\phi }_1{k}_2$/$({\phi }_1$ × $k_2$ + $k_1)$ (Pearl, 2022) [65], which reflects the degree of intermediate effects.

Firstly, this paper adopts the absolute value of the SA index to measure the degree of financing constraints, the index of the reverse index. A smaller index indicates a more severe degree of financing constraints faced by the enterprise. According to the results in Column (1) of

Table 8. Results of mediating effect test model.

Items	Mediating Path1		Mediating Path2		Mediating Path3
	SA	ESG	GP	ESG	TFP_LP	ESG
did	0.016 ***	0.092 *	1.101 **	0.147 ***	0.146 ***	0.108 **
	(0.006)	(0.056)	(0.530)	(0.056)	(0.040)	(0.050)
SA		1.295 ***
		(0.210)
GP				0.002 **
				(0.001)
TFP_LP						0.218 ***
						(0.050)
R2	0.967	0.733	0.707	0.683	0.883	0.734
N	9173	9173	13,508	13,508	12,799	12,799
Test Conclusion	partial mediating effect		partial mediating effect		partial mediating effect
Mediated Effect Proportion	18.54%		1.58%		20.55%

Note: clustering standard errors at the firm level are shown in parentheses, *** p < 0.01, ** p < 0.05, and * p < 0.1.

, the index itself is negative, and the coefficients of did and SA are significant at the 1% level, indicating that the pilot policy effectively alleviates enterprises’ financing constraints. The positive and significant coefficient of SA and ESG in Column (2) implies that the Sci-Tech Finance pilot policy can provide enterprises with diverse financing channels, enabling them to allocate funds towards sustainable investments, thereby enhancing their corporate ESG performance.

Secondly, this paper analyzes corporate green innovation as a potential influence channel. Columns (3) and (4) of Table 8 show the regression results of corporate green innovation effect. The coefficients of the independent variable and mediating variable in Column (3) are significantly positive at the level of 1%, and the case is the same for the coefficients of the intermediate variable and dependent variable in Column (4) at the level of 1%, which suggests that the implementation of the Sci-Tech Finance pilot policy has stimulated enterprises to implement green innovation, guided capital flow into the green technology field, and finally improved the corporate ESG performance.

Finally, this paper delves into the impact of the third mechanism for the enterprise total factor productivity, the first column (5) and (6) of Table 8 as the return of the resource allocation effect. Through the above analysis, the mechanism of improving the total factor productivity of enterprises turns out to be the most obvious, accounting for 20.55%. The second mechanism aims to alleviate corporate financing constraints, accounting for 18.54% of the overall impact. The third mechanism of increasing the number of green patents granted accounted for the lowest proportion (1.58%), which may be blocked due to the following reasons. Firstly, despite the policy’s support for innovation, it lags behind due to issues such as inadequate implementation and limited policy coverage during the policy implementation process, resulting in enterprises not fully benefiting from the Sci-Tech Finance policy support. Secondly, enterprises, particularly small and medium-sized enterprises, may encounter challenges such as technological bottlenecks or insufficient funds, as they lack the necessary technical personnel or resources to undertake green technology research and development. Thirdly, even when an enterprise obtains green patent authorization, the technology may fail to translate into significant commercial or environmental benefits due to limited market demand and industrialization capacity, thereby diminishing the positive impact of green patents on corporate ESG performance.

6.2. Heterogeneity Analysis

In order to deeply explore the differences in the impact of different types of enterprises and different environmental characteristics on the pilot policy of Sci-Tech Finance, analysis is conducted mainly on two aspects, the external environment and enterprises’ characteristics, so as to evaluate the impact of the pilot policy of Sci-Tech Finance on the ESG of enterprises more comprehensively and reveal the differences in the implementation process of the policy, thus providing an important basis for policy improvement and promotion.

6.2.1. Heterogeneity Analysis on Geographical Regions

Based on China’s geographical regions, enterprises are classified into the eastern, central, and western regions, and a heterogeneity regression analysis is conducted to explore this classification. The regression setup is as shown in Equation (6), where $I_k$ stands for the identification variables of the region to which the enterprise belongs, with $I_1$ representing the eastern region and $I_2$ representing the non-eastern regions; the description of the remaining controlling variables can be seen in Section 4.3.

$${ESG}_{i,t}={\alpha }_0+{\sum }_{k=1}^2({\lambda }_k{did}_{i,t}\ast I_k)+\beta {Controls}_{i,t}+{\gamma }_i+{\mu }_t+{\epsilon }_{i,t}$$

(6)

The estimation results in

Table 9. Estimation result for the eastern and non-eastern regions.

Region	${\mathit{\lambda }}_{\mathit{k}}$	Null Hypothesis	F-Statistics	Test Conclusions
East (k = 1)	0.190 ***	${\lambda }_1$ = ${\lambda }_2$	5.01 **	Rejection
Non-east (k = 2)	0.029

Note: *** p < 0.01 and ** p < 0.05.

show that the coefficient of the impact of Sci-Tech finance policy on corporate ESG performance is significantly positive in the eastern region, while it is insignificant in the central and western regions. The F-test further verifies the difference in coefficient estimates between the eastern and non-eastern regions, suggesting that enterprises in the eastern region derive greater benefits from Sci-Tech Finance pilot policy compared to those in the central and western regions. The possible reason is that enterprises in the eastern region are concentrated in coastal areas and developed cities, where the concept of Sci-Tech Finance is widely disseminated. Many enterprises in these regions have implemented ESG-related policies earlier and formed a relatively mature governance system. In contrast, the central and western regions lag behind in economic development, with fewer talents and infrastructure resources, leading to a later introduction and implementation of the ESG concept among enterprises and thus a lower degree of policy incentivization.

6.2.2. Heterogeneity Analysis on Digital Economic Environment

To reveal the different impacts of Sci-Tech Finance pilot policy on corporate ESG performance under various digital economic contexts, cities are categorized into high and low digital economy levels based on their annual average digital economic situation, with those above the corresponding mean defined as high-digital-economic regions, and those below the mean as low-digital-economic regions. Due to space limitations, the detailed construction of indicators for measuring the regional digital economy is not presented here; however, it can be obtained from the authors upon request.

We conducted heterogeneous analysis based on the level of digital economy development. The regression equation is set as (7), and $I_k$ represents the indicator variable reflecting the region where the enterprise is located, with $I_3$ indicating high-digital-economy regions and $I_4$ indicating low-digital-economy regions.

$${ESG}_{i,t}={\alpha }_0+{\sum }_{k=3}^4({\lambda }_k{did}_{i,t}\ast I_k)+\sum {Controls}_{i,t}+{\gamma }_i+{\mu }_t+{\epsilon }_{i,t}$$

(7)

Table 10. Estimation result for high-level and low-level digital economy regions.

Region Indicator	${\mathit{\lambda }}_{\mathit{k}}$	Null Hypothesis	F-Statistics	Test Conclusions
High area (k = 3)	0.214 ***	${\lambda }_3$ = ${\lambda }_4$	10.08 ***	Rejection
Low area (k = 4)	0.024

Note: *** p < 0.01.

shows the estimation results of Formula (7). The policy effect in the regions at a high digital economic level appears significant, with a coefficient of 0.214, which passes the test at the significance level of 1%. In contrast, the policy effect in the low regions turns out to be insignificant, indicating that the regions with faster digital economy development are more conducive to the positive impact of Sci-Tech Finance pilot policy on the corporate ESG performance. The F-test results show that the difference between the two groups is significant, rejecting the hypothesis that the coefficients of the two groups are equal. These results indicate that regions at a high level of digital economy may have better infrastructure and market environment, which makes it easier for enterprises to benefit from Sci-Tech Finance pilot policy and improve their ESG performance.

6.2.3. Heterogeneity Analysis on Industrial Technological Attributes

The high-tech enterprises (HTEs) in this paper are defined according to the high-tech fields stipulated in the Administrative Measures for the Identification of High-tech Enterprises (No. [2008]172) issued by China National Sci-Tech Finance Co., Ltd., corresponding to the Guidelines for Industry Classification of Listed Enterprises revised by China Securities Regulatory Commission in 2012. Compared with traditional enterprises, HTEs have asset-light, high-growth, high-risk, and high-competition characteristics. The pilot policy of Sci-Tech Finance can provide rich resources for the innovation and development of high-tech enterprises, which is greatly conducive to the growth of high-tech enterprises. It can be seen from

Table 11. Results of heterogeneity analysis on enterprise level.

Items	Technological Attributes		Factor Intensity			Pollution Attributes		Property Attributes
	HTEs	Non-HTEs	LIEs	CIEs	TIEs	HPIs	Non-HPIs	SOEs	Non-SOEs
did	0.228 ***	0.093	0.024	0.300 **	0.119 *	0.262 **	0.066	0.207 **	0.084
N	6481	5905	3440	3360	5536	3638	8778	4918	7064

Note: *** p < 0.01, ** p < 0.05, and * p < 0.1.

that the coefficient of Sci-Tech Finance policy is significantly positive in the sample group of high-tech enterprises, with a coefficient of 0.228. Correspondingly, it is not significant in the sample group of non-high-tech enterprises (non-HTEs), which reveals that implementing Sci-Tech Finance policy exerts a more significant effect on improving ESG in HTEs.

6.2.4. Heterogeneity Analysis on Industry Factor Intensity

According to the standard classification of 2012 industries, intensive industries of all samples are divided into three types, technology-intensive, capital-intensive, and labor-intensive industries, in line with the production factors. Labor-intensive enterprises (LIEs), less dependent on technology and equipment in their daily business activities, are mainly concentrated in handicrafts and textile industries which are dominated by labor. Therefore, the Sci-Tech Finance pilot policy has little impact on these types of enterprises. Capital-intensive enterprises (CIEs) possess large equipment and fixed assets, with high capital utilization efficiency. They can take advantage of the policy to make sustainable investments, thus improving their ESG performance. Technology-intensive enterprises (TIEs) are characterized by high technology dependence, high R&D investment, and high knowledge intensity. The pilot policy of Sci-Tech Finance provides a variety of innovation supports for technology-intensive enterprises, enhancing their technological innovation ability and further improving their long-term sustainable investment.

Table 11 shows the heterogeneous results of LIEs, CIEs, and TIEs, respectively. The estimated coefficient for LIEs is 0.024, which is not significant. The coefficients for CIEs and TIEs are 0.3 and 0.119, respectively, which are significant at the level of 5% and 10% as well, indicating that the Sci-Tech Finance pilot policy is more conducive for CIEs and TIEs to improve their ESG level.

6.2.5. Heterogeneity Analysis on Industry Pollution Attributes

Based on the Guidance on Industry Classification of Listed Companies issued by the China Securities Regulatory Commission (CSRC) in 2012 and the Industry Classification Management List for Environmental Protection Verification of Listed Companies formulated by the Ministry of Environmental Protection in 2008, enterprises are classified into different categories. Among them, 16 industries, such as coal, mining, textiles, leather, paper-making, petrochemicals, pharmaceuticals, chemicals, and metallurgy, are classified as heavy-polluting industries (HPIs), while other industries are classified as non-heavy-polluting industries (non-HPIs).

According to Table 11, the coefficient for enterprises in heavy-polluting industries (HPIs) is 0.278, significant at the 5% level, whereas the coefficient for non-heavy-polluting industries (non-HPIs) is 0.076, which is statistically insignificant. This may be attributed to the fact that HPIs are typically subject to higher environmental protection requirements and stricter external supervision. The public and investors are more sensitive to their corporate behaviors, prompting them to adopt more measures for environmental protection and governance. At the same time, the Sci-Tech Finance pilot policy provides more financial channels for relevant enterprises to accelerate their green transformation and promote their ESG level. Non-HPIs have the characteristics of diversified industry types, a strong awareness of green development, and high knowledge intensity, while the pressure of environmental regulation is relatively light. Moreover, enterprises in non-HPIs may have laid a solid ESG foundation, rendering the policy’s impact on them insignificant.

6.2.6. Heterogeneity Analysis on Enterprise Property Rights

The far-right column of Table 11 presents the estimation results for both state-owned enterprises (SOEs) and non-state-owned enterprises (non-SOEs). Under the category of SOEs, the coefficient of did is 0.207, which is significant at the 5% level. In contrast, the coefficient for non-SOEs is not significant, suggesting that the Sci-Tech Finance pilot policy has a more pronounced ESG impact on SOEs than non-SOEs. The possible reasons behind this heterogeneous result mainly lie in the fact that SOEs have received greater government support and investment, enabling them to better benefit from the Sci-Tech Finance policy. Furthermore, the planning of ESG strategies within enterprises is primarily guided by government regulatory authorities. In comparison to non-SOEs, SOEs typically demonstrate a stronger response to this policy. Moreover, SOEs attach greater significance to social responsibility than non-SOEs, transcending the mere fulfillment of shareholders’ anticipations for corporate profitability.

6.3. Spatial Spillover Effect of Sci-Tech Finance Pilot Policy

With the improvement of China’s financial market integration and digitization level, the flow of factor resources, capital, and information across different regions is increasingly enhanced. As a result of the deep integration of digital information technology and financial support for innovation activities, Sci-Tech Finance could create cross-regional spillover effects through the spatial diffusion and dissemination of related financial services and digital technologies. This implies that Sci-Tech Finance policies may also have spatial spillover effects on corporates’ ESG behaviors. To specifically explore these spillover effects, with reference to the practice of Grupp (1996) [66] and Gu et al., (2024) [67], this paper constructs the following model:

$${ESG}_{i,t}={\alpha }_0+\rho w{x}_{i,t}+\beta {Controls}_{i,t}+{\gamma }_i+{\mu }_t+{\epsilon }_{i,t}$$

(8)

In Equation (8), the variable ${wx}_{i,t}$ represents the implementation status of Sci-Tech Finance pilot policy in the neighboring regions of the city where enterprise i is located at time t. Given the scarcity of geographically contiguous pilot cities, we have broadened the scope of neighboring regions. For city A, cities that are within the same province or in spatially adjacent provinces are deemed as its neighboring areas. The specific construction of ${wx}_{i,t}$ is as follows: if no cities in the neighboring regions of enterprise i implement Sci-Tech Finance pilot policy at time t, we set ${wx}_{i,t}=0$; if there are m cities in these neighboring regions implementing the policy, then ${wx}_{i,t}=m$. Consequently, the coefficient ρ in Equation (8) can reflect the spatial spillover effect of city-level Sci-Tech Finance pilot policy on enhancing the ESG performance of nearby enterprises. To facilitate a comparison between the direct local effect of Sci-Tech Finance policy on the ESG performance of local enterprises and the spatial spillover effect on nearby enterprises, we incorporate the pilot variable ${did}_{i,t}$ into Equation (8), yielding Equation (9). Thus, ρ and λ represent the spatial spillover impact and the direct regional effect of the pilot policy on enterprises’ ESG activities, respectively, with ρ/λ indicating the ratio of these two effects.

$${ESG}_{i,t}={\alpha }_0+\lambda {did}_{i,t}+\rho w{x}_{i,t}+\sum {Controls}_{i,t}+{\gamma }_i+{\mu }_t+{\epsilon }_{i,t}$$

(9)

The estimation results for Equations (8) and (9) are presented in

Table 12. Spatial spillover effects of pilot policy in nearby cities on corporate ESG.

	(1)	(2)
Items	ESG	ESG
did		0.147 ***
wx	0.012 ***	0.012 ***
${\mathrm{R}}^2$	0.683	0.683
N	13,507	13,507

Note: *** p < 0.01.

, with the coefficient ρ in both equations exhibiting a significant positive impact. The Sci-Tech Finance pilot policy has generated a positive spillover effect on the ESG activities of enterprises in surrounding areas. Furthermore, the coefficient ρ in Equations (8) and (9) remains relatively stable, with a value approximately equal to 0.012. Comparing the estimated value of the direct impact coefficient λ in Equation (9), it is observable that the spatial radiation effect of the Sci-Tech Finance policy on corporate ESG accounts for approximately 8% (0.012/0.147) of the direct impact within the same region.

Based on the spatial spillover analysis design derived from Equation (9), we have conducted grouped research on different types of enterprises, with specific results presented in

Table 13. Spatial spillover effects on different types of enterprises.

Items	Technological Attributes		Factor Intensity			Pollution Attributes		Property Attributes
	HTEs	Non-HTEs	LIEs	CIEs	TIEs	HPIs	Non-HPIs	SOEs	Non-SOEs
wx	0.012 **	0.010	0.010	0.013	0.011 *	0.009	0.012 ***	0.010	0.013 ***
N	6476	5887	3427	3356	5531	3634	8759	4899	7060

Note: *** p < 0.01, ** p < 0.05, and * p < 0.1.

. It can be observed that the coefficient of xw3 exhibits more pronounced statistical significance in HTEs, TIEs, non-HPIs, and non-SOEs. This indicates that Sci-Tech finance policy exhibits more pronounced spatial spillover effects on such enterprises.

From a practical perspective, the Sci-Tech Finance policy can exert a profound impact on enterprises in adjacent areas through the spatial diffusion and radiation of digital financial services and technology financing. The reason for the particularly evident spillover effect in high-tech enterprises is twofold: firstly, they are the primary targets of support under the Sci-Tech Finance policy; secondly, these enterprises possess greater advantages in industrial clustering, remote digital collaboration, and regional cooperation. This results in a stronger spillover effect of the policy on these enterprises, promoting the ESG performance of corresponding enterprises.

For non-HPIs, with natural advantages over HPIs in environmental protection and green technology development, they are positively affected to a greater extent by financial support policies in terms of boosting ESG activities. In addition, non-HPIs tend to have asset-light characteristics, making it easier for them to allocate resources across regions, which results in a significant spatial radiation effect of Sci-Tech Finance policy on the ESG performance of corresponding enterprises.

For non-SOEs, they possess a higher degree of marketization compared to SOEs, which enables non-SOEs to quickly respond to the innovation opportunities and resource support brought about by the Sci-Tech Finance policy. Additionally, non-SOEs have more flexible operational and decision-making mechanisms, which facilitate the allocation and support of cross-regional financial resources by relevant enterprises. For these reasons, the spatial radiation effect of the Sci-Tech Finance pilot policy is more prominent among non-SOEs.

7. Conclusions

Based on the quasi-natural experiment of “Sci-Tech Finance pilot policy” conducted in some cities in China, this paper discusses the impact of Sci-Tech Finance policy on corporate ESG. The empirical results obtained from the multi-period DID model demonstrate that the Sci-Tech Finance pilot policy has significantly enhanced corporate ESG performance. Compared with the non-pilot cities, the ESG score of enterprises in the pilot cities has increased by 3.71%; this conclusion still holds after a variety of robust tests. This paper also investigates the mechanism through which the Sci-Tech Finance pilot policy impacts corporate ESG performance. The study finds that alleviating corporate financing constraints, promoting green innovation, and enhancing corporate total factor productivity play significant intermediary roles in the promotion of corporate ESG performance by Sci-Tech Finance policy. Among these intermediary pathways, alleviating corporate financing constraints and enhancing corporate total factor productivity exhibit relatively stronger effectiveness.

Heterogeneity analysis reveals that the impact of Sci-Tech Finance pilot policy on ESG performance varies with the external environment and enterprise attributes. Specifically, in the eastern region and the regions at high levels of digital economic development, the impact of pilot policy on corporate ESG performance turns out to be more prominent. In technology-intensive enterprises, high-tech enterprises, heavily polluting enterprises, and state-owned enterprises, the impact of pilot policy on ESG performance appears more significant. In addition, spatial spillover effect analysis was conducted to demonstrate that the implementation of Sci-Tech Finance pilot exerts a significant positive radiation effect on the enhancement of corporate ESG performance in the neighboring cities. This radiation effect accounts for about 8% of the direct effect of the region.

Based on the quantitative conclusions presented in this paper, the following policy recommendations are provided to promote corporate ESG practices through the utilization of Sci-Tech Finance.

Firstly, the implementation of Sci-Tech Finance pilot policy plays a crucial role in promoting corporate ESG practices. In the digital economy era, the support of modern digital financial services for scientific and technological innovation continues to strengthen. Government administrators, when advancing the layout and implementation of Sci-Tech Finance pilot policy, should actively focus on the integration of these policies with corporate ESG practices. This can be achieved through incentive measures, such as establishing special funds for ESG innovation practices and strengthening the role of ESG evaluations as a bridge between corporate and capital investments. These efforts aim to provide a high-quality financial support environment for the implementation of ESG concepts and corporates’ sustainable development.

Secondly, government ESG management and related functional departments should actively strengthen green financial supervision, mitigate green financial arbitrage behavior, and promote innovative models and activities for ESG practices. This will better alleviate financing constraints in corporate ESG practices and advance genuine green innovation, therefore enhancing the promoting effect of Sci-Tech Finance activities on corporate ESG practices. Meanwhile, attention should be paid to the support provided by Sci-Tech Finance for corporate innovation activities in multiple dimensions and forms, such as management techniques innovation, business models innovation, and production processes innovation. This will assist in enhancing corporate TFP and operational performance, providing support for the sustainability of corporate ESG practices. Additionally, policymakers need to fully recognize the heterogeneity and development characteristics of different types of enterprises and adopt tailored policies to enhance the efficiency of Sci-Tech Finance support for enterprises’ vitality and ESG practices. For instance, in the case of non-high-tech and labor-intensive enterprises, it is advisable to encourage tech-finance to support the adoption of new technologies, secondary innovation, and business model innovation. For non-state-owned and small-and-medium-sized enterprises, it is recommended to appropriately relax the restrictions on the terms of Sci-Tech Finance policy support. Furthermore, industry-specific ESG performance evaluation systems, practical guidelines for ESG activities, and standards for guiding innovative behaviors should be established in order to more effectively leverage technological finance to promote ESG activities across diverse enterprise types.

Lastly, Sci-Tech Finance exhibits a notable spatial radiation effect on corporate ESG activities. Therefore, it is advisable for adjacent regions to establish and enhance cross-regional financial cooperation mechanisms, thereby facilitating cross-regional financial support for corporate ESG practices and associated innovation initiatives. From the perspective of enterprises themselves, they need to actively embrace Sci-Tech Finance policy and effectively utilize Sci-Tech Finance services and resources both within and outside their respective regions, leveraging them as effective catalysts to promote their ESG behavior and sustainable development.