Sustainable Finance Meets FinTech: Amplifying Green Credit’s Benefits for Banks

Abstract

In recent years, green credit has significantly supported the development of the sustainable economy. However, the existing literature presents differing views on the impact of green credit on bank performance, which is crucial for the sustainability of green credit business. Meanwhile, FinTech is comprehensively empowering green credit business. This paper investigates whether FinTech influences the effect of green credit on bank performance. Based on an analysis of data from 127 Chinese commercial banks from 2007 to 2022, we find that green credit significantly enhances bank performance, and FinTech further amplifies this positive effect. This finding partially explains the conflicting views in the existing literature, as the impact of green credit on bank performance varies under different levels of FinTech. We believe that FinTech exerts its influence through three mechanisms: cost reduction, reputation enhancement, and risk mitigation. Heterogeneity analysis reveals that the impact of FinTech is more pronounced in city commercial banks, in samples with better green credit development, and during banking industry downturns. Finally, we recommend that banks actively develop FinTech and apply it to green credit businesses to maximize the positive effects of green credit. Simultaneously, regulators and governments should provide necessary support for banks.

1. Introduction

Green credit, a vital component of green finance, refers to loans provided by financial institutions that prioritize environmental protection and resource conservation [1,2]. These loans support projects and enterprises that contribute to environmental sustainability, climate change mitigation, and efficient resource use. The main goal of green credit is to drive sustainable development through financial incentives. While definitions may vary by region, the core principle of promoting environmental protection through financing activities remains consistent globally.

China’s green credit market, despite its relatively recent development, has rapidly developed and is now the largest in the world [3]. By the end of 2022, China’s green loans reached CNY 22.3 trillion (approximately USD 3.2 trillion), reflecting a 38.5% year-on-year increase from 2021. This growth outpaced total loan growth by 28.1 percentage points, with green credit accounting for 10.8% of the total loan balance, up 0.5 percentage points from 2021. The impact of green credit has been substantial, supporting the saving of over 600 million tons of coal and reducing more than 1 billion tons of CO₂ emissions annually [4]. The Chinese banking sector has established a comprehensive green credit system and developed unique strategies to promote sustainable finance [5,6].

Why should banks engage in green credit? Some may argue that the recent growth of green credit in China is primarily driven by government initiatives. However, if green credit depends solely on government support, its sustainability could be at risk if that support diminishes in the future. If green credit proves to be profitable, banks are more likely to adopt these practices, which would promote sustainable development and create a positive feedback loop. Therefore, understanding the impact of green credit on bank performance is crucial. However, the existing literature on the relationship between green credit and bank performance is extensive but inconclusive. While most studies suggest a positive impact [7,8,9], others indicate a negative effect [10,11,12], leading to continuing debate among scholars. Further studies are needed to explore the reasons behind these differing results.

In recent years, the rapid advancement of FinTech has transformed the financial industry. Banks now use artificial intelligence, blockchain, cloud computing, big data, and mobile internet technologies to enhance service efficiency and quality. The impact of FinTech on the banking sector is deep and comprehensive, changing not only the way banks operate but also their business models, risk management, customer relationships, and market positioning [13,14]. Firstly, integrating advanced digital technologies allows banks to offer faster and more secure services. Secondly, automation and technological optimization help reduce operational costs and improve efficiency, particularly in credit approval and risk management. Additionally, FinTech significantly enhances the user experience by providing convenient and specialized banking services, thereby increasing customer loyalty and satisfaction [15]. However, FinTech also introduces new challenges, such as data security and privacy issues, and competitive pressure from emerging tech companies [16]. Overall, FinTech is driving the banking industry towards greater efficiency, security, and customer friendliness.

With the application of FinTech in green credit businesses [17], a natural question arises: Does FinTech influence the impact of green credit on bank performance? Can FinTech explain the divergent views in the literature regarding the relationship between green credit and bank performance? These are the primary issues this article aims to explore. While previous studies have examined how FinTech promotes green credit growth [18,19,20], the interaction between FinTech, green credit, and bank performance remains insufficiently studied. This study addresses this gap by investigating the synergistic effects of FinTech and green credit on bank performance, providing valuable insights for banks in developing appropriate FinTech and green credit strategies.

The potential marginal contributions of this study are as follows. First, while the existing literature has examined how FinTech influences green credit, few studies have explored the joint impact of FinTech and green credit on bank performance. This paper attempts to fill this gap and verifies that FinTech indeed affects the impact of green credit on bank performance. Additionally, it reveals that the impact of green credit on bank performance varies under different levels of FinTech, partially explaining the differing opinions about the impact of green credit on bank performance in the existing literature.

Secondly, this study confirms three channels through which FinTech influences the impact of green credit on performance: reducing costs, enhancing reputation, and mitigating risks. This conclusion helps to understand the specific mechanisms of how FinTech promotes the development of sustainable finance.

Thirdly, the heterogeneity analysis reveals that FinTech has a stronger impact on the performance of green credit for city commercial banks and banks with higher levels of green credit development, as well as during periods of lower banking sector prosperity. This provides theoretical evidence for banks to determine the extent to which they should develop FinTech based on their own and the industry’s environmental conditions.

Lastly, this paper constructs a more comprehensive dataset of green credit in China. Previous studies on Chinese green credit have primarily focused on listed banks. In contrast, we collected and integrated data from multiple databases and manually verified green credit data from approximately 127 commercial banks, including many non-listed city and rural commercial banks. As a result, we can explore the impact of green credit on non-listed banks, thereby improving the generalizability and reliability of the conclusions in this paper.

The remainder of this paper is structured as follows: Section 2 provides the literature review and research hypotheses. Section 3 outlines the research design, including the empirical model and data sources. Section 4 presents the main empirical results. Section 5 offers the heterogeneity analysis. Section 6 discusses the robustness tests. Finally, Section 7 concludes with a summary of the findings, policy recommendations, and directions for future research.

2. Literature Review and Hypothesis

2.1. Determinants of Bank Performance

Bank performance is a critical aspect of the financial sector. Numerous studies have found it is influenced by internal financial factors such as bank size, capital adequacy ratio, liquidity, loan growth, asset quality, and management efficiency [21,22,23,24,25,26]. Additionally, customer satisfaction and service quality play significant roles as well [27,28]. Internal corporate governance factors also influence bank performance, including the efficiency of bank branches [29], ownership types [30], cultural diversity, board composition [31,32], executive compensation and human resource management practices [33,34]. These factors significantly impact bank operational efficiency, further affecting overall performance.

External environmental factors should also not be overlooked in their impact on bank performance. These include macroeconomic factors such as interest rates, inflation, exchange rates, and GDP growth [35,36]; policy factors such as regulatory rules, sustainability initiatives, and anti-money-laundering practices [37,38]; market factors such as changes in consumer behavior; and technological factors such as advancements in digital banking.

2.2. Green Credit and Bank Performance

The previous literature has examined the impact of corporate social responsibility and green finance on bank performance [39,40]. As global green credit businesses evolve and data become more comprehensive, research has become increasingly detailed, with many studies focusing on the role of green credit. Numerous papers have identified a positive impact of green credit on bank performance. Luo et al. [41] found that green credit positively influences the core competencies of commercial banks in China, thereby improving their performance. Abuatwan [42] found that integrating green finance into the strategic frameworks of financial institutions can yield environmental benefits and enhance both long-term and short-term sustainability performance, especially in emerging economies such as Palestine. Comparable results have been observed in Indonesia [43], Bangladesh [44], and the Eurozone [45].

Regarding the mechanism, green credit directly improves banks’ financial performance by aligning their activities with environmental sustainability goals [7]. Yuan and Zeng [46] emphasized that green credit supports China’s green development strategy by enabling banks to benefit from government and central bank subsidies and preferential measures, thus increasing profitability. Fenchel et al. [47] suggested that using the environmental performance of borrowers as an indicator for credit risk is advantageous, as it reduces the effort required during the recovery phase and meets cost-effectiveness criteria. Lian et al. [48] demonstrated that green credit enhances financial performance by raising the return rate on interest-bearing assets. Additionally, Xi et al. [49] argued that green credit improves the financial performance of listed banks through value creation and better systems. By embracing sustainability, banks can not only improve their reputation but also contribute positively to environmental and social causes.

Green credit also indirectly improves financial performance by enhancing risk management and credit allocation. Cui et al. [8] highlighted the importance of green credit policies, which mandate banks to offer green credit for environmentally friendly projects while restricting loans to polluting industries. This, in turn, reduces credit risk in the banking sector. Al-Qudah et al. [50] found similar results. Zhou et al. [51] conducted an empirical analysis on the relationship between banks’ green lending and credit risk, demonstrating that green finance regulations can enhance banks’ solvency. Umar et al. [45] highlighted that carbon-neutral lending contributes to lowering credit risk for banks in the Eurozone. As a result, banks can gain advantages by expanding their green lending portfolios, reducing loan loss provisions and the economic capital tied to credit risk, ultimately enhancing their overall performance.

However, some studies suggest that green credit may negatively impact banks’ financial performance. Andaiyani et al. [12] observed that green credit can reduce bank performance, particularly when projects prioritize public welfare over financial gains. This finding emphasis the need to balance environmental goals with financial sustainability in green credit programs. Galán and Tan [11] and Del Gaudio et al. [52] noted that, although green initiatives offer environmental benefits, they can potentially lower banks’ operational efficiency. This trade-off between environmental responsibility and operational effectiveness highlights the complexity of integrating green credit into banking practices. Fata [5] indicated that green credit requires higher initial investments and longer return periods, which can reduce banks’ short-term profitability.

2.3. FinTech and Green Credit

Earlier studies on the impact of FinTech on sustainable development have primarily focused on areas such as green growth, climate change, and energy efficiency [53,54,55]. However, research on the influence of FinTech on green credit and the underlying mechanisms at the micro level remains limited.

Some studies examine the relationship between FinTech and the growth of green credit businesses, generally concluding that FinTech applications help advance the development of green credit [18,56]. Mirza et al. [57] examine a case within the Eurozone banking sector and find a positive correlation between investments in financial technology and green lending. This connection is attributed to the improved search, diligence, and monitoring capabilities offered by new technologies. Gomber et al. [58] argue that banks can enhance their green financial services, optimize transactions and services for greater efficiency and convenience, and increase their motivation to engage in green credit businesses by adopting advanced technological approaches. Muganyi et al. [59] suggest that FinTech accelerates the development of green financial products and enhances the ability of financial institutions to provide green credit. FinTech platforms facilitate the sourcing and use of funds, thereby promoting green project finance. Lv et al. [60] emphasize that FinTech increases the efficiency of green credit distribution, leading to greater adoption of green finance practices. Vergara and Agudo [61] find numerous commonalities between sustainable finance and FinTech, suggesting that FinTech has the potential to enhance the sustainability of the financial sector as a whole by advancing green finance businesses.

Additionally, several studies have explored the impact of fintech on the performance of green credit businesses, with most reporting a positive influence. Hidayat-ur-Rehman and MN Hossain [62] found that fintech adoption and digital transformation have a direct and significant impact on the sustainable performance of banks. Fuster et al. [63] note that FinTech-powered automated credit systems have the potential to improve banks’ efficiency in pre-loan assessments and enable adaptable adjustments to borrower criteria. Yan et al. [64] investigated how FinTech adoption affects the sustainability performance of banking institutions in an emerging economy like Bangladesh. Their findings revealed that green finance and green innovation fully mediate the relationship between FinTech adoption and the sustainability performance of these banks. Bayram et al. [16] and Shang and Niu [19] argued that higher levels of bank digitization significantly increase green credit growth. FinTech also enhances the operational efficiency of green credit in banking by reducing costs, improving risk management, and increasing capital efficiency. Liu and You [20] demonstrated that FinTech addresses critical issues like information asymmetry, thereby improving the allocation efficiency of green credit. Furthermore, Wen [65] emphasized that digital technologies foster the growth of green credit and enable banks to offer a variety of innovative products to meet the needs of environmentally conscious consumers, thus enhancing performance. Guang-Wen and Siddik [66] analyzed data from the pandemic period and suggested that integrating FinTech into the regular operations of banking institutions is essential for achieving sustainable environmental performance during the crisis. Bhuiyan et al. [67] proposed that the adoption of FinTech enhances sustainability by expanding access to green financing, improving transparency, reducing costs, and promoting broader financial inclusion.

2.4. Literature Discussion

While the existing literature offers substantial insights into the determinants of bank performance, the impact of green credit, and the role of FinTech in enhancing financial operations, several critical gaps remain. First, there is still no consensus on the impact of green credit on bank performance. Second, most studies have focused on isolated aspects of these relationships and have not fully explored how FinTech influences the impact of green credit on bank performance. Although some studies have examined the relationship between FinTech and green credit, suggesting that FinTech can promote the growth of green credit, this is different from the question our study addresses. Their research investigates how FinTech affects the green credit business, while our study examines how FinTech and green credit jointly impact bank performance, focusing on their synergistic effect. This study in this paper has significant implications for banks in formulating and selecting suitable FinTech and green credit development strategies. Therefore, this study examines how FinTech influences the impact of green credit on bank performance, highlighting the importance of FinTech in promoting sustainable finance and attempting to explain the differing perspectives in the existing literature on the impact of green credit on bank performance.

2.5. Research Hypothesis

2.5.1. The Impact of Green Credit on Bank Performance

The existing literature suggests that green credit can significantly enhance the financial performance of banks. By funding environmental projects, banks can improve their reputation and attract environmentally conscious customers who prefer green banking products and services [68]. This leads to a stronger brand image, increased customer loyalty, and greater market competitiveness [69]. A stronger brand image contributes to higher customer retention and acquisition rates, directly increasing the bank’s revenue streams.

Additionally, green credit reduces banks’ risk exposure [70]. Environmental projects often have lower default risks due to their sustainable nature and long-term viability, decreasing the likelihood of non-performing loans and improving asset quality [71,72]. Furthermore, green credit aligns with government policies and incentives, providing banks with additional financial benefits through subsidies, tax breaks, and other forms of support. These incentives enhance the profitability of green credit and help offset some of the initial operational costs associated with environmental projects.

However, it is important to acknowledge the challenges posed by green credit. The strict evaluation and management processes required for environmental projects can lead to increased operational costs [10]. Moreover, the longer return cycles associated with some environmental projects may delay the realization of expected returns, potentially affecting short-term financial performance.

Despite these challenges, the positive impacts of green credit—such as enhanced reputation, reduced risk exposure, and alignment with governmental incentives—are likely to outweigh the negative effects. Therefore, we propose the following hypothesis:

H1. 

Green credit can enhance bank financial performance.

2.5.2. The Role of FinTech in the Impact of Green Credit on Bank Performance

The integration of advanced FinTech solutions into green credit processes can significantly enhance the positive impact of green credit on bank performance. FinTech simplifies complex and time-consuming approval processes, improving operational efficiency and reducing both internal costs and external regulatory pressures. By utilizing technologies such as big data analytics and artificial intelligence, banks can more effectively process and analyze environmental data, leading to more informed and accurate lending decisions.

In addition, FinTech supports the creation of innovative financial products that meet the growing demand for sustainable investments. These products, such as green bonds and sustainability-linked loans, attract environmentally conscious customers, enhancing satisfaction and loyalty. This, in turn, broadens the customer base and generates additional revenue streams for banks [73].

Furthermore, FinTech improves banks’ ability to manage risks associated with green credit. Technologies like AI-driven risk assessment tools and blockchain enhance data security and provide more accurate methods for evaluating and monitoring loans. This enables banks to respond more effectively to potential issues, reducing financial risks and increasing the likelihood of successful project outcomes [20].

In summary, the combination of FinTech and green credit not only enhances operational efficiency and risk management but also improves green reputation, leading to increased income, all of which contribute to better bank performance. Therefore, we propose the following hypothesis:

H2. 

FinTech amplifies the positive impact of green credit on bank performance.

Hypothesis H2 is central to our study. Building on the above discussion, we suggest that FinTech influences the impact of green credit on performance primarily through cost reduction, enhanced reputation, and improved risk management. These aspects are further explored in detail in the following hypotheses.

2.5.3. Cost Reduction Mechanism

Implementing green credit involves various costs for banks, which can be categorized into internal and external expenses. Internal costs include employee training, risk assessment and management of green projects, the creation and maintenance of credit approval systems, and research and development of green products [8]. External costs include fees for third-party environmental assessments and certifications, legal consultations to ensure compliance with environmental regulations, and expenses related to regulatory communications and reporting [74].

The adoption of FinTech technologies has the potential to significantly reduce both internal and external costs associated with green credit. Internally, FinTech enables the integration of financial institution systems with green information platforms, simplifying complex approval procedures in green credit businesses and enhancing overall efficiency. The use of big data for real-time data collection provides robust, data-driven support for evaluating green credit projects, lowering costs associated with identifying and assessing potential risks. Additionally, automated credit processes enabled by FinTech reduce the need for manual labor, cutting labor costs and accelerating decision-making processes [75]. Moreover, the implementation of cloud computing and blockchain technologies can significantly reduce the maintenance costs of internal systems while simultaneously enhancing data security [76].

Externally, FinTech provides banks with advanced data collection and analysis tools, reducing the dependency on costly third-party environmental assessments. Furthermore, the application of regulatory technologies (RegTech) simplifies and automates compliance processes, thus decreasing legal consultation fees and minimizing the risk of non-compliance penalties. The availability of online tools for market research allows banks to acquire critical market and customer data more cost-effectively than traditional methods.

By reducing both internal and external costs, FinTech not only enhances the operational efficiency of green credit businesses but also contributes to the overall financial performance of banks involved in green credit. The cost savings achieved through FinTech adoption can be redirected to other strategic initiatives, thereby amplifying the positive impact of green credit on bank performance.

Based on the above analysis, we propose the following hypothesis:

H3. 

FinTech amplifies the positive impact of green credit on bank performance by reducing the associated costs with green credit businesses.

2.5.4. Reputation Enhancement Mechanism

Green credit businesses can significantly enhance a bank’s green reputation, which positively impacts overall bank performance. By positioning themselves as environmental leaders, banks engaged in green credit activities can shape a positive brand image. As environmental awareness grows among consumers and enterprises, there is an increasing preference for banks with strong green reputations. Consequently, offering green credit products enables banks to attract environmentally conscious clients, increasing deposits, loan activities, and other services, which enhances overall income [77].

The development of FinTech further amplifies the impact of green credit on building a green reputation. FinTech provides banks with advanced tools like targeted advertising and intelligent marketing, significantly increasing their exposure and public recognition. Through digital platforms and social media, banks can engage in extensive promotional activities, cultivating a more positive corporate image. Additionally, blockchain technology plays a crucial role by ensuring transparency in green financial activities, bolstering customer trust and confidence in the bank [78].

Moreover, FinTech enhances the development and delivery of green financial products and services, adding substantial value to the green credit business. Technologies such as big data and cloud computing enable in-depth data analysis, allowing banks to better identify and meet diverse green financial needs [37]. By aligning customer-specific needs with risk preferences, banks can develop scenario-based, intelligent green credit products that are more tailored and appealing to a broader customer base, thus expanding their market reach and enhancing revenue. Examples of such innovations include blockchain-based green bonds and smart contract-supported sustainable investments [64,79,80]. Furthermore, advanced green finance data analysis tools provide critical insights into both environmental impact and financial returns, enriching the bank’s product offerings and driving revenue growth [81].

The integration of FinTech with green credit businesses thus builds a more robust green reputation, which in turn has a compounded positive effect on bank performance, particularly in revenue generation. Given these considerations, we propose the following hypothesis:

H4. 

FinTech amplifies the positive impact of green credit on bank performance by establishing a stronger green reputation.

2.5.5. Risk Mitigation Mechanism

Green credit is associated with several risks, both pre-loan and post-loan. Pre-loan risks include information asymmetry, where borrowers may not disclose accurate information about their projects; greenwashing risk, where borrowers might falsely claim their projects are environmentally friendly; and project assessment risk, where banks may overestimate a project’s environmental benefits or market potential [8]. Post-loan risks arise when market or macroeconomic conditions change after the loan is issued. These include technological risks from immature or unfeasible technology, market risks from changes in product demand, environmental risks from natural disasters or climatic changes, and regulatory risks from changes in relevant policies and laws [9].

FinTech significantly enhances banks’ ability to manage pre-loan risks associated with green credit by leveraging big data and artificial intelligence. These advanced technologies improve the precision of risk assessment by integrating diverse data sources, enabling banks to better evaluate the environmental and market potential of projects. As a result, banks can avoid high-risk investments, increasing the likelihood of project success and minimizing the occurrence of non-performing loans [82,83]. Furthermore, FinTech tools are instrumental in detecting fraud and ensuring compliance with regulations, reducing the chances of banks falling victim to greenwashing. By addressing information asymmetry and lowering the costs associated with identifying greenwashing risks, FinTech improves the efficiency of credit resource allocation [84].

In addition, FinTech plays a crucial role in managing post-loan risks through continuous monitoring and early warning systems. These technologies enable banks to track real-time data on the operations of funded projects, such as shifts in market demand and cash flows. Continuous monitoring allows banks to promptly identify and respond to potential issues, reducing the likelihood of significant losses. Moreover, the dynamic nature of FinTech-enabled risk management tools allows for the early detection of anomalies, enhancing banks’ ability to mitigate risks during the loan period, reduce risk-related losses, and accelerate the resolution of emerging risks.

Based on the analysis, it is evident that FinTech can efficiently manage the risks of green credit, thereby amplifying its positive effect on bank performance. Hence, we propose the following hypothesis:

H5. 

FinTech amplifies the positive impact of green credit on bank performance by reducing the risks associated with green credit businesses.

Drawing from the preceding analysis and hypotheses, we have constructed a mechanism design diagram that illustrates the relationships between green credit, FinTech, and bank performance, as shown in Figure 1, highlighting the significant interactions among these variables.

3. Methodology and Data

3.1. Econometric Modeling

To test hypotheses H1 and H2, we constructed the following two regression models:

$${Performance}_{it}={\beta }_0+{\beta }_1{GC}_{it}+{\beta }_2{Control}_{it}+{\mu }_i+{\rho }_t+{\epsilon }_{it},$$

(1)

$$\begin{gathered} {Performance}_{it}={\beta }_0+{\beta }_1{GC}_{it}+{\beta }_2{FT}_{it}+{\beta }_3{GC}_{it}\times {FT}_{it}+{\beta }_4{Control}_{it}+{\mu }_i+ \\ {\rho }_t+{\epsilon }_{it}, \end{gathered}$$

(2)

where i represents the bank, t represents time, the dependent variable ${Performance}_{it}$ is the bank’s performance indicator, ${GC}_{it}$ is the bank’s green credit indicator, ${FT}_{it}$ is the bank’s FinTech indicator, and ${Control}_{it}$ represents control variables. Furthermore, to mitigate potential estimation biases caused by unobserved individual and temporal factors, we introduce individual fixed effects ${\mu }_i$ and time fixed effects ${\rho }_t$ into the regression model.

According to Hypothesis H1, if green credit improves bank performance, the coefficient ${\beta }_1$ in the regression results of Equation (1) should be significantly positive. According to Hypothesis H2, if the development of FinTech further enhances the positive impact of green credit on bank performance, the coefficient ${\beta }_3$ in the regression results of Equation (2) should also be significantly positive. Additionally, since the inclusion of interaction terms alters the meaning of the main effect coefficient ${\beta }_1$ in Equation (2), we adjust the green credit and FinTech variables by subtracting their respective sample means before regression. After this adjustment, the main effect coefficient ${\beta }_1$ indicates the marginal impact of green credit on bank performance when FinTech is at its average level.

Next, we use different methods to examine the three channels through which FinTech enhances the impact of green credit on bank performance. To test the cost reduction channel (Hypothesis H3), we constructed the following regression equation:

$${Cost}_{it}={\beta }_0+{\beta }_1{GC}_{it}+{\beta }_2{FT}_{it}+{\beta }_3{GC}_{it}\times {FT}_{it}+{\beta }_4{Control}_{it}+{\mu }_i+{\rho }_t+{\epsilon }_{it}$$

(3)

Equation (3) mirrors Equation (1), except that the dependent variable has been replaced by the bank cost indicator ${Cost}_{it}$. The equation suggests that the marginal cost of green credit is ${\beta }_1+{\beta }_3{FT}_{it}$. The derivative of the marginal cost of green credit with respect to FinTech, denoted by ${\beta }_3$, indicates the effect of FinTech on the marginal cost of green credit. According to the Hypothesis H3, if FinTech reduces the costs associated with green credit businesses, then as the level of FinTech increases, the negative impact of green credit on costs will become more significant. Therefore, the coefficient ${\beta }_3$ in Equation (3) is expected to be significantly negative.

Regarding the reputation enhancement channel (Hypothesis H4), we employ a grouped regression method for testing. Green reputation is often difficult to quantify since it refers to the public image and perception of a bank regarding environmental protection and sustainable development. If a bank’s FinTech can enhance its green reputation, customers will have greater confidence in the bank’s capabilities, leading them to consume more non-interest services at the bank. Using the median as the dividing line, we split the entire sample into two groups based on the proportion of fee and commission income. We then perform grouped regression by Equation (2). If Hypothesis H4 holds, the sample with a higher proportion of fee and commission income (more reliant on non-interest business) will show a more pronounced effect of FinTech in strengthening the impact of green credit on bank performance; i.e., ${\beta }_3$ will be larger. It is important to note that this paper does not group based on the non-interest income ratio because several components of non-interest income, such as investment income, fair value gains, exchange gains, and other business income, are unrelated to green reputation.

We employ a similar method to examine the risk mitigation channel (Hypothesis H5), categorizing banks into high-risk and low-risk groups based on their risk indicators and using Equation (2) for grouped regression. According to Hypothesis H5, FinTech can reduce pre-loan and post-loan risks in green credit businesses. Therefore, we believe that the risk-mitigating effect of FinTech is more pronounced in high-risk banks. In other words, in the high-risk group, the effect of FinTech in enhancing the impact of green credit on bank performance will be stronger, as indicated by a larger ${\beta }_3$.

3.2. Variable Definitions

First, we select bank performance indicators. In this study, we assess the financial performance of commercial banks using three widely recognized indicators: Return on Assets (ROA), Return on Equity (ROE), and Net Profit Margin (NPM). These indicators were selected because they provide a comprehensive view of bank performance from different financial dimensions. Specifically, ROA reflects the efficiency with which a bank utilizes its total assets to generate profits, offering insights into overall asset management. ROE, on the other hand, measures the return generated on shareholders’ equity, which is crucial for assessing the value delivered to investors. NPM provides a perspective on the bank’s operational efficiency by evaluating the profitability relative to revenue. These indicators are commonly used in the literature to capture different aspects of financial performance, ensuring a robust evaluation. In our empirical analysis, we conduct separate regressions using each of these indicators as dependent variables to comprehensively explore the relationship between green credit, FinTech, and financial performance.

Next is the selection of the bank green credit indicator (GC). The existing literature mainly employs two indicators: the logarithmic value of the green credit balance and the ratio of green credit to total loans. The logarithmic value of the green credit balance (lnGC) represents the scale of a bank’s green credit business, while the green credit ratio (GCR) measures the proportion of green credit relative to the total loan portfolio, providing a scale-independent perspective. However, preliminary data analysis revealed significant fluctuations in the green credit ratio for small banks. In these banks, issuing or recovering a single green credit loan can cause large fluctuations in the green credit ratio due to their smaller loan portfolios. These fluctuations might misleadingly suggest substantial changes in the bank’s green credit business when no such change has occurred. To mitigate this risk and provide a more stable measure, we have chosen the logarithmic value of the green credit balance (lnGC) as the primary indicator. This choice is further supported by the observation that banks with larger green credit balances generally exhibit stronger development in their green credit businesses. To ensure the robustness of our results, we also include the green credit ratio (GCR) in subsequent robustness checks, thereby validating the consistency of our findings across different measurement approaches.

Regarding bank FinTech, we employ three indicators for measurement, detailed as follows:

• The news-based FinTech index (FTNews): The existing literature supports the use of internet search indices to measure banks’ FinTech levels [85,86]. Since banks frequently advertise their FinTech capabilities and involvement in FinTech projects, a higher number of reports related to a bank’s FinTech activities generally indicates a higher level of FinTech expertise. In China, Baidu News is a prominent news aggregation platform that sources information from over 500 authoritative websites, providing a broad spectrum of coverage. We employed web crawling technology to conduct searches combining each commercial bank’s name with FinTech-related keywords. This methodology allowed us to quantify the number of news pages referencing a bank’s FinTech activities annually. We then summed the news articles containing FinTech keywords for each bank within a given year and took the logarithm of this total to derive the annual FinTech index for each bank. For selecting FinTech keywords, we referred to significant documents such as the “13th Five-Year National Science and Technology Innovation Plan″ and the “China FinTech Development Report (2018)”, along with relevant key news and conferences. These keywords span fundamental FinTech technologies and applications under the “ABCD+” framework [87], with a total of 48 keywords listed in

  Table 1. FinTech keywords.

  Major Category	Subcategory	Keywords
  ABCD + Core Technologies	Artificial Intelligence	Artificial Intelligence, Machine Learning, Deep Learning, Cognitive Computing, Natural Language Processing, Voice Recognition, Image Understanding, Intelligent Data Analysis
  	Blockchain	Blockchain, Smart Financial Contracts, Digital Currency
  	Cloud Computing	Cloud Computing, Converged Architecture, Distributed Computing, In-Memory Computing, Stream Computing, Green Computing, Billion-level Concurrency
  	Big Data	Big Data, Data Mining, Data Visualization, Text Mining, Business Intelligence, Cyber–Physical Systems, Heterogeneous Data, Brain-Inspired Computing, Graph Computing
  	Security Technology	Differential Privacy Technology, Multi-party Secure Computing, Biometric Technology, Identity Verification, EB-level Storage
  FinTech Applications	Payment Technologies	NFC Payments, Third-Party Payments, Mobile Payments
  	Investment and Financing Technologies	Equity Crowdfunding, Quantitative Finance, Investment Decision Support Systems, Intelligent Investment Advisory
  	Intelligent Services and Data Management	Internet Finance, Open Banking, Intelligent Customer Service, Credit Reporting, Internet of Things, Connected Networks, Virtual Reality, Mobile Internet

  . The FTNews index captures the bank’s public-facing FinTech efforts, which are indicative of its FinTech development and market presence.
• The Peking University China Commercial Bank Digital Transformation Index (FTDigital): The FTDigital index, developed by Peking University, provides a comprehensive depiction of digital transformation trends among Chinese commercial banks. This index evaluates banks across three dimensions and eight aspects, offering a nuanced measure of FinTech development [88]. It has been utilized in previous studies as a reliable proxy for bank FinTech engagement [19].
• The patent application-based FinTech Index (FTPatent): We constructed a FinTech index based on bank FinTech patent application data. First, we collected patent application information from the China National Intellectual Property Administration’s patent search database for banks up to 2022. We collected a total of 38,407 patent application records. Each record includes details such as the patent applicant, application date, and a patent abstract, which discloses the primary purpose and technology of the patent. Second, we matched the patent applicants with commercial bank samples. Specifically, patent applications from bank branches were treated as the whole bank’s applications. Third, we searched each patent application abstract for FinTech-related keywords from Table 1. If an abstract contained any of these keywords, the patent was classified as a FinTech patent. Finally, we calculated the cumulative number of FinTech patents applied for by each bank and took the logarithm of this count as the FinTech index. A higher index reflects better development and application in FinTech by the bank.

To summarize, these three FinTech variables cover various aspects of financial technology, each providing a comprehensive reflection of the overall level of FinTech development in banks. However, due to differing construction methods, these variables vary in data volume. The Peking University China Commercial Bank Digital Transformation Index, based on external data such as annual reports and patent information from commercial banks, offers extensive coverage and includes many non-listed banks in China. However, its data update slowly, with the most recent update only extending to 2021. The patent application-based FinTech index directly captures a bank’s FinTech achievements, but since some commercial banks have not applied for FinTech patents, this index may not fully represent their actual level of FinTech development. In contrast, the news-based FinTech index is more comprehensive in terms of data coverage, as most banks are featured in relevant news reports, offering a more accurate representation of their FinTech development. Considering these factors, we select the news-based FinTech index as the primary indicator of FinTech, while the other two indices are used for robustness analysis.

We include a range of control variables that capture both bank-specific characteristics and macroeconomic conditions, as these factors are commonly recognized in the literature as significant determinants of bank performance. Bank-specific characteristic variables include bank size (Size), capital-to-asset ratio (CAP), interest-generating asset ratio (IGAR), total asset turnover ratio (TAT), loan concentration Herfindahl–Hirschman index (LoanHHI), and short-term asset liquidity ratio (LIQ). Macroeconomic variables include the real GDP growth rate (RGDP) and the consumer price index (CPI). Note that the two macroeconomic control variables in this study are provincial-level indicators, so they are not absorbed by time fixed effects.

Table 2. Variable definitions.

Type	Variable	Description	Calculation Method
Explained Variables	ROA	Return on Assets	Net profit/Total assets
	ROE	Return on Equity	Net profit/Total equity
	NPM	Net Profit Margin	Net profit/Total revenue
Green Credit Variables	lnGC	Scale of green credit	The natural logarithm of green loans
	GCR	Green Credit Ratio	Green loans/Total loans
FinTech Variables	FTNews	FinTech Index Based on News	The natural logarithm of one plus the number of FinTech news articles
	FTDigital	Peking University Commercial Bank Digital Transformation Index	Obtained from Xie and Wang [88]
	FTPatent	FinTech Index Based on Patent Application Data	The natural logarithm of one plus the cumulative number of FinTech patent applications
Bank Control Variables	Size	Bank Size	The natural logarithm of total assets
	CAP	Capital-to-asset ratio	Total Equity/Total assets
	IGAR	Interest Generating Asset Ratio	Income generating assets/Total assets
	TAT	Total Asset Turnover	Total revenue/Total assets
	LoanHHI	Loan Concentration	Herfindahl-Hirschman Index based on bank loans from 18 industries
Macroeconomic control variables	GDPR	Real GDP Growth Rate	Annual growth rate of real GDP
	CPI	Consumer Price Index	Annual change in consumer price index
Other variables	CIR	Cost-to-Income Ratio	Operating Expenses/Total Operating Income
	AdminCost	Administrative Expense Ratio	Administrative Expenses/Total Operating Income
	FeeIncome	Net fee and commission income ratio	Net fee and commission income/Total Operating Income
	NPLR	Non-Performing Loan Ratio	Non-Performing Loans/Total Loans
	ODLR	Overdue Loan Ratio	Overdue Loans/Total Loans
	RWAR	Risk-Weighted Asset Ratio	Total Capital/Risk-Weighted Assets

illustrates the definitions and calculation methods of the primary variables.

In our mechanism analysis, we carefully selected variables corresponding to different channels. For the cost channel, we used the cost-to-income ratio (CIR) and the administrative cost ratio (AdminCost) as key indicators. CIR measures a bank’s operational efficiency by comparing operating costs to operating income, with a lower CIR indicating better efficiency. AdminCost reflects the proportion of operating expenses attributable to administrative activities, highlighting the bank’s ability to control overhead costs. For the reputation channel, we used the fee and commission income ratio (FeeIncome) to evaluate a bank’s reliance on non-interest income activities. A higher FeeIncome ratio enables the bank to derive greater performance benefits from its reputation. For the risk channel, we incorporated three risk indicators: the non-performing loan ratio (NPLR), the overdue loan ratio (ODLR), and the risk-weighted asset ratio (RWAR). NPLR measures the proportion of loans not generating income due to borrowers’ failure to repay. ODLR represents the percentage of loans overdue but not yet classified as non-performing. RWAR reflects the level of risk in the bank’s asset portfolio, considering the varying risk levels of different asset types. These variables comprehensively reflect the bank’s risk profile.

3.3. Data Sources and Sample Selection

The fundamental information and financial data of banks, along with macroeconomic variables, are derived from the Wind database. For green credit data, we used a combination of database extraction and manual verification. The Wind, CSMAR, and CNRDS databases, which are considered authoritative economic and financial sources in China, include green credit data of Chinese banks. However, these databases exhibit variations in sample coverage and frequency. They may occasionally show inconsistencies in data for the same bank within the same year. Therefore, we combined green credit data from these three databases and conducted manual checks and modifications based on annual reports and ESG disclosures from banks.

Based on the available public data, this study’s sample includes 127 commercial banks, covering the period from 2007 to 2022. This timeframe aligns with significant developments in green credit and FinTech in China. The sample consists of 6 state-owned commercial banks, 9 joint-stock commercial banks, 80 city commercial banks, 29 rural commercial banks, and 3 private banks. The number of banks in this sample is substantial compared to previous studies. During the sample period, the total assets of these commercial banks accounted for 85% to 92% of the total assets of all commercial banks in China, demonstrating the strong representativeness of the sample. Furthermore, to mitigate the negative effects of outliers on regression analysis, variables containing outliers were winsorized at the top and bottom 1%.

3.4. Descriptive Statistics

Figure 2 illustrates the trends in green credit balances across different types of banks from 2007 to 2022. The y-axis uses a logarithmic scale to better display data covering multiple levels of magnitude. From 2009 to 2022, different types of banks experienced rapid growth in green credit. The Compound Annual Growth Rate of green credit balances (CAGR) was 24.67% for state-owned banks, 35.81% for joint-stock banks, 71.57% for city commercial banks, and 91.23% for rural commercial banks. Private banks, starting from a smaller base in 2016, also showed a robust CAGR of 49.65% through 2022.

By the end of 2022, state-owned banks held the largest share of green credit at 74.03%. Joint-stock banks followed with 16.51%, and city commercial banks accounted for 8.12%. Although smaller, rural commercial banks and private banks held 1.33% and 0.01%, respectively. This distribution highlights the leading role of state-owned banks in China’s green finance sector, while smaller banks are gradually increasing their participation.

Figure 3 shows the changes in the ratio of green credit to total loans among various types of banks from 2007 to 2022. Starting in 2007, most bank types exhibited a steady increase in the proportion of green credit within their total loan portfolios. State-owned banks led, reaching a ratio of 13.05% by 2022, followed by joint-stock banks at 7.06%. City commercial banks and rural commercial banks also raised their green credit ratios to 5.87% and 3.65%, respectively. Private banks initially experienced growth in green credit, but their ratio declined over time, falling to 0.58% in 2022. This decrease is due to the slower growth of green credit compared to the rapid expansion of their total loan portfolios.

Overall, the banking sector in China is rapidly developing its green credit business. The involvement of various banks highlights their strong commitment to promoting sustainable economic growth and contributing to China’s environmental goals.

Table 3. Descriptive statistics.

Variables	Obs	Mean	Std	Min	P25	Median	P75	Max
ROA (%)	569	0.766	0.311	0.043	0.547	0.772	0.974	1.821
ROE (%)	569	11.274	5.275	0.505	7.411	10.881	14.501	28.579
NPM (%)	569	29.726	9.496	2.548	22.553	31.447	36.864	52.877
lnGC	569	20.213	2.302	14.509	18.644	20.030	21.442	26.709
GCR (%)	568	3.721	3.583	0.042	1.314	2.708	4.924	28.470
FTNews	529	4.232	1.034	0.693	3.526	4.205	4.927	6.926
FTDigital	468	81.213	34.814	3.297	55.877	78.447	99.598	197.142
FTPatent	569	0.487	1.293	0.000	0.000	0.000	0.000	8.067
Size	569	8.564	1.737	4.458	7.360	8.262	9.512	12.889
CAP (%)	569	7.207	1.582	3.093	6.245	7.136	8.042	17.205
IGAR (%)	569	88.646	10.698	44.523	82.338	92.890	97.423	99.233
TAT (%)	569	2.545	0.581	1.050	2.160	2.530	2.900	5.160
LoanHHI	569	0.195	0.063	0.111	0.152	0.179	0.225	0.534
LIQ (%)	569	61.838	21.569	28.280	45.810	56.680	73.540	133.490
RGDP (%)	569	6.310	2.616	−5.000	3.900	6.900	8.000	14.231
CPI	569	2.080	0.943	−0.700	1.525	2.003	2.500	5.900
CIR (%)	569	33.987	7.627	19.184	28.570	32.960	37.680	66.470
AdminCost (%)	566	33.889	7.787	18.987	28.515	32.922	37.533	65.674
FeeIncome (%)	569	8.803	8.259	−9.248	3.055	8.224	14.637	28.692
NPLR (%)	566	1.526	0.633	0.015	1.090	1.470	1.810	4.690
ODLR (%)	515	2.390	1.808	0.297	1.277	1.957	2.784	18.868
RWAR (%)	549	65.209	8.014	27.278	60.928	65.572	70.434	85.668

presents the descriptive statistics of the primary variables in this study. The results indicate that all variable values fall within a reasonable range. The average Return on Assets (ROA) is 0.77%, the average Return on Equity (ROE) is 11.27%, and the average Net Profit Margin (NPM) is 29.73%. There are significant numerical discrepancies among performance indicators due to their different denominators. Moreover, there is a notable disparity in the scale and proportion of green credit among different banks. The natural logarithm of the green credit balance (lnGC) ranges from 14.51 to 26.71, indicating that the green credit balance varies from approximately CNY 20 million to nearly CNY 400 billion. The green credit ratio (GCR) fluctuates between 0.04% and 28.47%. Despite the rapid development of green credit in China’s banking industry, its proportion remains relatively low, with an average value of only 3.72%. This suggests considerable room for growth and market opportunities in China’s green credit sector.

Regarding the control variables, the average value of CAP is 7.21%, indicating a high leverage ratio in the banking industry. The average IGAR is 88.65%, while FeeIncome has an average of 8.80%, demonstrating that the Chinese banking sector still heavily relies on traditional interest-based business. LoanHHI averages at 0.195, suggesting a high level of diversification and low concentration in loan industries. The average value of NPLR is 1.53%, with a maximum value of 4.69%. According to Chinese regulations, the non-performing loan ratio should be below 5%, indicating that the credit risk in the Chinese banking industry is controlled and remains at a manageable level.

Table 4. Pearson correlation between key variables.

	ROA	ROE	NPM	lnGC	GCR	FTDigital	FTNews	FTPatent
ROA	1.000
ROE	0.901 ***	1.000
NPM	0.827 ***	0.764 ***	1.000
lnGC	0.345 ***	0.291 ***	0.269 ***	1.000
GCR	0.130 ***	0.039	0.132 ***	0.535 ***	1.000
FTDigital	−0.035	−0.214 ***	−0.006	0.434 ***	0.177 ***	1.000
FTNews	0.023	−0.121 ***	0.035	0.617 ***	0.241 ***	0.670 ***	1.000
FTPatent	0.212 ***	0.075 *	0.156 ***	0.637 ***	0.261 ***	0.482 ***	0.514 ***	1.000

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

lists the correlation coefficients of the key variables. First, the correlation coefficients among indicators of the same type are relatively high. For instance, the performance indicators of banks are significantly positively correlated, with all correlation coefficients exceeding 0.7. Second, there is a significant positive correlation between bank performance and the scale of green credit. In contrast, the positive correlation between bank performance and the ratio of green credit is weaker. This may indicate that expanding the scale of green credit is more effective in enhancing bank performance than increasing the proportion. Furthermore, the relationship between FinTech and bank performance is not clearly defined. Some FinTech indicators are positively correlated with performance, while others are negatively correlated. This suggests that the impact of FinTech on bank performance may depend on the nature of the technology and its implementation within bank operations. Lastly, there is a positive correlation between FinTech and green credit. This indicates that banks with higher levels of FinTech tend to have better development in green credit business, consistent with existing literature findings [19,20].

4. Empirical Results and Analysis

4.1. Benchmark Regression

Table 5. The impact of green credit on bank performance and the role of FinTech.

	(1)	(2)	(3)	(4)	(5)	(6)
	ROA	ROE	NPM	ROA	ROE	NPM
lnGC	0.029 ***	0.506 ***	1.309 ***	0.009	0.266	0.332
	(2.76)	(3.26)	(2.84)	(0.82)	(1.58)	(0.67)
lnGC × FTNews				0.013 ***	0.146 ***	0.680 ***
				(5.11)	(3.81)	(6.05)
FTNews				0.002	0.139	0.131
				(0.10)	(0.59)	(0.19)
Size	0.017	0.497	2.994	0.146 **	1.900 **	9.622 ***
	(0.35)	(0.70)	(1.42)	(2.58)	(2.28)	(3.94)
CAP	−0.011	−1.269 ***	−0.528	−0.014	−1.446 ***	−0.598
	(−1.45)	(−11.09)	(−1.55)	(−1.63)	(−11.63)	(−1.64)
IGAR	0.002 ***	0.011	0.047	0.003 ***	0.022 **	0.095 ***
	(2.91)	(1.05)	(1.53)	(4.21)	(2.04)	(3.05)
TAT	0.219 ***	2.624 ***	−1.923 **	0.228 ***	2.833 ***	−1.847 **
	(10.70)	(8.81)	(−2.17)	(11.06)	(9.39)	(−2.08)
LoanHHI	0.059	1.732	4.926	−0.032	0.892	0.621
	(0.36)	(0.72)	(0.69)	(−0.19)	(0.37)	(0.09)
LIQ	0.000	−0.001	−0.000	0.001	0.006	0.020
	(0.45)	(−0.18)	(−0.01)	(1.41)	(0.88)	(1.05)
RGDP	−0.004	−0.077	−0.308	−0.001	−0.004	−0.179
	(−0.86)	(−1.04)	(−1.41)	(−0.19)	(−0.05)	(−0.81)
CPI	−0.003	0.182	−0.284	0.003	0.331	−0.091
	(−0.16)	(0.77)	(−0.41)	(0.21)	(1.37)	(−0.13)
Constant	0.027	10.369 *	16.665	−1.011 **	−1.522	−34.739 *
	(0.07)	(1.76)	(0.95)	(−2.14)	(−0.22)	(−1.70)
Ind Effect	Yes	Yes	Yes	Yes	Yes	Yes
Year Effect	Yes	Yes	Yes	Yes	Yes	Yes
Obs	569	569	569	529	529	529
R2	0.620	0.806	0.320	0.652	0.824	0.378
Adj-R2	0.483	0.736	0.076	0.529	0.763	0.161

Notes: T-Statistics are reported in parentheses. *, **, and *** represent significance levels of 10%, 5%, and 1%, respectively.

presents the benchmark regression results regarding the impact of green credit on bank performance and the interaction effect of green credit and FinTech. Columns (1) to (3) show results based on Equation (1), using different bank performance indicators in each column. The results show that the coefficients of green credit are positive and statistically significant at the 1% level, indicating that an increase in green credit enhances the financial performance of banks, thus supporting Hypothesis H1. Specifically, a 1% increase in green credit raises ROA by 0.029%, ROE by 0.506%, and NPM by 1.309%, demonstrating the economic significance of green credit on bank performance.

Columns (4) to (6) present regression results based on Equation (2). They show that the interaction coefficients between green credit and FinTech are positive and statistically significant at the 1% level. This indicates that as the level of FinTech increases, the positive impact of green credit on bank performance is enhanced, supporting Hypothesis H2.

To more intuitively illustrate the interaction effect, we plotted the marginal effects of green credit on bank performance at different levels of FinTech, as shown in Figure 4. Subplots (a) to (c) represent the marginal effects with different bank performance indicators. The solid lines denote the estimated marginal effects, and the dashed lines represent the upper and lower bounds of the 95% confidence intervals. Subplot (d) depicts the kernel density curve of FinTech. The results reveal significant differences in the impact of green credit on performance at varying levels of FinTech. When FinTech is at a low level, the impact of green credit on bank performance is negative but not significant. When FinTech is at an average level, the impact is positive but not significant, consistent with the main effect coefficients in Table 5. Only at a high level of FinTech does the impact of green credit on bank performance become significantly positive.

Based on the results of Figure 4, we believe the different perspectives in the literature on the impact of green credit on bank performance may come from different levels of FinTech among banks. Advanced FinTech can completely mitigate the negative effects of green credit on bank performance. Conversely, less developed FinTech fails to fully offset these adverse effects. In summary, FinTech enhances the positive impact of green credit on bank performance. To amplify these positive effects, banks should focus on developing their FinTech.

Regarding control variables, the coefficients for IGAR are mostly significantly positive, indicating that traditional interest-earning activities remain the primary contributors to banking performance. A higher proportion of interest-earning assets enables a bank to receive more interest income on the same scale, leading to better performance. The coefficients for CAP are mostly insignificant, except for a significantly negative coefficient in the regression of ROE. This suggests that the bank’s capital structure has little impact on overall performance, but more capital tends to dilute the Return on Equity.

4.2. Mechanism Identification

To explore how FinTech influences the impact of green credit on bank performance, this section examines three channels: cost reduction, reputation enhancement, and risk mitigation.

4.2.1. Test of Cost Reduction Mechanism

To examine the cost reduction mechanism, we directly analyze whether FinTech reduces the marginal cost of green credit. We regress cost indicators on FinTech, green credit, and their interaction terms. Columns (1) and (2) in

Table 6. The impact of FinTech on reducing marginal costs of green credit.

	(1)	(2)
	CIR	AdminCost
lnGC × FTNews	−0.176 **	−0.176 **
	(−2.14)	(−2.44)
lnGC	−0.402	−0.241
	(−1.12)	(−0.76)
FTNews	0.840	0.601
	(1.64)	(1.34)
Control	Yes	Yes
Ind Effect	Yes	Yes
Year Effect	Yes	Yes
Obs	529	526
R2	0.498	0.569
Adj-R2	0.323	0.418

Notes: T-Statistics are reported in parentheses. *, **, and *** represent significance levels of 10%, 5%, and 1%, respectively.

show results using the cost-to-income ratio and management expense ratio as dependent variables, respectively.

The results show that the coefficients of the interaction term between green credit and FinTech are negative and significant at the 5% level. This suggests that as the level of FinTech increases, the marginal cost of green credit decreases, validating Hypothesis H3. In addition, the coefficients of the interaction terms in columns (1) and (2) are nearly identical. This is probably because the operating costs of banks are primarily management expenses, resulting in a high correlation between these two cost indicators.

4.2.2. Test of Reputation Enhancement Mechanism

In this section, we divide the samples into two groups based on the ratio of net income from fees and commissions to examine the reputation enhancement mechanism. If FinTech amplifies the positive effect of green credit on bank performance through reputation enhancement, this amplification should be larger in the group with a higher ratio of net income from fees and commissions. This is because reputation enhancement typically brings more non-interest-related income, which is a significant component of overall bank performance [89].

The results of the grouped regression are shown in

Table 7. The amplifying effect of FinTech in different groups based on FeeIncome.

	(1)	(2)	(3)	(4)	(5)	(6)
	ROA	ROA	ROE	ROE	NPM	NPM
	Low Fee	High Fee	Low Fee	High Fee	Low Fee	High Fee
lnGC × FTNews	0.009	0.015 ***	0.064	0.236 ***	0.462	0.717 ***
	(1.13)	(4.63)	(0.50)	(4.46)	(1.28)	(5.07)
lnGC	0.012	0.021	0.221	0.352	0.076	1.013 *
	(0.53)	(1.56)	(0.66)	(1.63)	(0.08)	(1.76)
FTNews	−0.003	−0.023	0.195	−0.169	0.313	−0.589
	(−0.13)	(−1.13)	(0.48)	(−0.52)	(0.27)	(−0.67)
Control	Yes	Yes	Yes	Yes	Yes	Yes
Ind Effect	Yes	Yes	Yes	Yes	Yes	Yes
Year Effect	Yes	Yes	Yes	Yes	Yes	Yes
Obs	264	265	264	265	264	265
R2	0.462	0.787	0.611	0.895	0.290	0.517
Adj-R2	0.030	0.702	0.299	0.853	−0.279	0.325

Notes: FeeIncome refers to the ratio of net fee and commission income to total operating income. T-Statistics are reported in parentheses. *, **, and *** represent significance levels of 10%, 5%, and 1%, respectively.

. Columns (1), (3), and (5) display the regression results for the group with a lower ratio of net income from fees and commissions, while columns (2), (4), and (6) show the results for the higher ratio group. The coefficients of the interaction term between FinTech and green credit are significantly positive in the higher ratio group but not significant in the lower ratio group. This implies that FinTech more effectively enhances the impact of green credit on bank performance in banks that rely more on non-interest business. This finding is consistent with the predictions of the reputation enhancement mechanism. Thus, Hypothesis H4 is validated.

4.2.3. Test of Risk Mitigation Mechanism

We divided the samples into two groups based on the magnitude of risk indicators for grouped regression to examine the risk mitigation mechanism. This mechanism suggests that the development of FinTech in banks helps reduce the risks associated with green credit business, thereby improving asset quality and bank performance. For banks with higher risks, FinTech’s effect on reducing green credit risk is more pronounced. In this scenario, the amplifying effect of FinTech on the positive impact of green credit on bank performance is greater. Therefore, we anticipate that in the sample of banks with higher risk levels, the coefficient of the interaction term between FinTech and green credit will be larger.

Table 8. The amplifying effect of FinTech in different groups based on risk indicators.

	(1)	(2)	(3)	(4)	(5)	(6)
	ROA	ROA	ROE	ROE	NPM	NPM
	Low NPLR	High NPLR	Low NPLR	High NPLR	Low NPLR	High NPLR
lnGC × FTNews	0.009	0.015 ***	0.064	0.236 ***	0.462	0.717 ***
	(1.13)	(4.63)	(0.50)	(4.46)	(1.28)	(5.07)
lnGC	0.012	0.021	0.221	0.352	0.076	1.013 *
	(0.53)	(1.56)	(0.66)	(1.63)	(0.08)	(1.76)
FTNews	−0.003	−0.023	0.195	−0.169	0.313	−0.589
	(−0.13)	(−1.13)	(0.48)	(−0.52)	(0.27)	(−0.67)
	ROA	ROA	ROE	ROE	NPM	NPM
	Low ODLR	High ODLR	Low ODLR	High ODLR	Low ODLR	High ODLR
lnGC × FTNews	−0.001	0.014 **	−0.046	0.229 ***	−0.037	0.718 ***
	(−0.37)	(2.37)	(−0.83)	(2.75)	(−0.30)	(2.70)
lnGC	0.012	0.025	0.118	0.467 *	0.291	1.364
	(0.84)	(1.32)	(0.47)	(1.74)	(0.51)	(1.59)
FTNews	0.024	−0.005	0.651 *	0.152	0.193	0.484
	(1.23)	(−0.18)	(1.94)	(0.39)	(0.25)	(0.39)
	ROA	ROA	ROE	ROE	NPM	NPM
	Low RWAR	High RWAR	Low RWAR	High RWAR	Low RWAR	High RWAR
lnGC × FTNews	−0.007 **	0.041 ***	−0.117 **	0.449 ***	−0.201	1.558 ***
	(−2.18)	(4.97)	(−2.09)	(3.97)	(−1.44)	(4.32)
lnGC	0.040 ***	−0.022	0.637 ***	0.043	1.668 ***	−0.548
	(2.99)	(−0.93)	(2.86)	(0.13)	(3.01)	(−0.53)
FTNews	0.032	0.000	0.794 **	−0.094	1.601 **	−0.266
	(1.65)	(0.01)	(2.45)	(−0.21)	(1.99)	(−0.19)

Notes: T-Statistics are reported in parentheses. *, **, and *** represent significance levels of 10%, 5%, and 1%, respectively. All regressions include control variables, individual fixed effects, and time fixed effects.

presents the regression results of the test on the risk mitigation mechanism. Regardless of the risk indicators used, the results consistently show that within the higher-risk group, the amplifying effect of FinTech is stronger. This confirms the existence of the risk mitigation channel and verifies Hypothesis H5.

5. Heterogeneity Analysis

Different bank characteristics and macroeconomic environments can lead to significant differences in their operational behaviors. These differences affect how FinTech enhances the impact of green credit on bank performance. In this section, we discuss the heterogeneity analysis of bank types, green credit development, and the prosperity of the banking industry.

5.1. Bank Types

In China, there are many types of banks, each exhibiting significant differences in scale, customer base, strategic positioning, business capabilities, and risk management. Consequently, it is essential to analyze the heterogeneous effects of FinTech across different types of banks to gain a nuanced understanding of its impact. In this section, we conducted regression analyses on samples from national commercial banks, city commercial banks, and rural commercial banks. For clarity, we grouped state-owned and joint-stock commercial banks under the category of national commercial banks. Due to the small sample size and distinct characteristics of private banks, they were excluded from this analysis.

Table 9. Heterogeneity analysis of bank types.

	(1)	(2)	(3)
	ROA	ROA	ROA
	National Banks	City Banks	Rural Banks
lnGC × FTNews	−0.004	0.010 **	−0.006
	(−0.53)	(1.99)	(−0.19)
lnGC	0.031	0.022	0.084
	(1.47)	(1.38)	(0.66)
FTNews	0.067 *	0.011	−0.033
	(1.70)	(0.53)	(−0.53)
	ROE	ROE	ROE
	National Banks	City Banks	Rural Banks
lnGC × FTNews	−0.111	0.134 *	0.155
	(−0.95)	(1.92)	(0.40)
lnGC	0.358	0.527 **	1.002
	(0.97)	(2.39)	(0.58)
FTNews	1.808 ***	0.096	0.413
	(2.65)	(0.32)	(0.49)
	NPM	NPM	NPM
	National Banks	City Banks	Rural Banks
lnGC × FTNews	0.006	0.457 **	−0.327
	(0.02)	(2.12)	(−0.33)
lnGC	1.151	1.009	3.281
	(1.38)	(1.48)	(0.77)
FTNews	3.086 **	0.522	−2.060
	(2.00)	(0.57)	(−0.99)
Control	Yes	Yes	Yes
Ind Effect	Yes	Yes	Yes
Year Effect	Yes	Yes	Yes
Obs	142	336	51

Notes: T-Statistics are reported in parentheses. *, **, and *** represent significance levels of 10%, 5%, and 1%, respectively.

presents the heterogeneity results for different bank types. Unlike previous tables, each panel here represents different bank performance indicators, while the columns represent the different types of banks. The results reveal that the interaction between FinTech and green credit is not significant for national and rural commercial banks. However, for city commercial banks, the interaction term between FinTech and green credit shows a significantly positive coefficient.

National commercial banks are the most developed entities within China’s banking system. These banks have made substantial investments in FinTech and have achieved a high level of technological integration. As a result, FinTech is likely already deeply embedded in their green credit businesses, which means that its marginal contribution is relatively limited. This saturation effect explains the insignificant interaction coefficient between FinTech and green credit in these banks.

City commercial banks, on the other hand, occupy a middle ground between national and rural commercial banks in terms of scale and competitiveness. They primarily serve local customers who demand high-quality financial services. In recent years, these banks have made significant progress in both green credit and FinTech. By integrating FinTech into their green credit businesses, they have effectively reduced costs and risks, built strong green reputations, and enhanced their overall performance. Unlike national commercial banks, city commercial banks have not yet reached a stage where the marginal benefits of FinTech diminish, which explains the significantly positive interaction coefficient observed.

Rural commercial banks lag behind other types of banks in FinTech and green credit development, and the integration of FinTech and green credit is insufficient. Their primary clientele consists of individuals and enterprises in rural areas, where the demand for green credit is relatively low. Given the limited volume of business, adopting high-cost FinTech solutions is not financially justified for these banks. Traditional credit approval and management methods remain more profitable. As a result, FinTech struggles to achieve economies of scale in rural commercial banks, leading to an insignificant interaction effect between FinTech and green credit.

In summary, the impact of FinTech on green credit businesses has reached a saturation point for national commercial banks, while city commercial banks continue to experience positive effects. However, rural commercial banks have not yet seen a significant impact. In the long term, as green credit businesses evolve, the demand for FinTech is expected to grow. This will help address the saturation of FinTech efficiency in national commercial banks, and rural commercial banks are likely to adopt FinTech more extensively. Consequently, the positive impact of FinTech on green credit businesses is expected to become significant across all types of banks.

5.2. Green Credit Development

We are also interested in whether FinTech’s impact on enhancing green credit’s influence on bank performance varies at different levels of green credit development. We selected two indicators to assess green credit development levels: the size of the bank’s green credit (lnGC) and the green credit index (GCI) of the bank’s region. The GCI represents the ratio of total environmental project credit to total credit in the province where the bank is located. The first indicator measures the bank’s green credit development, while the second assesses regional green credit development. Examining these two dimensions provides a more comprehensive perspective on the question we proposed. Next, we conduct grouped regressions based on these two indicators separately.

Table 10. Heterogeneity analysis of green credit development.

	(1)	(2)	(3)	(4)	(5)	(6)
	ROA	ROA	ROE	ROE	NPM	NPM
	Low GC	High GC	Low GC	High GC	Low GC	High GC
lnGC × FTNews	0.004	0.013 **	0.066	0.198 **	0.084	0.936 ***
	(0.37)	(2.55)	(0.48)	(2.40)	(0.22)	(3.83)
lnGC	0.022	0.064 ***	0.585 **	0.880 **	0.815	2.289 **
	(1.13)	(2.78)	(2.15)	(2.41)	(1.05)	(2.11)
FTNews	0.019	−0.040 *	0.242	−0.096	0.586	−1.877 *
	(0.65)	(−1.79)	(0.59)	(−0.27)	(0.50)	(−1.78)
	ROA	ROA	ROE	ROE	NPM	NPM
	Low GDI	High GDI	Low GDI	High GDI	Low GDI	High GDI
lnGC × FTNews	0.004	0.029 ***	0.034	0.334 **	0.180	1.277 ***
	(0.83)	(3.03)	(0.51)	(2.55)	(1.15)	(2.68)
lnGC	0.006	0.018	0.389	0.137	0.152	0.853
	(0.31)	(0.66)	(1.38)	(0.36)	(0.23)	(0.62)
FTNews	−0.029	0.072 **	−0.289	1.046 **	−0.853	3.253 *
	(−1.02)	(2.07)	(−0.68)	(2.17)	(−0.85)	(1.86)

Notes: T-Statistics are reported in parentheses. *, **, and *** represent significance levels of 10%, 5%, and 1%, respectively. All regressions include control variables, individual fixed effects, and time fixed effects.

presents the grouped regression results based on the development level of green credit. The results indicate that, whether from the perspective of the bank itself or its region, a higher level of green credit development increases FinTech’s impact on enhancing green credit’s influence on bank performance. For banks, FinTech exhibits a “scale effect” in the green credit business. As the scale of green credit expands, the average technological cost per unit of green loan decreases, thus improving the bank’s performance. For regions, a higher level of green credit development indicates a larger green credit market, greater policy support and encouragement, and a more mature green financial ecosystem. In such markets, the numerous green projects more significantly require FinTech to empower green credit businesses. Therefore, FinTech plays a more important role in enhancing the impact of green credit on bank performance.

In conclusion, from both the banks’ and the regional perspectives on green credit development, there is also a synergistic effect between FinTech and green credit. To maximize the benefits of green credit, banks and the local markets must collaborate to create a favorable green financial environment.

5.3. The Prosperity of the Banking Industry

The prosperity of the banking industry reflects its overall operational status, covering aspects such as asset quality, profitability, and risk management. Examining the heterogeneous effects of banking prosperity can deepen our understanding of FinTech’s impact on improving green credit performance across different banking cycles. This insight can guide banks on when to enhance their FinTech investments for greater benefits. We grouped the data based on the Banking Prosperity Index (BPI) into downturn and upturn states. The results of the grouped regression are shown in

Table 11. Heterogeneity analysis of Banking Industry Prosperity Index.

	(1)	(2)	(3)	(4)	(5)	(6)
	ROA	ROA	ROE	ROE	NPM	NPM
	Downturn	Upturn	Downturn	Upturn	Downturn	Upturn
lnGC × FTNews	0.030 ***	0.011 ***	0.347 ***	0.152 ***	1.369 ***	0.558 ***
	(4.72)	(3.00)	(3.80)	(2.69)	(4.76)	(3.53)
lnGC	−0.046 **	0.031 *	−0.499	0.533 **	−1.827 *	1.059
	(−2.14)	(1.81)	(−1.63)	(2.08)	(−1.89)	(1.47)
FTNews	−0.073 **	0.038	−0.764	0.635 *	−3.000 **	1.501
	(−2.25)	(1.57)	(−1.66)	(1.74)	(−2.06)	(1.47)
Control	Yes	Yes	Yes	Yes	Yes	Yes
Ind Effect	Yes	Yes	Yes	Yes	Yes	Yes
Year Effect	Yes	Yes	Yes	Yes	Yes	Yes
Obs	248	281	248	281	248	281
R2	0.666	0.722	0.798	0.869	0.496	0.397
Adj-R2	0.356	0.499	0.610	0.764	0.027	−0.090

Notes: T-Statistics are reported in parentheses. *, **, and *** represent significance levels of 10%, 5%, and 1%, respectively.

. The results indicate that whether the banking industry is in an upturn or downturn, the coefficient of the interaction term between FinTech and green credit is significantly positive. Notably, this coefficient is higher during a downturn, suggesting that FinTech more effectively enhances the impact of green credit on bank performance during downturns.

The results can be interpreted from a risk aversion perspective. During prosperous periods in the banking industry, the market maintains a positive outlook, and participants’ risk tolerance increases. With abundant capital, banks may take on higher risks, reducing the demand for FinTech’s risk management and prudent assessment in green credit businesses.

However, during a downturn, market participants’ risk awareness rises significantly. Banks then emphasize risk management and prefer lower-risk green projects. At this time, FinTech becomes crucial. It helps banks more effectively evaluate and manage green credit risks, improve asset allocation efficiency, and reduce potential losses. Therefore, during banking industry downturns, FinTech can more significantly enhance the positive impact of green credit on bank performance.

6. Robustness Tests

6.1. Another Green Credit Measure

In the previous analysis, we used the balance of green credit (lnGC) indicator. In this section, we use the green credit ratio (GCR) to further verify the conclusions. The regression results are shown in

Table 12. Robustness check: using green credit ratio.

	(1)	(2)	(3)	(4)	(5)	(6)
	ROA	ROE	NPM	ROA	ROE	NPM
GCR	0.014 ***	0.218 ***	0.742 ***	0.003	0.073	0.123
	(4.30)	(4.53)	(5.25)	(0.49)	(0.92)	(0.52)
GCR × FTNews				0.005 ***	0.070 **	0.306 ***
				(2.80)	(2.47)	(3.67)
FTNews				−0.023	−0.187	−1.215
				(−1.33)	(−0.74)	(−1.62)
Control	Yes	Yes	Yes	Yes	Yes	Yes
Ind Effect	Yes	Yes	Yes	Yes	Yes	Yes
Year Effect	Yes	Yes	Yes	Yes	Yes	Yes
Obs	568	568	568	529	529	529
R2	0.629	0.810	0.350	0.645	0.825	0.373
Adj-R2	0.496	0.742	0.116	0.520	0.764	0.153

Notes: T-Statistics are reported in parentheses. *, **, and *** represent significance levels of 10%, 5%, and 1%, respectively.

, structured similarly to Table 5. In columns (1) to (3), the coefficient of the green credit ratio is significantly positive. In columns (4) to (6), the coefficient of the interaction term between the green credit ratio and FinTech is also significantly positive. These results validate Hypotheses H1 and H2, demonstrating that green credit enhances bank performance and that FinTech amplifies this effect. For every 1 percentage point increase in the green credit ratio, ROA increases by 0.014%, ROE by 0.218%, and NPM by 0.742%. Therefore, the green credit ratio’s impact on bank performance is economically significant.

6.2. Other FinTech Measures

Different FinTech indicators often show significant variations due to their calculation methods. In this section, we use the Peking University China Commercial Bank Digital Transformation Index (FTDigital) and a FinTech indicator based on patent applications (FTPatent).

Table 13. Robustness check: other FinTech measures.

	(1)	(2)	(3)	(4)	(5)	(6)
	ROA	ROE	NPM	ROA	ROE	NPM
lnGC	0.023 *	0.451 ***	0.879 *	0.025 **	0.451 ***	1.099 **
	(1.96)	(2.64)	(1.78)	(2.44)	(2.98)	(2.50)
lnGC × FTDigital	0.000 ***	0.004 ***	0.019 ***
	(4.05)	(3.24)	(5.12)
FTDigital	0.001 ***	0.008	0.046 ***
	(2.66)	(1.42)	(2.94)
lnGC × FTPatent				0.006 **	0.094 **	0.375 ***
				(2.12)	(2.29)	(3.17)
FTPatent				0.010	0.073	0.208
				(0.66)	(0.35)	(0.34)
Control	Yes	Yes	Yes	Yes	Yes	Yes
Ind Effect	Yes	Yes	Yes	Yes	Yes	Yes
Year Effect	Yes	Yes	Yes	Yes	Yes	Yes
Obs	468	468	468	569	569	569
R2	0.669	0.826	0.429	0.644	0.818	0.389
Adj-R2	0.528	0.751	0.184	0.514	0.751	0.165

Notes: T-Statistics are reported in parentheses. *, **, and *** represent significance levels of 10%, 5%, and 1%, respectively. All regressions include control variables, individual fixed effects, and time fixed effects.

presents the regression results. Columns (1) to (3) show the results with FTDigital as the FinTech variable, while columns (4) to (6) display the results with FTPatent. The results show that the coefficient of the interaction term between green credit and FinTech remains significantly positive, demonstrating the robustness of Hypothesis H2.

6.3. Lagged Explanatory Variables

We applied a one-period lag to all explanatory variables as a robustness check. This ensures that the current dependent variable is not directly influenced by explanatory variables from the same period, helping to mitigate potential endogeneity problems and enhancing the model’s credibility.

Table 14. Robustness check: lagged explanatory variables.

	(1)	(2)	(3)	(4)	(5)	(6)
	ROA	ROE	NPM	ROA	ROE	NPM
L.lnGC	0.024 **	0.623 ***	1.685 ***	0.013	0.516 ***	0.918 *
	(2.03)	(3.67)	(3.50)	(1.07)	(2.78)	(1.82)
L.lnGC × L.FTNews				0.011 ***	0.071 *	0.684 ***
				(3.74)	(1.66)	(5.86)
L.FTNews				0.011	0.108	0.346
				(0.60)	(0.40)	(0.47)
Control	Yes	Yes	Yes	Yes	Yes	Yes
Ind Effect	Yes	Yes	Yes	Yes	Yes	Yes
Year Effect	Yes	Yes	Yes	Yes	Yes	Yes
Obs	489	490	489	457	458	457
R2	0.582	0.793	0.323	0.611	0.807	0.386
Adj-R2	0.405	0.706	0.037	0.451	0.728	0.134

Notes: T-Statistics are reported in parentheses. *, **, and *** represent significance levels of 10%, 5%, and 1%, respectively.

presents the regression results after lagging all explanatory variables. The results indicate that the positive impact of green credit on bank performance, as well as the significant effect of FinTech in enhancing the influence of green credit on bank performance, remain consistent with previous findings.

6.4. DID Approach

In this section, we employed a difference-in-differences (DID) method to re-examine the problem. In 2016, the former China Banking Regulatory Commission issued the “Regulatory Guidance on the Development Plan for Information Technology in the Banking Industry during the 13th Five-Year Plan.” This document aimed to strengthen cybersecurity and information security capabilities within the banking sector. It also focused on using technological innovation to improve financial services and support economic development. This policy can be seen as a shock that increased FinTech levels in the banking industry.

We believe that banks with higher levels of FinTech were more significantly influenced by this policy. We identified banks with FinTech indicators (FTNews) in 2015 above the median of that year’s sample as the experimental group, while the remaining banks were set as the control group. Subsequently, we constructed the following DID regression model:

$$\begin{gathered} {\hspace{1em}\hspace{1em}\hspace{0.17em}Performance}_{it}={\beta }_0+{\beta }_1{GC}_{it}\times {Treat}_i\times {Post}_t+{\beta }_2{GC}_{it}\times {Treat}_i \\ \hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}+{\beta }_3{GC}_{it}\times {Post}_t+{\beta }_4{Treat}_i\times {Post}_t+{\beta }_5{GC}_{it} \\ \hspace{0.17em}+{\beta }_6{Controls}_{it}+{\mu }_i+{\rho }_t+{\epsilon }_{it} \end{gathered}$$

(4)

In Equation (4), ${Treat}_i$ represents the experimental group dummy variable, where ${Treat}_i=1$ if the bank is in the experimental group, and ${Treat}_i=0$ otherwise. ${Post}_t$ denotes the policy dummy variable, where ${Post}_t=1$ for the years 2016 and later and ${Post}_t=0$ otherwise.

It is worth mentioning that our DID model differs from typical ones. We incorporated an interaction term between three variables: green credit, the experimental group dummy variable, and the policy dummy variable. The coefficient of this interaction term indicates the extent to which FinTech enhances the impact of green credit on bank performance. To ensure the model’s completeness, we also included interaction terms for the pairwise combinations of these three variables.

The validity of the DID model depends on satisfying the parallel trend assumption. To test this, we replace ${\beta }_1{GC}_{it}\times {Treat}_i\times {Post}_t$ in Equation (4) with ${\sum }_k{\beta }_k{GC}_{it}\times {Treat}_i\times D_k$, where $D_k$ denotes the year dummy variable. This allows us to compare the impact of green credit on bank performance before and after policy implementation. We set 2009 as the baseline year and 2010–2022 as the testing period. Figure 5 illustrates the results of the parallel trend test. The horizontal axis represents the distance from the policy implementation year, 2016, and the vertical axis denotes the marginal effect of green credit on bank performance. The results show no significant change in the marginal effect of green credit on bank performance before policy implementation. However, after policy implementation, the marginal effect of green credit on bank performance significantly increases, validating the parallel trend assumption.

Finally,

Table 15. Robustness check: DID approach.

	(1)	(2)	(3)
	ROA	ROE	NPM
lnGC × Treat × Post	0.100 ***	1.377 ***	3.989 ***
	(3.43)	(3.23)	(3.12)
Control	Yes	Yes	Yes
Ind Effect	Yes	Yes	Yes
Year Effect	Yes	Yes	Yes
Obs	524	524	524
R2	0.666	0.832	0.379
Adj-R2	0.547	0.773	0.158

Notes: T-Statistics are reported in parentheses. *, **, and *** represent significance levels of 10%, 5%, and 1%, respectively.

presents the results of the difference-in-differences estimation. The coefficient of the interaction term between green credit, the experimental group dummy variable, and the policy dummy variable is significantly positive. This indicates that the implementation of FinTech policies has enhanced the positive impact of green credit on bank performance, validating Hypothesis H2 from the perspective of policy shock.

7. Conclusions

Green credit, as a financial product supporting environmental protection and sustainable development projects, has a complex impact on bank performance. The existing literature presents diverse viewpoints on this impact. In recent years, FinTech has gradually spread into various banking domains, demonstrating its unique value. With the ongoing integration of FinTech and green credit, this paper explores whether FinTech influences the impact of green credit on bank performance. Theoretically, we try to explain the diverse viewpoints in the existing literature regarding the effect of green credit on bank performance. Practically, we study the extent to which FinTech can enhance the sustainability of banks’ green financial businesses. This paper provides theoretical foundations and practical guidance for banks to apply FinTech in their green finance businesses.

Based on an empirical analysis of data from Chinese commercial banks, this paper finds that green credit generally enhances bank performance, and FinTech further amplifies this positive effect. Marginal effect analysis indicates that when a bank’s FinTech level is low, the impact of green credit on bank performance is negative but not significant. However, when the FinTech level is high, the impact is significantly positive. This suggests that FinTech can explain the conflicting views in the existing literature, as differences in the level of bank FinTech lead to different conclusions about the impact of green credit on bank performance. This study offers a new perspective for understanding the relationship between FinTech and green credit, enriching relevant research. While the existing literature points out that FinTech can influence the level of green credit, this paper proposes that FinTech can synergize with green credit to affect bank performance.

We propose three mechanisms through which FinTech influences the impact of green credit on bank performance. First, FinTech reduces costs associated with green credit business, such as approval, maintenance, and external agency expenses. Second, FinTech enhances the reputation of green credit by improving customer experience and loyalty through precise marketing and personalized services. Third, FinTech mitigates the risks of green credit by improving the accuracy of risk assessments and strengthening dynamic information monitoring and early warning systems. Empirical results validate these mechanisms.

Furthermore, heterogeneity analysis reveals that the strengthening effect of FinTech on the impact of green credit on bank performance is more pronounced in city commercial banks. Additionally, as green credit development increases and during periods of banking industry downturns, FinTech demonstrates greater advantages, making the impact of green credit on bank performance more apparent. Finally, this paper conducts various robustness tests, including replacing green credit and FinTech indicators, using lagged explanatory variables, and employing the DID method. All these tests support the above conclusions.

Based on this study, we propose the following policy recommendations:

First, the regulator should continue to support the development of green credit in banks. This can be achieved by collaborating with relevant institutions to establish scientific and objective standards for evaluating green credit, enabling banks to better assess the risks and returns of green projects. Additionally, the regulator should revise policy documents to provide clear guidelines for banks, including specific details on operational processes, approval standards, and post-loan management. These measures will ensure the compliance and effectiveness of green credit activities. Furthermore, the regulator can organize green finance training to help banks improve their management capabilities, particularly in project evaluation and risk control.

Second, the government should support the development of banking FinTech through various policy measures. It can encourage banks to invest more in FinTech and acquire advanced equipment and technologies by offering tax deductions. Additionally, the government can provide financial support for FinTech infrastructure, such as big data processing centers, cloud computing service centers, and blockchain technology platforms. These infrastructures will ensure that financial institutions receive the necessary technical support, thereby enhancing their service efficiency.

At last, banks should enhance the development of FinTech and actively integrate it into their green credit business. They need to establish dedicated FinTech departments to drive innovation across various business areas. Rural commercial banks should increase investment in FinTech to achieve synergy with green credit services. State-owned and joint-stock commercial banks should maintain their current pace of FinTech development and explore new profit opportunities in green finance. Additionally, banks can accelerate FinTech advancement by collaborating with tech companies and research institutions, utilizing their expertise and resources. The widespread application of FinTech can enhance the economic and environmental benefits of banks, significantly contributing to sustainable finance.

Future research could expand on this study by exploring the impact of specific FinTech components on green credit activities. With the wide variety of FinTech technologies available, analyzing how individual elements like blockchain, artificial intelligence, and mobile payments enhance the effectiveness of green credit could offer deeper insights into FinTech adoption in banking. Currently, the development of green credit businesses is still in its early stages, and detailed data, such as revenues, costs, and risks associated with each bank’s green credit activities, remain limited. As more comprehensive data become available in the future, more in-depth research on the impact of green credit on bank performance will be possible. While the core principles of green credit and FinTech are generally consistent across countries, the findings of this study may provide valuable guidance for their development in other nations. However, due to variations in economic development, financial regulation, and cultural factors, the effects of green credit and FinTech may differ between countries, requiring further research to explore the influence of these country-specific factors.