FinTech and Inclusive Green Growth: A Causal Inference Based on Double Machine Learning

Abstract

Based on city-level data from 287 Chinese prefecture-level administrative units spanning 2011 to 2021, this study employs text analysis to quantify FinTech development and applies a double machine learning model to empirically assess the impact of FinTech on Inclusive Green Growth, along with its underlying mechanisms. The key findings are as follows: (1) FinTech significantly promotes Inclusive Green Growth, particularly in the areas of payment systems, lending and capital raising, and investment management. This conclusion holds across a range of robustness checks, including alternative measurement indicators, different machine learning models, and tests for endogeneity. (2) Mechanism analysis demonstrates that FinTech drives Inclusive Green Growth by fostering financial employment, expanding financial supply, and facilitating green technological innovation. (3) Heterogeneity analysis reveals that while China still faces elements of the resource curse, the positive impact of FinTech on Inclusive Green Growth is more pronounced in regions with higher levels of digital infrastructure, environmental regulation, and green finance. These findings provide valuable insights into leveraging FinTech’s potential to support China’s high-quality economic development.

1. Introduction

Since the reform and opening-up, China has experienced rapid economic growth. According to data from the National Bureau of Statistics, China’s gross domestic product (GDP) achieved an average annual growth rate of 9.1% from 1978 to 2022. Nevertheless, despite this rapid expansion, China still faces significant challenges, including economic structural imbalances, social development disparities, and ecological degradation [1] (pp. 1–13). The report from the 20th National Congress of the Communist Party of China stressed the need to coordinate industrial restructuring, pollution control, ecological protection, and climate change mitigation while promoting efforts to reduce carbon emissions, cut pollution, expand green spaces, and foster growth. It further calls for raising the incomes of low-income earners and expanding the middle-income group. These guidelines seek to balance economic development with environmental protection and social equity. Addressing the lack of greening in traditional growth models and the insufficient inclusiveness in development—ultimately achieving Inclusive Green Growth—has thus become an essential direction for China’s future economic development.

The concept of Inclusive Green Growth originates from the ideas of green growth and inclusive growth, emphasizing sustainable development through enhanced economic productivity, job creation, poverty and inequality reduction, and the preservation of natural resources and ecosystems. Promoting Inclusive Green Growth has become a key area of widespread interest. Existing research has mainly focused on measuring Inclusive Green Growth and analyzing its spatial-temporal dynamics [2] (pp. 11025–11045), while also examining the influence of international macroeconomic factors such as global economic policy uncertainty [3] (p. 332) and economic globalization [4] (pp. 452–482). Additionally, studies have investigated the effects of sector-specific factors, including types of innovation [5] (p. 406) and network infrastructure [6] (pp. 113–135), on Inclusive Green Growth. As digital transformation advances, the application of big data technologies increasingly drives high-quality, sustainable development in China’s economy. Some scholars, adopting a micro-level perspective, propose that digital transformation within enterprises accelerates capital flows and labor mobility, thus indirectly promoting Inclusive Green Growth [7] (pp. 22–30).

FinTech, as a financial innovation driven by big data technologies, embodies the deep integration of emerging, disruptive technologies—including big data, cloud computing, artificial intelligence, blockchain, and mobile internet—with traditional financial services. Its emergence provides new avenues for fostering high-quality economic development. The FinTech Development Plan (2022–2025) explicitly outlines four core principles for FinTech development: digital empowerment, intelligence for the public, green and low-carbon growth, and equitable inclusiveness. Numerous studies have demonstrated that FinTech, leveraging its technological advantages, positively impacts economic growth, addressing the vanishing effect seen with traditional finance in economic expansion [8] (pp. 52–61). Moreover, FinTech possesses notable green attributes, fostering urban green technology innovation, reducing emissions, and improving alignment among technological innovation, financial development, and environmental governance. These advancements contribute to significant environmental governance effects [9] (pp. 116–130). Consequently, FinTech’s potential role as a key driver for Inclusive Green Growth warrants careful examination. Given the dual objectives of promoting inclusive economic growth and environmental sustainability, a critical question arises: Can FinTech foster Inclusive Green Growth, and through what mechanisms? This question deserves rigorous investigation. From a practical perspective, the establishment of the FinTech Committee by the People’s Bank of China in 2017 has clarified the policy direction toward financial innovation and risk prevention, supported by continuous top-level planning. Following the release of the FinTech Development Plan (2019–2021) in August 2019, the second iteration, the FinTech Development Plan (2022–2025), was issued in January 2022. Concurrently, various institutions—including financial entities, internet technology companies, and IT service providers—have actively pursued patents in areas such as big data, blockchain, cloud computing, and artificial intelligence. According to data from PatSnap (https://www.zhihuiya.com, accessed on 25 December 2022), over the past five years (2018–2022), China has ranked first globally in FinTech-related patents, with a total of 106,900 filings, establishing it as the primary global source of FinTech innovation. Clearly, as FinTech innovation flourishes, a supportive policy environment is gradually forming, the pace of industry digital transformation is accelerating, and FinTech is becoming a pivotal force driving high-quality economic development.

In this context, this paper focuses on 287 Chinese prefecture-level administrative units, employing a double machine learning (DML) model to examine the effects of FinTech on Inclusive Green Growth and investigate the heterogeneity and mechanisms underlying its influence. This study’s potential marginal contributions are as follows: (1) While prior research has mainly focused on the impact of FinTech on economic and green growth, it has largely overlooked its role in promoting social inclusiveness. Achieving high-quality economic growth necessitates not only efficiency improvements but also a balance between green economic development and social equity. This paper integrates the concepts of inclusiveness and greenness to explore FinTech’s impact on Inclusive Green Growth, thus broadening the research perspective within the FinTech literature and addressing a gap in the studies on advancing high-quality economic and social development. (2) This paper further enriches the understanding of the mechanisms through which FinTech affects urban Inclusive Green Growth, including its roles in promoting financial employment, expanding financial supply, and empowering green technological innovation. Additionally, it identifies heterogeneous effects across regions with varying resource endowments, digital infrastructure, environmental regulations, and green finance development, thereby providing a comprehensive complement to the existing literature [5,10] (p. 406, pp. 67–80). These findings offer valuable insights for policymakers aiming to develop tailored policies. (3) Previous studies on the impact factors and economic outcomes of FinTech or Inclusive Green Growth have primarily relied on traditional causal inference methods, such as multiple linear regression models, difference-in-differences, matching methods, regression discontinuity design, and synthetic control methods. However, these traditional approaches are often constrained by limitations, including hidden bias, stringent assumptions, endogeneity, and extrapolation issues, which limit their applicability in social science research. The double machine learning (DML) model offers several advantages over traditional methods, including the ability to handle complex nonlinear relationships, control potential biases, and provide adaptive and non-parametric features. By adopting a research perspective on FinTech and Inclusive Green Growth, this paper applies the DML model for empirical testing, aiming to advance the application of machine learning and other emerging research methodologies within the humanities and social sciences.

2. Literature Review

As the integration between advanced technologies and the financial industry continues to surge, the concepts involved are dynamically evolving, leading to the emergence of terms such as Internet-based Finance, TechFin, Digital Finance, and FinTech. However, these concepts have yet to be universally defined, resulting in their interchangeable use and application across various studies based on specific research contexts. The early literature typically described financial services operating through the internet as Internet-based Finance [11] (pp. 1–12). Other scholars have referred to the integration of technology and finance as TechFin, which emphasizes financial formats, services, and products designed to support technological innovation [12] (pp. 13–20). Unlike Digital Finance, which focuses on using accumulated financial data for analysis and decision-making, FinTech highlights technological attributes, specifically emphasizing the application of technological innovations in financial services [13] (pp. 1489–1502). Initially, FinTech was defined as the combination of banking expertise with modern management science and computer technology [14] (pp. 62–63). However, as global technological innovation has entered an unprecedented period of activity, FinTech has continuously been redefined. The Financial Stability Board (FSB) provides one of the most widely accepted definitions, describing FinTech as technology-based innovation in financial services that can generate new business models, applications, processes, or products, thereby having a significant impact on the delivery of financial services.

Current research on FinTech primarily focuses on enriching its theoretical framework, evaluating measurement indicators, and analyzing its influencing factors and economic impacts. At its core, FinTech is a comprehensive process that leverages advanced technologies and innovative approaches to transform and optimize financial services. It involves a wide range of stakeholders, from macro-level regulatory agencies to companies using modern internet technologies to operate financial services, as well as traditional commercial banks. This diversity makes the measurement of FinTech challenging, leading scholars to adopt various indicators to assess the degree of FinTech development at the regional level. Some have measured it through indicators such as innovation vitality and R&D expenditures [15] (pp. 46–53), the number of FinTech companies [16] (pp. 138–155), or the Digital Financial Inclusion Index [17] (pp. 177–193). Other researchers have employed text analysis, using FinTech-related keywords scraped from various databases to measure the level of FinTech development [18] (pp. 81–98)., with the analysis conducted using Python (version 3.x, developed by the Python Software Foundation, Wilmington, DE, USA).

Additionally, the factors driving FinTech development are highly diverse. Haddad and Hornuf suggest that a positive correlation often exists between a region’s level of economic development and the number of FinTech startups [19] (pp. 81–105). In regions with more developed financial systems, there is typically a higher concentration of technical talent and abundant social capital, which further facilitates the growth of FinTech [20] (pp. 1–18). At the same time, moderate financial regulation and proactive policy planning are critical in guiding and supporting the development of FinTech [21] (pp. 1490–1497).

On the other hand, FinTech also holds significant economic value. From a macro perspective, Broby et al. argue that FinTech enhances labor productivity, which positively impacts regional employment [22] (pp. 1–13). Additionally, the innovative financial infrastructure, new business models, and practices emerging from FinTech contribute to improvements in total factor productivity [23] (pp. 134–144). Li and Zhang suggest that FinTech facilitates the dual processes of industrial upgrading and rationalization by leveraging the Engel effect on the demand side and the Baumol effect on the supply side [24] (pp. 102–118). From a micro perspective, addressing the financing challenges faced by small- and medium-sized enterprises (SMEs) and boosting their innovation capabilities are critical for the vitality of the real economy, and this has become a focal point of current research. Numerous studies indicate that FinTech reduces information asymmetry between banks and businesses, alleviates issues related to short-term debt financing [25] (pp. 137–154), and fosters innovation by enhancing the effectiveness of tax refunds.

Additionally, some of the literature investigates the green development implications of FinTech. It has been noted that FinTech, through various online business models, reduces transaction costs and resource waste, improves the quality and efficiency of green financial resource allocation, and thus impacts regional environmental quality. Moreover, green policy pilots have been found to positively moderate the emission reduction effects of FinTech [26] (pp. 3–16). Thus, FinTech plays a crucial role in providing financial services to the real economy and supporting the green transformation of economic and social systems.

It can be observed that existing research has laid a solid foundation for this study. However, most of these works tend to focus on the economic outcomes of FinTech from a singular perspective, such as economic growth or environmental governance, lacking a systematic exploration that integrates both aspects. Investigating FinTech from the perspective of Inclusive Green Growth provides a valuable supplement to the current body of research on the economic consequences of FinTech.

3. Theoretical Analysis

FinTech plays a crucial role in contributing to urban Inclusive Green Growth. On the one hand, FinTech has inclusive value, narrowing income gaps across regions and populations by providing more accessible financial services to a broader range of groups [27] (pp. 141–151). As a phenomenon-driven innovation, FinTech also creates a range of internet-based financial services driven by data and technology, meeting diverse needs in areas such as employment, financing, and investment. This multi-sectoral application is gradually transforming the employment structure of the traditional financial industry, helping to address talent homogeneity and attracting more skilled professionals, thereby creating more income-generating opportunities [28] (pp. 150–157). This demonstrates FinTech’s capacity to promote inclusive economic and social growth. On the other hand, FinTech represents the deep integration of data-driven digital technologies with financial services, supporting technological innovation and enhancing core competitiveness through next-generation internet technologies. By reducing the information barriers and technical constraints associated with traditional resource use, FinTech helps micro-level actors improve their environmental awareness and capabilities. Additionally, by increasing the efficiency of green credit allocation and enhancing the effectiveness of green investments, FinTech fosters environmental investment and green technology research and development, leading to advancements in energy utilization and pollution control technologies [29,30] (pp. 207–224) (pp. 913–919). From this perspective, FinTech contributes to promoting green economic and social growth.

In fact, achieving Inclusive Green Growth through FinTech is a complex process, resulting from the organic integration, allocation, accumulation, and transformation of various production factors such as labor, capital, and technology. The specific theoretical analysis framework is illustrated in Figure 1. First, FinTech promotes Inclusive Green Growth by driving employment in the financial sector. Employment, as the foundation of national welfare and development, is critical for achieving economic growth. The integration of digital technology with finance continuously impacts labor market dynamics. The catfish effect suggests that when FinTech enters the market with its unique technological support, business models, and value-creation methods, traditional financial institutions feel competitive pressure and are compelled to accelerate internal transformations in their strategies, operations, and structures [31] (pp. 56–70). According to the China FinTech and Digital Inclusive Finance Development Report, since the establishment of the first banking FinTech subsidiary, Industrial Bank Digital Finance, in 2015, the number of banking FinTech subsidiaries has steadily increased, with more small and medium-sized banks joining this trend. By November 2022, 19 such subsidiaries had been established in China, demonstrating that the combination of finance and technology has effectively increased the demand for complex labor in the financial industry, thereby contributing to both the expansion and improvement of employment within the sector [32] (pp. 53–62).

Second, FinTech fosters Inclusive Green Growth by expanding financial supply. The rational allocation of financial capital is a critical pathway to achieving shared prosperity. Traditional financial markets often follow the 80/20 rule, where 20% of clients generate 80% of the profits, leading financial institutions to focus on high-end customers while neglecting the remaining 80%, who possess significant growth potential. This focus has hindered the development of inclusive finance. FinTech lowers the cost of acquiring information for banks, enhances risk assessment and control, and promotes the expansion of the financial sector’s scale, structure, and efficiency. It also encourages the accumulation of financial capital and strengthens lending willingness. Furthermore, FinTech has transformed the inclusive consumer market by offering products and services characterized by low costs, high efficiency, transparency, and personalized asset allocation. These features reduce the barriers to accessing loans for low-income groups and underserved regions, allowing more long-tail customers, such as small and micro-enterprises and low-income individuals, to gain access to financial resources. For example, Ping An Bank, leveraging AI and big data, implemented the Smart Bank 3.0 strategy in 2021, which introduced new models of retail business transformation, including open banking, AI banking, remote banking, offline banking, and comprehensive banking. By 2022, the bank had provided remote banking services to over 37 million customers, significantly expanding the customer service radius. The alignment of FinTech with the goals of digital inclusive finance—equality, priority, and sustainability—helps maximize the financial system’s redistributive function, reducing discrimination and inequality in financial markets and contributing to the realization of shared prosperity [33] (pp. 16–32).

Finally, FinTech accelerates Inclusive Green Growth by empowering green technological innovation. Technological advancement, as a key driver of environmental protection and emissions reduction, is regarded as a crucial means for achieving dual carbon goals. Green technological innovation is essential for greening socioeconomic development. However, due to the public good nature of green technology, market failures exist, compounded by weak initial accumulation and the need for breakthroughs in key areas. Additionally, the positive externalities of green technology innovation often lead to underinvestment. FinTech, as a disruptive innovation [34] (pp. 4–13), is not only a data-driven emerging industry but also represents the application of disruptive technologies such as AI, machine learning, IoT, automation, robotics, and blockchain in the financial industry and monetary policy. These technologies offer critical support for overcoming bottlenecks in green technology innovation. The application of FinTech in areas such as environmental risk management, ESG assessment, and the calculation of environmental costs and benefits is also deepening. Moreover, FinTech efficiently transforms large-scale societal savings into productive investments, addressing the funding constraints of green technology innovation. During the pre-lending phase of green credit by commercial banks, FinTech enhances risk assessment and control capabilities, efficiently identifying enterprises’ credit needs and directing green credit resources toward precise allocation, thereby improving the efficiency of green credit. In the post-lending phase, FinTech plays a supervisory role by enabling financial institutions to achieve full-chain transparency and traceability of green assets, enhancing the efficiency and security of green investments. This, in turn, promotes green technological innovation and facilitates the green, low-carbon development of society [35] (pp. 34–48).

Based on this, the following hypotheses are proposed in this paper:

H1. 

FinTech promotes Inclusive Green Growth in cities.

H2. 

FinTech promotes Inclusive Green Growth by driving financial employment.

H3. 

FinTech promotes Inclusive Green Growth by expanding financial supply.

H4. 

FinTech promotes Inclusive Green Growth by empowering green technological innovation.

4. Research Design

4.1. Model Specification

Previous empirical studies have primarily relied on traditional causal inference models for hypothesis testing. However, these models face several challenges when handling high-dimensional control variables, including the curse of dimensionality and multicollinearity, which can undermine the accuracy of the estimates. Additionally, traditional models often assume linear relationships between explanatory and dependent variables, but non-linear relationships may exist in practice, further complicating causal inference. To address these challenges, the double machine learning (DML) model proposed by Chernozhukov et al. [36] (pp. 1–68) provides a robust approach by effectively handling complex non-linear relationships, controlling for potential biases, and offering advantages such as adaptability and non-parametric flexibility.

Inclusive Green Growth, as a comprehensive measure of urban transformation and development, is influenced by various socio-economic factors. Therefore, accurately identifying causal effects requires careful control of the confounding factors that impact Inclusive Green Growth. Traditional causal inference models struggle with the curse of dimensionality and multicollinearity when managing high-dimensional control variables, which can distort causal identification. The regularization algorithm embedded in the DML model helps mitigate the instability caused by multicollinearity by constraining variable coefficients, preventing regression coefficients from becoming difficult to interpret due to high correlations among variables. Additionally, the regularization algorithm penalizes excessively large coefficients, which reduces the risk of overfitting the model to training data, particularly when the sample size is small. Moreover, pre-set control variables may contain redundant information, where some variables may not exert a significant influence on the dependent variable. Regularization helps shrink or adjust the coefficients of these redundant variables, thereby reducing their interference with the results and minimizing unnecessary noise. This enables the model to focus on the key control variables that have a more substantial impact on Inclusive Green Growth. Through these techniques, the DML model effectively improves the robustness and accuracy of the results.

When assessing the impact of FinTech on Inclusive Green Growth, non-linear relationships may exist between variables, which traditional causal inference models are ill-equipped to handle. These models may suffer from specification bias, leading to unstable and unreliable estimates. In contrast, the DML framework addresses these non-linear issues by combining instrumental variable functions, two-stage residual regression, and sample splitting. Specifically, instrumental variables help address endogeneity bias, while two-stage residual regression leverages machine learning algorithms to capture non-linear relationships and extract residuals. Sample splitting further enhances the model’s generalization ability. By integrating these techniques, the DML framework provides more reliable causal inferences in the presence of high-dimensional and non-linear data.

$$Y={\theta }_{0}D+g\left({\rm X}\right)+U,E\left(U|X,D\right)=0$$

(1)

$$D=m\left({\rm X}\right)+V,E\left(V|X\right)=0$$

(2)

$$\stackrel{⌢}{V}=D-m(X)$$

(3)

In this context, Equation (1) represents the main model, where Y is the dependent variable, indicating Inclusive Green Growth; θ₀ is the primary regression coefficient in this study; D is the explanatory variable, representing FinTech; X is a high-dimensional set of control variables, which includes both linear and quadratic terms, with g(X) computed using machine learning methods; and U is the error term with a conditional mean of zero. Given that machine learning estimation might introduce regularization bias when handling large datasets, an auxiliary regression model in Equations (2) and (3) is further constructed. Here, m(X) is the regression function of the explanatory variable on the high-dimensional control variables, which also requires machine learning algorithms to estimate its specific form. V is the error term, with a conditional mean of zero. After estimating m(X) using machine learning algorithms, the residual $\stackrel{⌢}{\mathit{V}}$ is taken as an instrumental variable for D to perform a regression and obtain an unbiased estimate of the treatment coefficient.

4.2. Indicator Setting

4.2.1. Dependent Variable: Inclusive Green Growth (IGG)

Inclusive Green Growth is a composite category focused on the sustainable development concepts of economic, social, and ecological sustainability. Scholars primarily measure it based on its theoretical connotations, extensions, and characteristics, while also considering the simplicity, typicality, and objectivity of the indicators. This study follows the approach of Zhang and Li [6] (pp. 113–135) by selecting four dimensions—economic growth, income distribution, welfare inclusiveness, and environmental protection and pollution reduction—along with 15 specific indicators for the scientific assessment of Inclusive Green Growth. Given the multiple advantages of the base-period range entropy weight method, including high computational precision, resistance to subjective bias, and effective depiction of spatiotemporal trends, this method is employed to measure Inclusive Green Growth. The definitions of specific indicators and their calculated weights are presented in

Table 1. Inclusive Green Growth measurement system.

Dimensions	Indicators	Directions	Weights
Economic Growth	GDP Growth Rate	Positive	0.0143
	Per Capita GDP	Positive	0.1121
Income Distribution	Urban Residents’ Per Capita Disposable Income	Positive	0.0762
	Rural Residents’ Per Capita Disposable Income	Positive	0.0453
	Urban–Rural Residents’ Per Capita Disposable Income Ratio	Negative	0.0152
Inclusive Welfare	Urban Basic Medical Insurance Coverage	Positive	0.2284
	Urban Employees’ Basic Pension Insurance Coverage	Positive	0.2913
	Unemployment Insurance Coverage	Negative	0.0029
	Hospital and Clinic Beds per 10,000 People	Positive	0.1274
Environmental Protection and Pollution Reduction	Carbon Dioxide Emissions	Negative	0.0079
	Industrial Wastewater Emissions	Negative	0.0042
	Industrial Dust Emissions	Negative	0.0017
	General Industrial Solid Waste Comprehensive Utilization Rate	Positive	0.0231
	Sewage Treatment Plant Centralized Treatment Rate	Positive	0.0215
	Harmless Treatment Rate of Household Garbage	Positive	0.0285

.

Furthermore, building on the methodologies of Zhang et al. [37] (pp. 35–44) and Li and Dong [38] (pp. 4–18), this study develops an indicator measurement system for inclusive green total factor productivity. Capital stock and energy consumption (total natural gas, liquefied petroleum gas supply, and total electricity consumption of the entire society, converted into ten thousand tons of standard coal) are used as proxy variables for capital and energy inputs, based on the indicators listed in Table 1. A data envelopment analysis (DEA) model with undesirable outputs is employed to measure the inclusive green total factor growth rate, providing a robustness check by substituting the dependent variable. The specific indicators are outlined in

Table 2. Measurement system for inclusive green total factor productivity indicators.

Level 1	Level 2	Level 3
Input	Input Factors	Capital Input: Capital Stock
		Labor Input: Number of Employees at the End of the Year
		Energy Input: Energy Consumption
Output	Expected Output: Economic Growth	GDP Growth Rate
		Per Capita GDP
	Expected Output: Greening	General Industrial Solid Waste Comprehensive Utilization Rate
		Sewage Treatment Plant Centralized Treatment Rate
		Harmless Treatment Rate of Household Garbage
	Expected Output: Inclusive	Urban Basic Medical Insurance Coverage
		Urban Employees’ Basic Pension Insurance Coverage
		Public Library Books per 10,000 People
		Hospital and Clinic Beds per 10,000 People
	Unexpected Output: Environmental Pollution	Carbon Dioxide Emissions
		Industrial Wastewater Emissions
		Industrial Dust Emissions
	Unexpected Output: Social Injustice	Urban–Rural Residents’ Per Capita Disposable Income Ratio
		Unemployment Insurance Coverage

. The data are sourced from the China Urban Statistical Yearbook.

4.2.2. Explanatory Variable: FinTech

Compared to traditional measurement methods, this study suggests that using big data to obtain relevant information can more accurately reflect the development path and true state of regional FinTech. The core logic behind this approach is that advancements in FinTech, the establishment of enterprises, and technological breakthroughs are frequently reported in media coverage. Baidu, as China’s leading search engine, holds a dominant position in the domestic market. Therefore, by searching for FinTech-related keywords in Baidu News, the number of returned pages can effectively indicate the level of FinTech development in specific regions.

Following the methodology of Li [18] (pp. 81–98), this study first established a set of 48 FinTech-related keywords based on important documents and conferences, such as the National Science and Technology Innovation Plan, the Big Data Industry Development Plan, and the China FinTech Operation Report. These keywords include terms such as EB-level storage, NFC payment, and differential privacy technology and are classified into four main categories: Payment and Settlement (Pay), Lending and Capital Raising (Capital), Investment Management (Investment), and Market Infrastructure (Market), as shown in

Table 3. FinTech keywords.

Classification	Definition	Keywords
FinTech—Pay	Fund transfer and transactions	NFC Payment, Mobile Payment, Digital Currency, Third-Party Payment.
FinTech—Capital	Financing acquisition	Equity Crowdfunding, Internet Finance, Open Banking, Credit Reporting.
FinTech—Investment	Optimizing investment decisions and risk management	Investment Decision Support Systems, Intelligent Investment Advisors, Smart Financial Contracts, Quantitative Finance, Machine Learning, Deep Learning, Data Mining, Data Visualization, Natural Language Processing.
FinTech—Market	Infrastructure and technology required to support market operations	Intelligent Customer Service, Intelligent Data Analysis, Differential Privacy Technology, Multi-Party Secure Computing, Big Data, Distributed Computing, EB-Level Storage, Brain-Like Computing, Stream Computing, Green Computing, In-Memory Computing, Blockchain, Fusion Architecture, Business Intelligence, Identity Verification, Biometric Technology, Graph Computing, Networking, Text Mining, the Internet of Things, Cyber-Physical Systems, Virtual Reality, Mobile Internet, Billion-Level Concurrency, Heterogeneous Data, Cloud Computing, Semantic Search, Image Understanding, Speech Recognition.

. Next, these keywords were matched with 287 prefecture-level administrative units in mainland China, which include prefecture-level cities, autonomous prefectures, regions, and leagues. In China, prefecture-level administrative units are subdivisions under provincial-level entities, covering all major cities and key regional centers nationwide. Using all 287 prefecture-level units as a sample provides a more comprehensive regional perspective to assess FinTech development. In Baidu News’ advanced search, we conducted annual searches using combinations of a prefecture-level unit + a keyword, such as Beijing + Blockchain. The advanced search function in Baidu News returns the number of news pages containing both Beijing and Blockchain within a specified timeframe, spanning from 2011 to 2021. Through web scraping techniques, we extracted the source code of Baidu News’ advanced search pages to obtain the number of search results for each keyword by year. Finally, we aggregated the results for all keywords at each prefecture-level unit on an annual basis, applying equal weighting across years to ensure consistency. We then calculated the logarithm of these summed results to quantify the FinTech development level in each region. All data were sourced from Baidu News.

4.2.3. Control Variables

The study incorporates the following control variables: (i) Level of Government Intervention, measured as the ratio of general budgetary expenditures to regional GDP; (ii) Level of Technological Investment, measured as the ratio of scientific research expenditures to regional GDP; (iii) Level of Fixed Investment, measured as the ratio of total fixed asset investment to regional GDP; (iv) Level of Education Investment, measured as the ratio of education expenditures to regional GDP; and (v) Industrial Structure, measured as the ratio of value-added in the tertiary industry to value-added in the secondary industry. To improve the precision of the model, quadratic terms for each control variable are included. Additionally, city and year fixed effects are introduced as dummy variables to prevent information loss across urban and temporal dimensions. The data for these variables are obtained from the China Urban Statistical Yearbook.

4.2.4. Mechanism Variables

This study aims to explore the mechanism effects of FinTech on urban Inclusive Green Growth through the promotion of financial employment, the release of financial supply, and the empowerment of green technology innovation. Financial resources are measured by the natural logarithm of the number of financial employees per ten thousand people and the year-end balance of loans from financial institutions. The level of urban green technology innovation is measured by the natural logarithm of the number of green invention patents and green utility model patents applied for in the current year. The data for this study are sourced from the National Intellectual Property Administration and the China Urban Statistical Yearbook.

4.3. Data Processing and Descriptive Statistics

For the construction of Inclusive Green Growth metrics, certain data, such as the number of insured individuals, have been recorded since 2011. To ensure consistency in statistical analysis, this study focuses on 287 Chinese prefecture-level administrative units, with data spanning from 2011 to 2021. However, when measuring the inclusive green total factor growth rate using the SBM model with undesirable outputs, the introduction of new indicators has led to some missing data in certain regions. Consequently, the study narrows the focus to 279 prefecture-level cities while maintaining the sample period unchanged. Missing data are addressed through interpolation. Detailed descriptive statistics are provided in

Table 4. Definition and descriptive statistics of main variables.

Variable Category		Variable Name and Specific Representation	Mean	Standard Deviation	Minimum	Maximum
Dependent Variables		Inclusive Green Growth	0.1841	0.1296	0.0467	1.4209
		Inclusive Green Total Factor Productivity	0.6818	0.2947	0.0005	1
Explanatory Variables		FinTech	3.3742	1.4615	0.0000	7.4911
Control Variables	Government Intervention	Ratio of general budgetary expenditures to regional GDP	0.1996	0.0844	0.0976	0.4022
	Technological Investment	Ratio of scientific expenditures to regional GDP	0.0641	0.0637	0.0008	0.1814
	Fixed Asset Investment	Ratio of total fixed asset investment to regional GDP	0.8306	0.2958	0.3740	1.5022
	Education Investment	Ratio of education expenditures to regional GDP	0.0353	0.0175	0.0135	0.1079
	Industrial Structure	Ratio of value-added in the tertiary industry to value-added in the secondary industry	1.0234	0.4437	0.4548	2.1301
Mechanism Variables	Financial Employment	Natural logarithm of the number of financial employees per ten thousand people	0.2672	0.9052	−2.3126	4.3442
	Financial Supply	Year-end balance of financial institution loans	16.5239	1.1633	13.7234	20.5984
	Green Technological Innovation	Natural logarithm of the number of green invention patents applied for in the current year	4.3878	1.6170	0.6931	8.2099
		Natural logarithm of the number of green utility model patents applied for in the current year	3.8999	1.7404	0.0000	8.1945

.

5. Empirical Analysis

5.1. Baseline Regression

This study employs a double machine learning (DML) model to assess the impact of FinTech on Inclusive Green Growth. The study utilizes an out-of-sample cross-validation approach, specifically the K-fold cross-validation method, to enhance data utilization, prevent overfitting, and ensure more robust parameter estimates. Based on the optimal 5-fold cross-validation identified by Chernozhukov et al. [36] (pp. 1–68), the study adopts a 5-fold cross-validation setup. Furthermore, the random forest algorithm is used for prediction in both the main and auxiliary regressions, with the results presented in

Table 5. Baseline regression results.

	(1) IGG	(2) IGG	(3) IGG	(4) IGG	(5) IGG
FinTech	0.0179 *** (0.0013)
FinTech—Pay		0.0193 *** (0.0014)
FinTech—Capital			0.0169 *** (0.0014)
FinTech—Investment				0.0154 *** (0.0013)
FinTech—Market					0.0092 *** (0.0014)
Controls (1st Order)	Yes	Yes	Yes	Yes	Yes
Controls (2nd Order)	Yes	Yes	Yes	Yes	Yes
Time Fixed Effects	Yes	Yes	Yes	Yes	Yes
City Fixed Effects	Yes	Yes	Yes	Yes	Yes
Sample Size	3157	3157	3157	3157	3157
R-Squared	0.0795	0.0723	0.0596	0.0581	0.0587

Note: *** p < 0.01. Standard errors in parentheses.

. Column 1 controls for city fixed effects, time fixed effects, as well as the linear and quadratic terms of the control variables, and the regression coefficients are significantly positive. This indicates a significant positive impact of FinTech on Inclusive Green Growth, validating hypothesis H1.

In addition, based on the classification of FinTech by the Financial Stability Board (FSB) and the Basel Committee on Banking Supervision (BCBS), this study subdivides FinTech keywords into four categories: Payment and Settlement (Pay), Deposit and Capital Raising (Capital), Investment Management (Investment), and Market Infrastructure (Market). This classification further explores the impact of segmented FinTech businesses on Inclusive Green Growth. Columns 2–5 present the regression results. It can be observed that all four categories of segmented businesses have a significant positive impact on Inclusive Green Growth. Specifically, core businesses such as Payment and Settlement, Deposit and Capital Raising, and Investment Management are closely related to the flow of funds and possess strong financial attributes. These businesses can promote the investment and development of green projects by optimizing resource allocation and enhancing the efficiency of fund flows. As a result, they have a more direct role in driving Inclusive Green Growth, with relatively larger regression coefficients. In contrast, Market Infrastructure businesses have smaller regression coefficients for Inclusive Green Growth, which may be due to the large investment scale and long construction period of such infrastructure. However, this type of infrastructure is crucial for the stability and sustainable development of the financial system and can still support Inclusive Green Growth in the long term.

5.2. Robustness Checks

5.2.1. Replacement of the Dependent Variable

Existing research on measuring Inclusive Green Growth can be categorized into two dimensions: level and efficiency. The level dimension typically uses a multidimensional indicator system for comprehensive evaluation, as validated in the baseline regression. In contrast, the efficiency dimension employs data envelopment analysis (DEA) to assess multiple indicators of inputs and outputs. To ensure the robustness and comprehensiveness of the research results, this study constructs a slacks-based measure (SBM) model with undesirable outputs to measure the efficiency aspect of inclusive green total factor growth. According to the regression results in

Table 6. Robustness checks.

	(1)	(2)				(3)
	Adjust Sample	Adjust Sample				Change Sample Interval
		1% Winsorization	5% Winsorization	Trim Part of Cities	Shorten Sample Interval	K-Folds = 10	Normal Sample Split 1:1
FinTech	0.0177 *** (0.0019)	0.0173 *** (0.0014)	0.0140 *** (0.0007)	0.0169 *** (0.0009)	0.0195 *** (0.0022)	0.0198 *** (0.0018)	0.0162 *** (0.0018)
Controls (1st Order)	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Controls (2nd Order)	Yes	No	Yes	No	Yes	Yes	No
Time Fixed Effects	Yes	Yes	Yes	Yes	Yes	Yes	Yes
City Fixed Effects	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Sample Size	3070	3157	3157	2838	2009	3157	3157
R-Squared	0.1348	0.0668	0.1256	0.1365	0.0561	0.0928	0.0613
	(4)			(5)
	Change ML Models			Endogeneity Check
	Bagging	Boosting	Neural Network	Distance from Each City to Hangzhou	Weak Endogeneity Subsample	Lagged Dependent Variable
FinTech	0.0183 *** (0.0013)	0.0205 *** (0.0013)	0.0151 *** (0.0009)	0.0597 *** (0.0138)	0.0098 *** (0.0005)	0.0162 *** (0.0015)
Controls (1st Order)	Yes	Yes	Yes	Yes	Yes	Yes
Controls (2nd Order)	Yes	No	Yes	No	Yes	Yes
Time Fixed Effects	Yes	Yes	Yes	Yes	Yes	Yes
City Fixed Effects	Yes	Yes	Yes	Yes	Yes	Yes
Sample Size	3157	3157	3157	3157	2244	2870
R-Squared	0.0812	0.1041	0.0941	0.1788	0.2076	0.0494

Note: *** p < 0.01. Standard errors in parentheses.

, column (1), the coefficient for the impact of FinTech on the inclusive green total factor growth rate remains significantly positive. Therefore, both level and efficiency measurements confirm the positive effect of FinTech on Inclusive Green Growth.

5.2.2. Adjustments to the Research Sample

The research sample is adjusted to ensure the reliability of empirical results using the following methods:

(1)

To address potential biases caused by outliers, this study applies truncation at the 1st and 99th percentiles, as well as the 5th and 95th percentiles, for all variables.

(2)

Given the significant differences in FinTech-related policy support and development conditions between provincial capitals, municipalities directly under the central government, and other regions, provincial capitals and municipalities are excluded to ensure the generalizability of the results.

(3)

Since 2015, following the release of the Guiding Opinions on Promoting the Healthy Development of Internet Finance by the People’s Bank of China and ten other ministries, the FinTech sector has entered a new phase of financial innovation, regulatory responsibility, and market order. This year is also referred to as the Year of FinTech Regulation in China. It is believed that FinTech development post-2015 has been more beneficial to socio-economic development. Therefore, to exclude the impact of other factors on Inclusive Green Growth within the sample period, the data sample is restricted to the period from 2015 to 2021. The regression results in Table 6, column (2), show that the coefficient for FinTech’s impact on Inclusive Green Growth remains significantly positive, reaffirming the conclusions of the baseline model.

5.2.3. Model Specification Adjustments

To avoid potential biases due to model specification errors, the robustness of the conclusions is tested through the following approaches:

(1)

Changing the sample split ratio in the double machine learning model from the previous 5-fold cross-validation to 10-fold cross-validation and a 1:1 training and testing sample split ratio.

(2)

Replacing the random forest algorithm previously used for prediction with bagging, boosting, and neural network algorithms. Specific regression results are shown in Table 6, columns (3) to (4). The results indicate that the conclusions remain robust across different sample split methods and machine learning algorithms.

5.2.4. Endogeneity Issues

To further address the potential issues of omitted variables and reverse causality, the following methods are used for the relevant tests:

(1)

Instrumental Variable Method: Following the approach of He and Song [39] (pp. 58–68), the distance from each city to Hangzhou is used as an instrumental variable for FinTech development. Hangzhou, the birthplace of FinTech giants like Alipay and Ant Financial, has a leading position in FinTech development. Thus, cities geographically closer to Hangzhou are expected to have better FinTech development, satisfying the relevance principle of instrumental variables. Additionally, the distance to Hangzhou is unlikely to have a direct impact on current city-level Inclusive Green Growth, satisfying the exogeneity principle.

(2)

Weak Endogeneity Sub-Sample: To further address endogeneity, the sample is divided into high and low Inclusive Green Growth level groups based on the mean level of Inclusive Green Growth. The high-level group is excluded, and only the low-level group is used for regression analysis.

(3)

Lagged Independent Variable: The independent variable (FinTech) is lagged by one period to mitigate endogeneity issues. The specific regression results are presented in Table 6 (5). The results show that FinTech significantly enhances Inclusive Green Growth, consistent with the conclusions of the baseline model.

5.3. Mechanism Analysis

The empirical results have robustly verified the positive impact of FinTech on Inclusive Green Growth. However, understanding how FinTech promotes Inclusive Green Growth is also a key focus of this study. Based on the theoretical analysis presented, the mechanisms through which FinTech may influence Inclusive Green Growth include the following: financial employment promotion, financial supply release, empowerment of green technological innovation, and green finance effects.

To investigate these mechanisms, this study builds upon the previous double machine learning model and adopts the testing procedures of Wen et al. [40] (pp. 614–620) to examine the transmission mechanisms of FinTech on Inclusive Green Growth. The specific tests are outlined as follows:

$$Y={\theta }_{1}D+g_1\left({\rm X}\right)+U_1,E\left(U_1|X,D\right)=0$$

(4)

$$M={\theta }_{2}D+g_2\left({\rm X}\right)+U_2,E\left(U_2|X,D\right)=0$$

(5)

$$Y={\lambda }_{1}D+{\lambda }_{2}M+g_3\left({\rm X}\right)+U_3,E\left(U_3|X,D\right)=0$$

(6)

Here, M represents the mediating variable, and the remaining variables are consistent with those defined in Equations (1)–(3) above. To ensure the robustness of the research conclusions, additional tests, including the Sobel test, Aroian test, and Goodman test, are conducted to provide a comprehensive assessment of the mechanisms through which FinTech impacts Inclusive Green Growth.

5.3.1. Financial Employment Promotion

To examine whether FinTech influences Inclusive Green Growth through its effect on financial employment, this study uses the natural logarithm of the number of financial professionals in various cities (Employment) as a measure of financial employment and conducts regression analysis. Column 1 in

Table 7. Mediation test of financial employment effect.

	IGG	Employment	IGG
FinTech	0.0179 *** (0.0013)	0.0678 *** (0.0096)	0.0143 *** (0.0012)
Employment			0.0348 *** (0.0026)
Sobel z			7.4592 ***
Aroian z			7.4521 ***
Goodman z			7.4666 ***
Controls (1st Order)	Yes	Yes	Yes
Controls (2nd Order)	Yes	Yes	Yes
Time Fixed Effects	Yes	Yes	Yes
City Fixed Effects	Yes	Yes	Yes
Sample Size	3157	3157	3157
R-Squared	0.0795	0.0201	0.1382

Note: *** p < 0.01. Standard errors in parentheses.

presents the results of the impact of FinTech on Inclusive Green Growth without considering financial employment. Column 2 shows the results of FinTech’s impact on financial employment, indicating that the coefficient for FinTech is significantly positive, suggesting that FinTech promotes financial employment. Column 3 reports the combined effect of financial employment and FinTech on Inclusive Green Growth, where all coefficients, including the Sobel z, Aroian z, and Goodman z statistics, are significantly positive. This indicates that FinTech promotes Inclusive Green Growth by enhancing financial employment, thereby supporting H2.

This finding suggests that FinTech effectively stimulates the financial employment factor. As FinTech advances, certain roles within traditional financial institutions face disruptions, leading to a reduction in demand for conventional labor to adapt to transformation and new technological developments. However, this does not imply a decrease in the overall employment or entrepreneurial opportunities for financial professionals. Instead, FinTech raises the threshold for labor supply while increasing the demand for financial employment positions under the new conditions. Thus, FinTech can effectively reallocate surplus labor in the market, ensuring employment market sufficiency and balance and injecting vitality into economic growth.

5.3.2. Financial Supply Release

To investigate whether FinTech influences Inclusive Green Growth through its effect on financial supply release, this study uses the natural logarithm of the year-end balance of loans from financial institutions in various cities (Loan) as a measure of financial supply and conducts regression analysis. Column 1 in

Table 8. Examination of the mediating mechanism of the financial supply effect.

	IGG	Loan	IGG
FinTech	0.0179 *** (0.0013)	0.2152 *** (0.0097)	0.0046 *** (0.0013)
Loan			0.0619 *** (0.0025)
Sobel z			18.0536 ***
Aroian z			18.0479 ***
Goodman z			18.0593 ***
Controls (1st Order)	Yes	Yes	Yes
Controls (2nd Order)	Yes	Yes	Yes
Time Fixed Effects	Yes	Yes	Yes
City Fixed Effects	Yes	Yes	Yes
Sample Size	3157	3157	3157
R-Squared	0.0795	0.1793	0.2800

Note: *** p < 0.01. Standard errors in parentheses.

is identical to Column 1 in Table 7. Column 2 presents the results of FinTech’s impact on the year-end balance of loans from financial institutions, showing that the coefficient for FinTech is significantly positive, indicating that FinTech promotes credit supply. Column 3 reports the combined effect of year-end loan balances and FinTech on Inclusive Green Growth, with all coefficients, including Sobel z, Aroian z, and Goodman z statistics, being significantly positive. This result suggests that FinTech impacts Inclusive Green Growth by stimulating credit provision, thereby supporting H3.

At the current stage, national policies strongly support FinTech, with major projects in payment and settlement, registration and custody, credit enhancement and rating, asset trading, and data management being implemented. Improving financial service levels through FinTech has become a policy focus. Additionally, rapid deployment of FinTech applications has led to disruptive innovations in areas such as client risk assessment and credit pricing. This efficient adjustment of credit structure and quality has made basic financial services more accessible, fostering an inclusive transformation in urban development.

5.3.3. Green Technological Innovation

To assess whether FinTech influences Inclusive Green Growth through its effect on green technological innovation, this study uses the natural logarithms of the number of green invention patents (M1) and green utility model patents (M2) applied for each year as measures of urban green technological innovation and conducts regression analysis.

Table 9. Examination of the green technology innovation effect.

(a) Number of Green Invention Patents				(b) Number of Green Utility Model Patents
	IGG	M1	IGG		IGG	M2	IGG
FinTech	0.0179 *** (0.0013)	0.4262 *** (0.0151)	0.0085 *** (0.0015)	FinTech	0.0179 *** (0.0013)	0.3138 *** (0.0166)	0.0120 *** (0.0014)
M1			0.0222 *** (0.0018)	M2			0.0189 *** (0.0016)
Sobel z			12.7410 ***	Sobel z			11.1251 ***
Aroian z			12.7347 ***	Aroian z			11.1502 ***
Goodman z			12.7473 ***	Goodman z			11.1353 ***
Controls (1st Order)	Yes	Yes	Yes	Controls (1st Order)	Yes	Yes	Yes
Controls (2nd Order)	Yes	Yes	Yes	Controls (2nd Order)	Yes	Yes	Yes
Time Fixed Effects	Yes	Yes	Yes	Time Fixed Effects	Yes	Yes	Yes
City Fixed Effects	Yes	Yes	Yes	City Fixed Effects	Yes	Yes	Yes
Sample Size	3157	3157	3157	Sample Size	3157	3157	3157
R-Squared	0.0795	0.2635	0.1415	R-Squared	0.0795	0.1379	0.1334

Note: *** p < 0.01. Standard errors in parentheses.

presents the results of the mediation effects of green invention patents and utility model patents on the impact of FinTech on Inclusive Green Growth. Columns 1 in Table 9 (a,b) show the results of FinTech’s impact on Inclusive Green Growth without considering green technological innovation. Column 2 shows the impact of FinTech on green technological innovation, with significantly positive coefficients, indicating that FinTech enhances green technological innovation. Column 3 reports the combined effects of green technological innovation and FinTech on Inclusive Green Growth, with the coefficients for both FinTech and green technological innovation being significantly positive. The Sobel z, Aroian z, and Goodman z statistics are also significantly positive, suggesting that FinTech affects Inclusive Green Growth through promoting green technological innovation, thus supporting H4.

These results indicate that FinTech contributes to strengthening green technology research and supporting the green transformation of the economy and society. Historically, industrial development has been dominated by natural resource extraction and labor-intensive traditional manufacturing, leading to severe ecological and environmental issues. Industrial upgrading and technological innovation are closely linked and require significant core technological advancements. As a driver of fundamental technological innovation, FinTech provides substantial technical support for green technology research and development. Additionally, FinTech helps guide social capital towards green industries, offering financial support for green technological innovation, and thus injects strong momentum into the green, low-carbon transformation of the economy.

5.4. Heterogeneity Analysis

5.4.1. The Differential Impact of Resource Endowments

Resource endowments play a significant role in shaping regional economic transformation. Regions with abundant resources often develop an industrial structure dominated by natural resource exploitation, generating substantial government revenue from resource wealth. This reduces incentives to develop the manufacturing sector, creating a crowding-out effect on endogenous drivers of growth, such as human capital and technology [41] (pp. 2314–2327). Additionally, regions with rich resource endowments are more prone to rent-seeking and corrupt behavior by local governments, leading to the unnecessary waste of social resources [42] (pp. 15–31). Therefore, we hypothesize that FinTech is more conducive to promoting Inclusive Green Growth in regions with poorer resource endowments.

To this end, based on the Notice on the Sustainable Development Plan for Resource-based Cities (2013–2020), this study classifies all sample cities into resource-based and non-resource-based cities. For analysis purposes, a dummy variable (Resource) is set, where the value is 1 if the sample city is a resource-based city and 0 otherwise. Additionally, this study introduces an interaction term between FinTech and the Resource variable to examine how FinTech impacts Inclusive Green Growth under different resource endowments. The regression results in column (1) of

Table 10. Heterogeneity analysis results.

	(1) IGG	(2) IGG	(3) IGG	(4) IGG
FinTech	0.0197 *** (0.0013)	0.0115 *** (0.0014)	0.0184 *** (0.0012)	0.0034 (0.0094)
Resource	−0.0072 (0.0064)
FinTech × Resource	−0.0054 *** (0.0016)
Digital		−0.0064 (0.0063)
FinTech × Digital		0.0606 *** (0.0081)
Environmental Regulation			−0.0170 (0.0094)
FinTech × Environmental Regulation			0.0077 ** (0.0024)
Green Finance				0.1236 *** (0.0331)
FinTech × Green Finance				0.0353 *** (0.0059)
Controls (1st Order)	Yes	Yes	Yes	Yes
Controls (2nd Order)	Yes	Yes	Yes	Yes
Time Fixed Effects	Yes	Yes	Yes	Yes
City Fixed Effects	Yes	Yes	Yes	Yes
Sample Size	3157	3157	3157	3157
R-Squared	0.1179	0.1187	0.1072	0.1626

Note: ** p < 0.05, *** p < 0.01. Standard errors in parentheses.

show that the interaction term FinTech × Resource is significantly negative at the 1% level, indicating that the degree of regional resource endowment weakens the positive effect of FinTech on Inclusive Green Growth. This suggests that in resource-based cities, the higher the level of resource endowment, the weaker the effect of FinTech on green growth, and that the role of FinTech in promoting inclusive and green economic growth is more limited in resource-rich cities.

This phenomenon can be further explained by the resource curse theory. Resource-based cities in China are often highly dependent on resource extraction and heavy industries, which leads to a relatively mono-structural and resource-intensive economy, resulting in environmental pollution and slow industrial upgrading. In this context, the innovative and transformative potential of FinTech may not be fully realized in resource-based cities. On the other hand, cities with lower resource dependence and more diversified economic structures within resource-based regions can more effectively leverage FinTech to promote a green economy and inclusive growth. Therefore, this result not only further highlights the resource curse faced by some resource-based regions in China but also reflects the regional disparities in the role of FinTech in driving sustainable development.

5.4.2. The Differential Impact of Digital Infrastructure Development

In recent years, digital infrastructure has become a focal point of government initiatives. The 14th Five-Year Plan explicitly calls for the strategic development of new infrastructure, emphasizing the accelerated construction of 5G networks, industrial internet, and big data centers. As a key component of new infrastructure, digital infrastructure not only facilitates the development of smart cities and digital villages—narrowing the urban–rural income gap—but also drives green technological innovation, achieving both emission reductions and efficiency improvements. This aligns with the core principles of green development. However, the impact of FinTech on Inclusive Green Growth may vary depending on the level of digital infrastructure development. Cities with more advanced digital infrastructure typically possess superior information technologies and communication networks, which provide a stronger foundation for FinTech innovation and its potential to support inclusive growth and sustainable development. Based on this premise, we hypothesize that the effect of FinTech on Inclusive Green Growth will be more pronounced in cities with higher levels of digital infrastructure. To further explore the role of digital infrastructure in promoting Inclusive Green Growth through FinTech, this study draws on the methodology of Zhao [43] (pp. 80–92) and selects six indicators to construct a digital infrastructure development index (Digital) from both the input and output dimensions of digital infrastructure. Specifically, the input of digital infrastructure includes fiber optic cable density, per capita broadband internet access ports, and the number of related industry professionals. The output of digital infrastructure includes telecom service revenue, mobile phone penetration rate, and internet penetration rate. These indicators comprehensively reflect the level of urban digitalization and its development potential. To further investigate the relationship between FinTech and digital infrastructure, this study introduces the interaction term between FinTech and the digital infrastructure development index (Digital).

The regression results in Column (2) of Table 10 show that the interaction term FinTech × Digital is significantly positive at the 1% level, indicating that improvements in local digital infrastructure levels significantly enhance the promoting effect of FinTech on Inclusive Green Growth. This result suggests that in cities with higher levels of digital infrastructure, FinTech innovation and development are more effective in promoting a green economy and inclusive growth. As digital infrastructure improves, urban residents, businesses, and governments are able to more easily access and process information, thereby promoting the innovation of green technologies, the widespread adoption of green financial services, and the enhancement of their social benefits.

This finding is consistent with expectations and reflects that digital infrastructure not only provides the technological and data support for FinTech, but also creates the necessary foundational conditions for promoting green development. The improvement of digital infrastructure enables cities to better address environmental and social challenges, facilitating the popularization and innovation of green financial services. This process not only drives the green transformation of urban economies but also provides strong technological support for enhancing inclusive growth. Therefore, strengthening the construction of digital infrastructure, particularly investments in information and communication technologies and internet penetration, becomes an important pathway for effectively combining FinTech and green growth.

5.4.3. The Differential Impact of Environmental Regulation

While resource endowments and digital infrastructure influence the impact of FinTech on Inclusive Green Growth from a production factors perspective, environmental regulation plays a role in determining this impact from a policy constraint perspective. Traditional economic theory posits that environmental protection and economic growth are mutually inhibitive. However, the Porter Hypothesis argues that environmental policies can stimulate technological innovation, creating a win-win scenario for both environmental protection and economic growth. This hypothesis has evolved into three variants: the Narrow Porter Hypothesis, the Weak Porter Hypothesis, and the Strong Porter Hypothesis. The context of this study provides an opportunity to re-examine the Porter Hypothesis, particularly by exploring how FinTech influences Inclusive Green Growth under different levels of environmental regulation. This can further contribute to the ongoing academic debate regarding whether environmental regulations help reconcile economic growth with environmental protection.

To verify the above hypothesis, this study draws on the methodology of Zhang and Chen [44] (pp. 78–93), constructing a keyword library of 27 terms covering three aspects: environmental goals, environmental work targets, and environmental measures. Text mining techniques are then employed to conduct statistical analysis of local government work reports. Specifically, this study calculates the ratio of the frequency of environmental-related terms to the total word count in each city’s government work report, which is used to measure the level of environmental regulation (Environmental Regulation). Additionally, this study introduces an interaction term between FinTech and Environmental Regulation to further explore the synergistic effects of these two factors in driving Inclusive Green Growth.

The regression results in Table 10, Column (3), show that the interaction term FinTech × Environmental Regulation is significantly positive at the 1% level, indicating that an increase in environmental regulation significantly enhances the promoting effect of FinTech on Inclusive Green Growth. This conclusion confirms the synergistic effects of FinTech and environmental regulation in driving Inclusive Green Growth in cities. Specifically, in cities with higher levels of environmental regulation, the promoting effect of FinTech on Inclusive Green Growth is more pronounced. This is primarily because in regions with stricter environmental regulations, local governments place more emphasis on environmental protection and sustainable development, while micro-level entities (such as businesses and consumers) face stricter environmental policy constraints. In this context, FinTech provides these entities with more flexible forms of economic activity, including innovative financial products and services such as green finance, environmental credit, and green bonds, helping businesses comply with environmental regulations while maintaining economic efficiency. At the same time, the development of FinTech in these cities can effectively address the shortcomings of traditional business models, enhancing the demand for green technologies and eco-friendly products among businesses and consumers, thereby further promoting green economic growth. Moreover, regions with stronger environmental regulation often have a clearer policy orientation, with local governments implementing more stringent environmental policies to further guide the allocation of social resources toward greener uses. Under this policy-driven framework, FinTech innovation can provide more targeted financial support and technological services for green industries, facilitating the innovation of green technologies and the development of green industry chains. Therefore, the synergistic effects of FinTech and environmental regulation can drive cities to better achieve Inclusive Green Growth under conditions of stronger environmental regulation.

5.4.4. The Differential Impact of Green Finance Development

Compared to the previous perspectives, green finance represents a policy incentive that influences the impact of FinTech on Inclusive Green Growth. As a critical tool for achieving high-quality economic development, green finance not only guides the green transformation of industries, expands employment opportunities, and reduces income inequality, but also improves the quality of economic growth by supporting green innovation. However, due to its relatively late development in China, Ning and She [45] (pp. 62–66) argue that green finance has not yet contributed to macroeconomic growth and has, in fact, hindered it. Additionally, there remains a lack of specific policy guidance for the integration of FinTech into green finance, as well as regulatory sandboxes for fostering green FinTech innovation. In practice, the integration of FinTech and green finance is still in its early stages, with challenges such as significant technological risks and data security issues. Whether this integration can effectively enhance Inclusive Green Growth requires further empirical verification.

To further explore the role of FinTech in promoting Inclusive Green Growth, this paper references the methods of Liu and He [46] (pp. 37–52), as well as Li [47] (pp. 65–77), constructing a Green Finance Index based on seven fundamental indicators: green credit, green investment, green insurance, green bonds, green support, green funds, and green equity. These indicators comprehensively reflect various aspects of green finance development, spanning from credit to investment, insurance, and capital markets, providing a systematic evaluation of the green finance system. The data sources primarily include local statistical yearbooks, environmental status bulletins, and specialized statistical yearbooks such as the China Science and Technology Statistical Yearbook, China Energy Statistical Yearbook, China Financial Yearbook, China Agricultural Statistical Yearbook, and China Industrial Statistical Yearbook, ensuring the authority and reliability of the data. To explore the relationship between FinTech and green finance, this paper also introduces an interaction term between FinTech (FinTech) and the Green Finance Index (Green Finance), using regression analysis to assess their synergistic effects. The regression results in Table 10, Column (4) show that the interaction term between FinTech and Green Finance is significantly positive at the 1% level, indicating that the combination of FinTech and green finance has a more pronounced effect on promoting Inclusive Green Growth in cities with higher levels of green finance. This suggests that, with the rapid development of FinTech and its deep integration with green finance in China, the potential of FinTech to empower green finance is continuously being unlocked, playing a significant role in promoting both green development and inclusive economic growth.

Behind this result, FinTech is providing more efficient technical support and service capabilities for green finance, driving innovation in green finance products and services. For example, through technologies such as blockchain, big data, and artificial intelligence, FinTech enhances the transparency and risk management capabilities of green financial products like green credit and green investment, reducing the financing costs of green projects and better supporting the growth of the green economy. Under the influence of FinTech, green finance can more effectively allocate resources, support green technological innovation, and increase the accessibility of green projects, thereby promoting a low-carbon, environmentally friendly economic transformation.

Thus, the findings of this paper indicate that the integration of FinTech and green finance provides a new driving force and path for achieving Inclusive Green Growth. FinTech injects new vitality into green finance and promotes the sustainable development of the green economy. As the integration deepens, green finance will have a broader and more positive impact, advancing China’s goal of achieving a green, low-carbon transformation.

6. Conclusions and Recommendations

6.1. Conclusions

Accurately identifying the driving mechanisms through which FinTech influences Inclusive Green Growth is critical for promoting high-quality economic development. This study utilizes data from 287 Chinese cities spanning from 2011 to 2021, employing text mining techniques for the precise measurement of FinTech and applying a double machine learning model to explore the mechanisms behind its impact on Inclusive Green Growth. The study also considers heterogeneous effects across different contexts, such as resource endowments, environmental regulation, and green finance. The findings reveal that FinTech significantly promotes Inclusive Green Growth, with this result holding true under a series of robustness and endogeneity tests. In terms of mechanisms, FinTech drives green technological innovation, expands financial supply, and generates financial employment, all of which influence Inclusive Green Growth at the city level. Furthermore, the impact of FinTech varies across different types of cities, being particularly beneficial in cities with high levels of digital infrastructure, environmental regulation, and green finance.

6.2. Recommendations

Based on these conclusions, several policy recommendations are proposed to enhance the role of FinTech in promoting Inclusive Green Growth. First, it is essential to establish risk funds, provide low-interest loans, and offer other forms of financial support to encourage FinTech startups and innovation projects, particularly those aimed at fostering sustainable development and green growth. Second, the adoption of FinTech should be vigorously promoted by developing infrastructure for digital financial services while ensuring robust cybersecurity and data privacy protections. In rural areas, targeted professional training in FinTech is necessary to enable more people to benefit from the opportunities FinTech presents. Third, the creation of green FinTech innovation centers is recommended to provide resources, training, and networking support for FinTech companies and innovators, enabling them to collaboratively address environmental and social challenges. Finally, particular attention should be given to resource-dependent cities and those with lower levels of digital infrastructure, environmental regulation, and green finance. Enhancing top-level design and promoting the adoption of FinTech technologies and low-carbon living concepts are crucial steps in better integrating FinTech into efforts to achieve Inclusive Green Growth at the macroeconomic level.