Will Digital Inclusive Finance Improve the Quality and Quantity of SMEs’ Green Innovation?

Abstract

Whether SMEs can become a significant player in green innovation and reshape the green innovation landscape in China largely depends on their ability to effectively address the lack of momentum for green innovation among SMEs. Utilizing data from China’s SMEs listed on the growth enterprise market between 2011 and 2022, this study empirically examines the effects and underlying mechanisms of digital financial inclusion on green innovation in SMEs, considering both the financial market supply and corporate financing constraints. The results indicate that digital inclusive finance significantly enhances both the quantity and quality of green innovation in SMEs. Moreover, digital financial inclusion particularly improves the quality and quantity of green innovation in SMEs with strong ESG performance and high equity concentration, compared to those with weaker ESG performance and lower equity concentration. Heterogeneity analysis reveals that digital inclusive finance can improve the quality and quantity of state-owned enterprises and enterprises in the eastern region. Still, it has no significant impact on the quantity of green innovation for enterprises in the central and western regions. Regarding the impact mechanism, digital inclusive finance encourages SMEs to engage proactively in green innovation by mitigating financing constraints and increasing R&D investments. The findings of this paper not only reveal how SMEs can overcome the restrictive mechanisms of green innovation through digital inclusive finance, but also provide critical insights for improving the imbalance in the structure of green innovation within China.

1. Introduction

The targets to reach the carbon emission peak and achieve carbon neutrality have been paramount for China’s green economic development, promoting green development and accelerating green transformation. Accelerating the comprehensive green transformation of economic and social development was included in the general objectives of deepening comprehensive reform at the Third Plenary Session of the 20th Central Committee of the Communist Party of China. The Central Committee of the Communist Party of China and The State Council promulgated the Opinions on Accelerating the Comprehensive Green Transformation of Economic and Social Development, proposing that it is necessary to take the work of peaking carbon neutrality as the guide, jointly promote carbon reduction, pollution reduction, green expansion, and growth, deepen the reform of the ecological civilization system, and thus improve the mechanism for green and low-carbon development. Different from traditional technological innovation, green innovation has economic, social, and environmental benefits [1]. It has turned into an important method to realize economic and social development and ecological environmental protection “win–win”, which is also the key to realizing high-quality economic development in China [2]. However, China’s green innovation is currently facing the dilemma of poor quality of innovation results and insufficient green innovation momentum due to excessive reliance on traditional technologies [3]. At present, China’s economy is at a critical stage of promoting high-quality development. Being an essential subject in microeconomic activities and an active participant in innovation, the quantity and quality of green technology innovation are directly related to the improvement of innovation efficiency. Therefore, discussing ways to improve the quality and increase the quantity of enterprise green innovation has vital significance, which can accelerate the transformation and upgrading of enterprise green development and actively and steadily promote the implementation of the national strategy related to carbon peak and carbon neutrality.

According to the Statistical Analysis Report on Green and Low Carbon Patents (2024) released by the State Intellectual Property Office in 2024, there were 14 enterprises from China in the top 20 applicants of green and low-carbon patent applications in China from 2016 to 2023, including 8 large enterprises and 6 universities and research institutions. From green patent innovation subjects in the report, by the end of 2023, 58 large enterprises accounted for 58 of the top 100 patentees holding effective green and low-carbon patents, which decreased by 6 compared to 2016; higher education institutions accounted for 42, an increase of 6 compared to 2016. It is not difficult to find that China’s green innovation is mainly dominated by large enterprises and higher education institutions, while most SMEs lack the motivation to carry out green innovation [4]. This reflects an imbalance in the structure of current green innovative entities in China, and challenges the promotion of enterprise-oriented and market-oriented green innovation strategies. Green innovation is a key link to promote China’s green development and a necessary engine to drive high-quality economic development. As Li et al. (2023) and He et al. (2022) suggest, there is a particular need for in-depth research into the causes of imbalance in the significant structure of green innovation [5,6]. On one hand, due to the dual externalities of both public knowledge and environmental protection [7], compared with traditional technological innovation, green innovation has higher risks, more capital investment, and higher uncertainty of expected returns, resulting in the lack of sufficient motivation for SMEs to promote green innovation [5]. On the other hand, the realistic challenge in SMEs is the low level of green technology innovation, insufficient self-owned funds, and discriminatory credit in the traditional financial market [8]. The information asymmetry between banks and enterprises is more serious; thus, SMEs suffer more serious financing constraints, limiting their ability to carry out green innovation [9]. Therefore, solving the central structural imbalance in green innovation is significant, as it includes enhancing the enthusiasm and initiative of small and medium-sized enterprises participating in green innovation. Thus, eliminating the financial barriers of small and medium-sized enterprises in the process of green innovation is vital [10].

Digital inclusive finance based on cutting-edge technologies such as big data, the Internet of Things, and blockchain can improve the efficiency of loan approval, lower the threshold for customer access, and provide new financing channels for SMEs to carry out green innovation activities. Digital inclusive finance breaks through the restrictions of traditional financial institutions in information transmission and regional coverage, effectively alleviates the information barriers between supply and demand of funds, significantly reduces the threshold and cost of SMEs to obtain financial services [11,12,13], optimizes the allocation efficiency of borrowing funds, and provides a strong impetus for SMEs to promote green innovation strategies. This would help increase the optimization and balance of the main structure of China’s green innovation.

Recent literature has also pointed to the relationship between digital financial inclusion and green innovation [14]. Digital financial inclusion has contributed significantly to enterprise innovation with improved credit accessibility, increased investment efficiency, and reduced financial risks according to Yang et al. (2025) [15]. For example, there is proof that digital financial inclusion based on technologies such as big data can effectively identify the credit status of SMEs, alleviate credit misallocation [16], and thus enhance the overall potential for green innovation of enterprises [17]. Additionally, digital lending platforms increase transparency and decrease transaction costs, reducing financial impediments to green technological advances [8,18]. Empirical research also shows that digital inclusive finance indirectly stimulates the ability of green innovation mainly through channels such as alleviating external financing constraints, increasing R&D investment, reducing the cost of debt financing, mitigating capital misallocation, improving financial efficiency, and optimizing the capital structure [5,19,20,21]. In addition, supportive policies for digital financial services have played an important role in promoting green investment. By reducing the financing costs of enterprises, these policies enable the implementation of green projects that were previously too costly or uncertain for enterprises [20,22]. Additionally, digital financial inclusion and environmental policy interaction further strengthen the enterprises’ motivation to engage in green innovation by offering fiscal incentives and sharing risks for developing green technologies. Based on the work of Wang et al. (2022) and Park et al. (2017), it can be said that governments can implement some supporting policies: preferential interest rates for green loans, insurance-backed guarantees for eco-friendly projects, and grants supporting research and development in sustainable technologies. Inclusion in digital finance enhances the potency of these policies through efficient disbursement, increased transparency, and reduced administrative costs [23,24]. Additionally, fintech solutions, such as AI-powered credit assessments and blockchain-based smart contracts, make green financing more efficient and highly secure, ensuring that capital reaches the most impactful projects.

This paper uses Chinese small and medium-sized board-listed companies as an example to explore, through theoretical and empirical methods, the extent of digital inclusive finance’s impact on the quality and quantity of green innovation and its mechanism. This research adds ESG and equity concentration as moderating variables, enriching the existing literature. Meanwhile, the research results also provide theoretical and empirical support for Chinese SMEs to carry out green innovation activities and are of great value and significance for achieving green transformation and upgrading.

This article consists of six sections. Section 1 introduces the background, significance, and objectives of the research, emphasizing the role of digital inclusive finance in green innovation. In the second section, based on theoretical analysis, corresponding research hypotheses are proposed. Section 3 presents the variables, and data sources of the paper, and constructs the empirical model. Section 4 provides the empirical results of the benchmark regression and the robustness test. Section 5 further analyzes the empirical results of the mediating effect and the moderating effect. Section 6 summarizes the main points of the paper, points out the limitations of the research, and suggests future research directions.

2. Theoretical Analysis and Research Hypothesis

2.1. Digital Inclusive Finance and Green Innovation

The development of digital inclusive finance helps to improve the breadth and depth of financial services, effectively make up for the shortcomings of the traditional financial model, optimize the allocation of financial resources, and improve the financing environment for enterprises, thus providing more financial support for enterprises to carry out green technological innovation [25]. First, digital inclusive finance broadens the financing channels of SMEs and solves the financing problems of SMEs carrying out green innovation [26]. Digital inclusive finance can absorb investors’ idle funds at a lower cost and transform them into financial supply for green innovation research and development [27], providing a source of funds for SMEs’ green innovation through digital platforms and enriching SMEs’ financing channels. Secondly, digital inclusive finance reduces the transaction costs of financial institutions, effectively eases the matching problems between the supply and demand sides of funds, and makes it possible for funds to flow to green innovation projects [28]. In the face of SMEs with incomplete information disclosure, small assets, and limited collateralized security value, financial institutions need to spend high search costs to identify whether enterprises meet the loan conditions [29], while digital inclusive finance can help financial institutions to identify the information of capital demanders in the mass of information, improve the efficiency of financial services, reduce the cost of searching, and improve the stable financial support for SMEs’ R&D activities. This in turn promotes SMEs to carry out green innovation, improve the quality and increase the quantity of innovation [30]. This paper proposes Hypothesis 1:

Hypothesis 1. 

Digital financial inclusion has a positive effect on the quality and quantity of SMEs’ green innovation.

2.2. Mechanisms for the Impact of Digital Financial Inclusion on Green Innovation

Digital inclusive finance, with the help of big data, artificial intelligence and other technical means, can accurately mine the information of the demand side of funds, alleviate the problem of information asymmetry between the borrower and the lender, optimize the efficiency and accessibility of financial services, and promote the inflow of funds into green research and development projects [31,32]. The long investment cycle and high uncertainty of green innovation projects make it difficult for traditional financial institutions to form an effective supervision of the enterprise development process [33]. At the same time, the low information transparency of SMEs and the inability of investors to judge the R&D and innovation ability of enterprises have led to a serious information asymmetry between green innovation subjects and financial market suppliers [34]. Under its efficient information screening and risk assessment capabilities, Digital P&F improves the information transparency of fund demanders, reduces the information asymmetry that may occur between SMEs and financial institutions in the borrowing and lending process, overcomes the problem of financing constraints, and ensures that the funds flow to enterprises with high innovation potential, thus stimulating SMEs’ green innovation capability, improving the quality and increasing the quantity of innovation [35,36]. Therefore, this study proposes the second hypothesis:

Hypothesis 2. 

Digital inclusive finance promotes the quality and quantity of green innovation in SMEs by alleviating financing constraints.

From the perspective of capital investment in green innovation, digital inclusive finance plays an important role in promoting R&D investment in SMEs [2]. The precondition for SMEs to improve their green innovation capacity is the sufficiency and stability of R&D inputs, such as stable resources, technologies, and talents. Digital inclusive finance enhances financial institutions’ pre- and post-supervision and management of innovation subjects, increases the default cost of enterprises, reduces the moral risk behavior of enterprises’ “free-riding”, and improves the financial environment [37]. As Wei et al. (2024) argue, digital inclusive finance breaks down the information wall between borrowers and lenders through the integration of digital technology, simplifies the loan approval process, and increases the opportunities for SMEs to raise funds [38]. This will be conducive to those enterprises that wish to carry out green innovation activities to reduce the financing threshold and financing costs, and obtain stable financial support, thus promoting the green innovation main body to increase the resources, technologies and talents needed for R & D investment, to improve the quality of green innovation and increase the number of innovations [39]. Based on this, Hypothesis 3 is proposed.

Hypothesis 3. 

Digital inclusive finance promotes the quality and quantity of green innovation through increased investment in R&D.

2.3. The Moderating Effect of ESG Rating and Ownership Concentration

The more outstanding an enterprise is in terms of ESG, the more it tends to focus on the pursuit of long-term value, rather than just focusing on the short-term gains and losses of managers. Therefore, enterprises with high ESG ratings actively respond to sustainable development strategies, accelerate the R&D and innovation process, and strengthen the output efficiency and effectiveness of technological innovation [40]. With higher ESG ratings, firms enjoy more favorable financing conditions, which reduces their financing costs and also improves the firms’ ability and motivation to implement green innovations [41]. Yang and Hui (2024) argued that ESG ratings moderated the relationship between digital inclusive finance and green innovations, and enhanced the quality and quantity of green innovations. This is because firms with better ESG performance benefit more from digital financial services because their responsible business practices attract capital, optimize resource allocation, and promote sustainable technological progress [22]. Based on this, Hypothesis 4 is proposed.

Hypothesis 4. 

ESG positively moderates the relationship between digital inclusive finance and corporate green innovation.

Ownership concentration is considered to be pivotal in corporate decision-making and investment strategy. The decision-making process within companies with higher ownership concentration goes more smoothly, as large shareholders influence corporation strategies and resource allocation more significantly. Ma (2023) states that this governance structure lowers coordination costs and ensures financial and managerial commitment to green innovation projects over time [42]. Because major shareholders directly benefit from the firm’s long-term success, they are more apt to pay attention to sustainable development and green innovation. Some key strengths of concentrated ownership are that it may reinforce corporate accountability and increase investor oversight. According to Macchiavello and Siri (2022), firms with higher ownership concentration are likely to accept green innovation since the shareholders ensure resource utilization is carried out efficiently and in a way that demonstrates a strategic fit with environmental purposes [43]. Ownership concentration improves the efficiency of digital inclusive finance in green innovation. While digital financial tools enhance capital access, their effects are significantly magnified if a firm has a governance structure that assures effective allocation of such capital. Large shareholders, oriented towards long-term value creation, can use digital financial services to channel investments towards environmentally sustainable projects and ensure that funds are utilized efficiently for R&D and technological development. Empirical analysis indicates that, with a more concentrated ownership structure background, firms would favor innovation projects with long-term environmental and financial benefits [42,44]. By exercising control over corporate strategies, major shareholders enable better congruence of digital financial inclusion with sustainable business objectives, and such congruence leads incrementally to enhancing the quality and effectiveness of green innovation. Based on this, Hypothesis 5 is proposed.

Hypothesis 5. 

Equity concentration enhances the positive impact of digital inclusion on the quality and quantity of firms’ green innovations.

3. Research Design

3.1. Sample Data and Sources

The sample used in this study consists of small and medium-sized board enterprises (SMEs) listed in China’s A-share market from 2011 to 2022. The SMEs were selected because they are the new force in China’s economic and social development; however, the financing constraints of SMEs restrain their innovation power and ability, so it is urgent to solve the insufficient innovation of SMEs, to provide reference value for promoting the main imbalance of green innovation in China. Data are sourced from multiple financial databases: the China Securities Markets and Accounting Research (CSMAR) database provides comprehensive enterprise-level financial data; the Wind Economy Database provides a wide range of economic and business performance data; Green patent data are provided by Chinese Research Data Services Platform (CNRDS). The data are cleaned thoroughly, including treatment for missing values by interpolation, while outliers were treated by winsorization at 1% and 99%. One of the ways to avoid bias in industry-specific effects was the exclusion of highly regulated industries, such as finance and utilities, thus providing a clean dataset to work with for digital finance and green innovation.

3.2. Variable Selection

The variables selected in this paper are based on the previous literature, ensuring empirical rigor and consistency with prior literature on green innovation and digital finance.

Table 1. Variable definitions.

Variable Type	Name	Symbol	Measurement Method	Data Sources	Relevant Literature
Explained variable	Green Innovation Quality	GRI	Ln (Number of green invention patent applications +1)	CNRDS	Wang & Wang (2021) [45] Li&Zheng (2016) [46]
	Green Innovation Quantity	GRU	Ln (Number of green utility model patent applications +1)	CNRDS	Wang & Wang (2021) [45] Li&Zheng (2016) [46]
Explanatory variable	Digital Inclusive Finance Index	Index	Reflecting the overall level zof digital inclusive finance, taking the logarithm of the index	Peking University Digital Finance Research Center	Du et al. (2024) [8]
	Digitalization level of digital inclusive finance	Digital	Reflecting the digitalization level of digital inclusive finance, taking away the logarithm		Han & Zhang (2023) [47]
	Depth of use of digital inclusive finance	Usage	Reflecting the depth of use of digital inclusive finance, taking away the logarithm		Du et al. (2024) [8] Tang et al. (2023) [10]
Intermediary variable	Financing constraints	SA	SA = −0.737Size + 0.043Size2 − 0.04Age	CSMAR	Cheng et al. (2024) [21]
	R&D investment	RD	The proportion of R&D personnel to the total number of employees	CSMAR	Shao & Chen (2023) [48] Ma & Yu (2020) [49]
Moderator variable	ESG rating	ESG	Huazheng ESG Rating	WIND	Yu et al. (2024) [50] Yang & Hui (2024) [22]
	Ownership concentration	CR1	The shareholding ratio of the largest shareholder	CSMAR	Zhao et al. (2024) [44]
Control variables	Enterprise age	Lnage	Observation year—establishment time +1 and then take away the logarithm	CSMAR WIND	Lu et al. (2024) [51]
	Return on total assets	ROA	Net profit/average total assets	CSMAR	Wang et al. (2024) [17]
	Net profit growth rate	JLR	(Profit for the current year—Profit for the previous year)/Profit for the previous year	CSMAR WIND	Qiao et al. (2022) [52]
	Revenue growth rate	AGR	(Current year’s income—previous year’s income)/previous year’s income	CSMAR WIND	Qiao et al. (2022) [52]
	Capital intensity	CA	Total assets/main business income	CSMAR	Qiao et al. (2022) [52]

provides the definition, measurement, corresponding studies, and data sources of variables.

3.2.1. Dependent Variables

Referring to relevant literature, this paper measures enterprises’ green innovation based on the number of green patent applications, and the level of green innovation is divided into the quality and quantity of green innovation for separate research [45,46]. Given that the number of patent applications of SMEs is zero in a certain number of years, when processing the variables, we add 1 to the number of patent applications and then take the logarithm of it.

Green Innovation Quality (GRI): The number of green invention patent applications indicates the quality of green innovation. According to relevant literature [8], green invention patents represent high-order technological innovation and reflect the level of development achieved by technological advances [45,46].

Green Innovation Quantity (GRU): The variable measures the quantity of green innovation with the number of green utility model patent applications [45,46]. Green utility models are typically innovations with lower technological content than invention patents [53].

3.2.2. Independent Variables

Digital Inclusive Finance (DIF): This is the holistic level of digital financial inclusion, measured concerning the Peking University Digital Financial Inclusion Index, which contains usage and digitization sub-indices. The overall index and its sub-components have found wide application in the studies of digital finance [18,43]. The data are drawn from the Peking University database to ensure reliability and reflect provincial-level inclusion.

3.2.3. Mediating Variables

R&D Investment (RD): R&D investment is measured by the share of R&D personnel in total employees, similar to prior research on innovation input [54]. This variable captures the impact of internal investment in innovation on greening innovation.

Financing Constraints (SAs): Financing constraints are proxied by the SA index, a standard measure used in previous studies to capture financial limitations [55]. The data come from the CSMAR database, which provides reliable financial metrics for Chinese firms.

3.2.4. Moderating Variables

ESG Rating (ESG): The ESG ratings are taken from the Huazheng ESG Rating system, with ratings graded from C to AAA. Liang et al. (2022) considered that ESG ratings might provide a generally accepted measure of the performance related to a firm’s environmental, social, and governance concerns [50,56]. ESG ratings are reclassified as numerical values from 1 to 9.

Ownership Concentration (CR1): Ownership concentration is defined as the shareholding ratio of the largest shareholder [44]. This variable describes the power of major shareholders to influence corporate decisions relevant to green innovation.

3.2.5. Control Variables

To control for confounding variables, this study follows other research [8,47,48,51,52,56] and includes the following variables:

Lnage: Enterprise Age, which is the observation year minus the year of establishment plus 1, which is then logged.

ROA: Return on Total Assets, which is a generally accepted measure of profitability.

JLR: Net Profit Growth Rate, which is the annual growth rate in net profit, computed as the difference between the current and previous year’s profit divided by the previous year’s profit.

AGR: Revenue Growth Rate, which is the annual growth rate in revenue, computed as net profit.

Capital Intensity (CA): Total assets divided by primary business income.

Table 1 gives an overview of all the variables, their definitions, relevant literature justifying their use, and their data sources.

3.3. Model Construction

3.3.1. Benchmark Regression Model

$${GRI}_{i,t}\left({GRU}_{i,t}\right)={\alpha }_0+{\alpha }_1{DIF}_{i,t}+{\alpha }_2{Controls}_{i,t}+{\mu }_i+{\gamma }_t+{\epsilon }_{i,t}$$

(1)

Among them, i and t represent enterprise individual and year, respectively, and GRI and GRU represent enterprise green innovation quality and quantity, respectively. DIF stands for digital inclusive finance. To comprehensively analyze the multi-level impact of digital inclusive finance, the level of digital inclusive finance is represented by the total index, usage, and digital, respectively. Control is the control variable, μ_i is the individual fixed effect, γ_t represents the time fixed effect, and ε_i,t represents the random error term. This model is used to test Hypothesis 1.

3.3.2. Intermediary Effect Model

To verify the mediating effect of financing constraints and R&D input in the promotion of green innovation by digital inclusive finance, the following intermediary effect model is constructed:

$${SA}_{i,t}={\beta }_0+{\beta }_1{DIF}_{i,t}+{\beta }_2{Controls}_{i,t}+{\mu }_i+{\gamma }_t+{\epsilon }_{i,t}$$

(2)

$${GRI}_{i,t}\left({GRU}_{i,t}\right)={\delta }_0+{\delta }_1{DIF}_{i,t}+{\delta }_2{SA}_{i,t}+{\delta }_3{Controls}_{i,t}+{\mu }_i+{\gamma }_t+{\epsilon }_{i,t}$$

(3)

$${RD}_{i,t}={\mathit{\partial }}_0+{\mathit{\partial }}_1{DIF}_{i,t}+{\mathit{\partial }}_2{Controls}_{i,t}+{\mu }_i+{\gamma }_t+{\epsilon }_{i,t}$$

(4)

$${GRI}_{i,t}\left({GRU}_{i,t}\right)={\vartheta }_0+{\vartheta }_1{DIF}_{i,t}+{\vartheta }_2{RD}_{i,t}+{\vartheta }_3{Controls}_{i,t}+{\mu }_i+{\gamma }_t+{\epsilon }_{i,t}$$

(5)

SA represents financing constraints, RD represents R&D investment, and the meanings of other variables are consistent with the Formula (1). Model (2) and model (3) are used to test Hypothesis 2, that is, whether financing constraints play an intermediary effect; Model (4) and model (5) are used to test Hypothesis 3, that is to test whether R&D investment plays a mediating effect.

3.3.3. Moderating Effect Model

This paper constructs the cross-term of digital inclusive finance and ESG, digital inclusive finance, and ownership concentration to explore the regulatory role played by ESG rating and ownership concentration. The model is as follows:

$${GRI}_{i,t}\left({GRU}_{i,t}\right)={\phi }_0+{\phi }_1{DIF}_{i,t}+{\phi }_2{DIF}_{i,t}\times {ESG}_{i,t}+{{\phi }_{3}ESG}_{i,t}+{\phi }_4{Controls}_{i,t}+{\mu }_i+{\gamma }_t+{\epsilon }_{i,t}$$

(6)

$${GRI}_{i,t}\left({GRU}_{i,t}\right)={\omega }_0+{\omega }_1{DIF}_{i,t}+{\omega }_2{DIF}_{i,t}\times {ESG}_{i,t}+{{\phi }_{3}ESG}_{i,t}+{\phi }_4{Controls}_{i,t}+{\mu }_i+{\gamma }_t+{\epsilon }_{i,t}$$

(7)

In the above model, ESG is the ESG rating of the enterprise, CR1 is the shareholding ratio of the largest shareholder, and the remaining variables are consistent with Formula (1). Model (6) is used to test whether ESG plays a moderating role in the process of digital financial inclusion affecting green innovation. Model (7) is used to test the moderating effect of ownership concentration.

4. Analysis of Empirical Results

4.1. Descriptive Statistical Analysis

Table 2. Descriptive statistical results.

Variable	Obs	Mean	Std. Dev.	Min	Max
GRU	8199	0.466	0.788	0	3.466
GRI	8199	0.456	0.817	0	3.761
Index	8199	2.758	1.091	0.331	4.528
Usage	8199	2.794	1.098	0.472	4.743
Digital	8199	3.251	1.216	0.157	4.641
Lnage	8199	1.829	0.702	0	2.833
ROA	8199	0.041	0.07	−0.255	0.232
JLR	8199	−0.341	3.143	−21.147	8.304
AGR	8199	0.154	0.339	−0.505	2
CA	8199	2.204	1.483	0.439	9.188
SA	8199	−3.769	0.244	−4.426	−3.245
RD	8199	0.106	0.12	0	0.59
ESG	8199	3.955	1.234	0	6
CR1	8199	0.329	0.141	0.089	0.702

shows the descriptive statistics of the main variables. Specifically, the average value of green innovation (GRU) is 0.466, the highest value is 3.466, and the lowest value is 0. In terms of green innovation quality (GRI), the mean is 0.456, the highest is 3.761, and likewise, the lowest is 0. The data reveal significant differences in the quality and quantity of green innovation among different companies. The data of the maximum and minimum values of the core explanatory variable digital financial Inclusion Composite Index (Index) are 4.528 and 0.331, respectively, indicating that there are significant differences in the development level of digital financial inclusion among regions, reflecting that the development imbalance between regions is still significant. The average value of the financing constraint (SA) is −3.769. The greater the absolute value of SA, the higher the degree of financing constraints. The average value of R&D investment (RD) for enterprises is 0.106, the minimum value is 0, and the maximum value is 0.59, indicating that there is a significant difference in the investment of SMEs in R&D innovation, and R&D investment is directly related to the process of green innovation. The average value of ESG is 3.955, the maximum value is 6, and the minimum value is 0. CR1 represents the shareholding ratio of the largest shareholder. As can be seen from Table 2, the average value is 0.329, indicating that the shareholding ratio of the largest shareholder of Chinese SMEs is relatively high at this stage, which is likely to have an important impact on the decision-making of enterprises’ green innovation activities.

4.2. Baseline Regression Analysis

Table 3. Baseline regression analysis.

	Green Innovation Quantity			Green Innovation Quality
	(1)	(2)	(3)	(4)	(5)	(6)
Index	0.415 ***			0.593 ***
	(0.093)			(0.098)
Usage		0.261 ***			0.243 ***
		(0.050)			(0.052)
Digital			0.247 ***			0.231 ***
			(0.040)			(0.042)
Lnage	0.009	0.004	0.011	−0.069 ***	−0.073 ***	−0.067 ***
	(0.025)	(0.025)	(0.025)	(0.026)	(0.026)	(0.026)
ROA	0.192	0.205	0.181	0.171	0.181	0.159
	(0.131)	(0.131)	(0.131)	(0.137)	(0.137)	(0.137)
JLR	−0.003	−0.003	−0.002	−0.001	−0.001	−0.001
	(0.002)	(0.002)	(0.002)	(0.002)	(0.002)	(0.002)
AGR	0.034 *	0.035 *	0.031	0.039 *	0.040 *	0.036 *
	(0.020)	(0.020)	(0.020)	(0.021)	(0.021)	(0.021)
CA	0.020 ***	0.019 ***	0.019 ***	0.013 *	0.012	0.012
	(0.007)	(0.007)	(0.007)	(0.007)	(0.007)	(0.007)
Constant	−0.178 ***	−0.118 **	−0.034	−0.215 ***	−0.037	0.040
	(0.064)	(0.048)	(0.037)	(0.067)	(0.051)	(0.039)
N	8199	8199	8199	8199	8199	8199
R-squared	0.172	0.172	0.174	0.119	0.117	0.118

Note: *, ** and *** represent significant coefficients at the levels of 10%, 5% and 1%, respectively, and the robust standard error is in parentheses, which is the same below.

shows the baseline regression results of the impact of digital inclusive finance on green innovation, where columns (1) to (3) show the impact of digital inclusive finance on the quantity of green innovation, and columns (4) to (6) show the impact of digital inclusive finance on the quality of green innovation. The results from column (1) to column (3) show that at the 1% confidence level, the total digital inclusive finance index, digitization degree, and depth of use have a significant positive impact on the number of green innovations of enterprises, which strongly supports the positive correlation between the development level of digital inclusive finance and the number of green innovations in the region where enterprises are located. That is, the booming development of digital financial inclusion has significantly promoted the increase in the number of green innovations in enterprises. The results in columns (4) to (6) show that the total index of digital inclusive finance, the degree of digitization, and the depth of use have a significant promoting effect on the quality of green innovation of enterprises, and they are all significant at the 1% confidence level, verifying Hypothesis 1. Digital inclusive finance has a significant promoting effect on the quantity and quality of green innovation of SMEs. The main reason is that digital inclusive finance improves the efficiency of financial services, reduces the financing cost of SMEs, provides SMEs with sufficient R&D and innovation funds, and then promotes the process of R&D and innovation of enterprises and improves the green innovation capability. To sum up, digital financial inclusion has a dual role in promoting the quality and quantity of green innovation and helps SMEs to improve the quality and quantity of green innovation, which aligns with Chen et al. (2023) and Zhu et al. (2024) [57,58].

4.3. Robustness Test

4.3.1. Delay the Accepted Variable by One Phase

Because of the gradual accumulation of digital inclusive finance to promote enterprises to carry out green innovation activities, its effect on green innovation may take a while to appear, so in this paper, the logarithm of green innovation is delayed by one stage [56].

Table 4. Robustness test results of independent variables that lag by one stage.

	Green Innovation Quality That Lags Behind by One Period			The Number of Green Innovations That Lag Behind by One Period
	L.GRI	L.GRI	L.GRI	L.GRU	L.GRU	L.GRU
Index	0.800 ***			0.807 ***
	(0.117)			(0.113)
Usage		0.282 ***			0.339 ***
		(0.066)			(0.064)
Digital			0.293 ***			0.427 ***
			(0.052)			(0.050)
Constant	−0.816 ***	−0.282 ***	−0.234 ***	−0.817 ***	−0.357 ***	−0.393 ***
	(0.140)	(0.100)	(0.068)	(0.136)	(0.097)	(0.066)
N	6343	6343	6343	6343	6343	6343
R-squared	0.122	0.117	0.119	0.165	0.162	0.169

Note: *** represents significant coefficients at the levels of 1%, and the robust standard error is in parentheses, which is the same below.

shows that even when the explained variable lags by one stage, the logarithm of green invention patents and utility model patents is delayed by one stage. Digital financial inclusion still has a significant positive effect on SMEs’ green innovation.

4.3.2. Model Replacement

In this study, the Tobit model is employed as a robustness test due to zero observations for the number of green patent applications in some years [59]. The Tobit model, first proposed by Tobin in 1958, is a widely adopted statistical estimation method when the dependent variable has censored points. This is a continuous green innovation quantity and quality, in this case, censored at zero since some firms do not have green patent applications for every year under study. This is useful when observations in the dataset cluster at a threshold, usually around zero, and if a normal linear regression model is applied, the estimates might become biased. This is achieved by explaining the censored (non-zero) and uncensored (zero) observations of the Tobit model. Therefore, more realistic estimates for the relationships of the independent and dependent variables can be derived.

In this context, the preferred model is a Tobit model that corrects for potential selection bias arising from left censoring of the green innovation outcomes. This is because zero patent applications by firms for any given year could imply zero innovation. In contrast, this may indicate that the firm has not participated in any green patent activity. This means that a lack of attention to such zeroes may result in biased results, and a relationship might exist between digital financial inclusion and green innovation. The Tobit model allows the research to consider all observations, either with zero patent applications or with non-zero values, considering the censored nature of data. The robust results of applying the Tobit model in this study are still significant, as depicted in

Table 5. Robustness test results of the replacement Tobit model.

	Green Innovation Quantity			Green Innovation Quality
	GRU	GRU	GRU	GRI	GRI	GRI
Index	0.470 ***			0.475 ***
	(0.023)			(0.026)
Usage		0.567 ***			0.531 ***
		(0.022)			(0.025)
Digital			0.336 ***			0.365 ***
			(0.021)			(0.024)
Constant	−1.984 ***	−2.269 ***	−1.783 ***	−2.223 ***	−2.398 ***	−2.102 ***
	(0.077)	(0.077)	(0.081)	(0.087)	(0.087)	(0.092)
N	8199	8199	8199	8199	8199	8199
R-squared	0.0254	0.0388	0.0150	0.0211	0.0276	0.0145

Note: *** represents significant coefficients at the levels of 1%, and the robust standard error is in parentheses, which is the same below.

, which suggests that central relationships hold after accounting for left censoring. More precisely, coefficients related to index, usage, and digital variables are statistically significant to ensure the robustness of results. That would mean the effect of digital financial inclusion on green innovation outcomes is not driven by potential biases from zero patent observations. Thus, the Tobit model confirms the reliability of the results provided with a standard regression approach.

5. Further Research

5.1. Intermediation Effect Test: Financing Constraints

Table 6. Test results of mediating effects of financing constraints.

	Green Innovation Quality			Green Innovation Quantity
	(1) GRI	(2) SA	(3) GRI	(4) GRU	(5) SA	(6) GRU
Index	0.588 ***	0.023 ***	0.566 ***	0.407 ***	0.023 ***	0.396 ***
	(0.098)	(0.008)	(0.097)	(0.093)	(0.008)	(0.093)
SA			0.965 ***			0.500 ***
			(0.143)			(0.136)
N	8198	8198	8198	8198	8198	8198
R-squared	0.635	0.972	0.638	0.644	0.972	0.645

Note: *** represents significant coefficients at the levels of 1%, and the robust standard error is in parentheses, which is the same below.

demonstrates a strong positive relationship between digital financial inclusion and green innovation through the mediating effect of financing constraints. This aligns with Xie (2024), Yao and Yang (2022), and Du et al. (2024), who found that digital inclusive finance indirectly stimulates green innovation capabilities by alleviating financing constraints [8,19,60]. From Table 6, it can be observed that the coefficients of the green innovation index with digital financial inclusion in columns 1 and 3 are observed to be significant and positive, i.e., 0.588 and 0.566, respectively. Additionally, the mediation of SA has a highly significant positive relation to the quantity of green innovation with a coefficient value of 0.965 in column (4). These confirm that financial inclusion mechanisms directly contribute to the outcome of innovation by way of better access to funds and reduced capital barriers.

Furthermore, the findings demonstrate that the firms with better digital financial support are more likely to participate in green innovation activities, which the significant coefficients in columns could justify (2) and (5). The R-squared values, ranging from 0.635 to 0.972, signify that digital financial inclusion is instrumental in reducing financing constraints. These results are in line with the literature. For example, Cheng et al. (2024) found that digital inclusive finance supports green innovation mainly through easing financing constraints, and point out that better liquidity from digital finance increases the ability of enterprises to make more effective investment decisions, hence being more innovative [21]. With these findings, expanding digital financial tools and thus ensuring greater credit access across diverse industries could engender broader stability in the economy. Further studies may well gauge the long-term effects of financial inclusions across sectors, with specific attention to the changes in the performance and sustainability of firm innovations based on digital financial policies within economic cycles.

5.2. Intermediate Effect Test: R&D Investment

Table 7. Test results of mediating effect of R&D input.

	Green Innovation Quality			Green Innovation Quantity
	(1) GRI	(2) RD	(3) GRI	(4) GRU	(5) RD	(6) GRU
Index	0.588 ***	0.061 ***	0.553 ***	0.407 ***	0.061 ***	0.393 ***
	(0.098)	(0.012)	(0.098)	(0.093)	(0.012)	(0.093)
RD			0.578 ***			0.229 **
			(0.096)			(0.091)
N	8198	8198	8198	8198	8198	8198
R-squared	0.635	0.745	0.637	0.644	0.745	0.644

Note: *** represents significant coefficients at the levels of 1%, and the robust standard error is in parentheses, which is the same below.

shows that digital inclusive finance improves the quality and quantity of green innovation by increasing R&D investment. Hypothesis 3 is tested. The findings align with Liang et al. (2023) and Li et al. (2023), who note that digital inclusive finance has increased the level of R&D investment, creating favorable conditions for the green innovation of enterprises [2,56,61]. Li and Yang (2024) further explain that digital inclusive finance facilitates corporate green innovation by increasing R&D investment, such as providing financial support for technical support and personnel training of green innovation [62].

Specifically, the findings suggest that financial accessibility is among the decisive factors in the firms’ level of R&D investment and thus their performance in green innovation. Better financial regulations and digital financial platforms offer firms the means to invest in sustainable projects for better innovation efficiency. Specifically, such lending platforms have created, in most cases, an alternative source of funds for those enterprises that often find it challenging to access traditional loans, hence fastening the speed of green R&D.

Government interventions, such as digital financial inclusion and R&D incentives, could thus facilitate broader corporate involvement in green innovation. Policymakers are recommended to develop regulatory frameworks that enable finance accessibility, especially for R&D-intensive sectors. Further, the role of digital financial tools interacting with firm-specific factors such as size and industry type in green R&D investments can be an area of further research. In addition, future research might explore how varied financing mechanisms, such as government subsidies or green bonds, complement digital finance in driving sustainable corporate innovation.

5.3. Moderating Effect Test: Ownership Concentration and ESG Rating

Table 8. Test results of adjustment effect.

	(1)	(2)	(3)	(4)	(5)	(6)
	GRI	GRI	GRI	GRU	GRU	GRU
Index	0.588 ***	0.604 ***	0.573 ***	0.407 ***	0.417 ***	0.394 ***
	(0.098)	(0.098)	(0.098)	(0.093)	(0.093)	(0.093)
Index * CR1		0.151 ***			0.098 **
		(0.044)			(0.042)
CR1		0.131			0.150
		(0.108)			(0.103)
Index * ESG			0.014 ***			0.011 **
			(0.005)			(0.005)
ESG			0.023 ***			0.030 ***
			(0.007)			(0.006)
Constant	−0.217 ***	−0.263 ***	−0.302 ***	−0.108 *	−0.163 **	−0.220 ***
	(0.061)	(0.073)	(0.066)	(0.058)	(0.070)	(0.063)
N	8199	8199	8199	8199	8199	8199
R-squared	0.117	0.118	0.120	0.170	0.171	0.174

Note: *, ** and *** represent significant coefficients at the levels of 10%, 5% and 1%, respectively, and the robust standard error is in parentheses, which is the same below.

reports the empirical results of ESG ratings and equity concentration on the moderating effect of digital inclusive finance on green innovation in enterprises. The empirical results of this paper are consistent with Yu et al. (2024) and Zhao et al. (2023) [44,50], which verify hypothesis 4 and hypothesis 5. Zhao et al. (2024) note that higher ownership concentration can coordinate the internal resources of enterprises, optimize the management process, and provide stable technology, manpower, and capital for green innovation research and development, thus improving the quality and quantity of green innovation [44]. When the ownership is highly concentrated, the interests of major shareholders are aligned with those of the company. This will drive major shareholders to attach importance to resource consumption and environmental pollution and encourage enterprises to carry out green innovation in pursuit of sustainable development [63]. The ESG concept is highly compatible with China’s new development concepts (innovation, coordination, green, openness, and sharing), and jointly advocates that enterprises should balance environmental protection and social responsibility in economic development. Research shows that ESG rating can not only motivate enterprises to develop green innovation strategies but also improve managers’ environmental awareness, thus enhancing the quantity and quality of enterprises’ green innovation achievements [50].

5.4. Heterogeneity Test: Property Rights

Table 9. Property rights heterogeneity analysis results.

	Green Innovation Quality		Green Innovation Quantity
	State-Owned Enterprise	Non State-Owned Enterprises	State-Owned Enterprise	Non State-Owned Enterprises
Index	0.658 ***	0.657 ***	0.570 ***	0.378 ***
	(0.237)	(0.112)	(0.206)	(0.109)
Lnage	−0.032	−0.059 **	−0.072	0.020
	(0.088)	(0.027)	(0.077)	(0.027)
ROA	−0.192	0.158	0.168	0.202
	(0.483)	(0.144)	(0.419)	(0.140)
JLR	−0.003	−0.000	−0.008	−0.002
	(0.006)	(0.003)	(0.005)	(0.003)
AGR	0.036	0.039 *	0.012	0.038 *
	(0.060)	(0.022)	(0.052)	(0.021)
CA	−0.053 **	0.022 ***	−0.016	0.028 ***
	(0.024)	(0.008)	(0.021)	(0.007)
Constant	0.052	−0.311 ***	−0.012	−0.194 ***
	(0.182)	(0.076)	(0.157)	(0.074)
N	1285	6914	1285	6914
R-squared	0.144	0.115	0.209	0.163

Note: *, ** and *** represent significant coefficients at the levels of 10%, 5% and 1%, respectively, and the robust standard error is in parentheses, which is the same below.

shows the benchmark regression results based on the equity heterogeneity of SMEs. In terms of green innovation quality, the impact of digital financial inclusion on state-owned enterprises and non-state-owned enterprises does not show a significant difference. However, in terms of the number of green innovations, digital financial inclusion has a stronger promoting effect on state-owned enterprises. The possible reason is that state-owned enterprises fulfill their social and economic responsibilities to a large extent, and they are more actively responding to the sustainable development concept called by the state, thus paying more attention to the development of green innovation of enterprises, and thus strengthening the positive role of digital inclusive finance in promoting their green innovation process. At the same time, state-owned enterprises have a natural advantage in the credit market. Compared with non-state-owned enterprises, state-owned enterprises are more likely to obtain government support and resources and face different pressures, which makes them accelerate the process of carrying out green innovation activities [51]. On the other hand, non-state-owned enterprises are in a relatively disadvantageous position and have difficulty in financing in the credit market. They also face higher financing constraints and may regard green innovation as a project with high investment and low return, so they lack sufficient motivation to use the funds raised for corporate green innovation. Therefore, digital inclusive finance has little positive impact on the number of green innovations.

5.5. Heterogeneity Test: Region

Due to the uneven level of economic development between regions, digital financial inclusion may have regional differences in promoting green innovation among SMEs. Therefore, this paper divides the sample into the eastern region and the central and western regions to explore their different effects. As is shown in

Table 10. Results of regional heterogeneity analysis.

	Green Innovation Quality		Green Innovation Quantity
	Midwest	East	Midwest	East
Index	0.626 ***	0.428 ***	0.230	0.629 ***
	(0.211)	(0.152)	(0.192)	(0.147)
Lnage	−0.004	−0.083 ***	0.040	0.009
	(0.050)	(0.030)	(0.045)	(0.029)
ROA	−0.180	0.228	−0.128	0.281 *
	(0.291)	(0.156)	(0.265)	(0.150)
JLR	0.011 **	−0.003	0.004	−0.004
	(0.005)	(0.003)	(0.005)	(0.003)
AGR	0.006	0.045 *	−0.004	0.046 *
	(0.038)	(0.024)	(0.034)	(0.024)
CA	−0.007	0.018 **	−0.011	0.029 ***
	(0.014)	(0.008)	(0.013)	(0.008)
Constant	−0.052	−0.152	0.038	−0.367 ***
	(0.102)	(0.107)	(0.093)	(0.104)
N	1967	6232	1967	6232
R-squared	0.074	0.132	0.123	0.189

Note: *, ** and *** represent significant coefficients at the levels of 10%, 5% and 1%, respectively, and the robust standard error is in parentheses, which is the same below.

, on the impact of green innovation quality, there is no significant difference between the role of digital financial inclusion in the central and western regions and the eastern regions. On the impact of the number of green innovations, the coefficient of digital financial inclusion in the eastern region is 0.629 and is significant at the level of 1%. In contrast, the coefficient of digital financial inclusion in the central and western regions was 0.230, which did not reach statistical significance. The results show that digital inclusive finance is more effective in promoting the green innovation of SMEs in the eastern region, which is consistent with Lu et al. (2024) and Rao et al. (2022) [30,51]. Exploring the reasons behind this, it is likely that the eastern region holds significant technological advantages. These advantages enable it to more effectively integrate and utilize various resources, which in turn promotes the agglomeration of innovative industries, thereby accelerating the implementation and development of green innovation activities [2]. Therefore, although digital inclusive finance has shown some potential in both regions, its impact is more pronounced in the eastern region.

Specifically, firms located in the eastern region, which generally enjoy higher maturity levels in financial infrastructure and economic prosperity, tend to benefit more from digital finance in terms of green innovation quality and quantity. The significant coefficient in the eastern region suggests that digital finance is more effective in supporting corporate innovation efforts where economic conditions are favorable. In contrast, while digital finance improves the quality of green innovation in the midwestern region, its effect on the quantity of green innovation is relatively weaker. This suggests that, even when digital financial services are used, firms in less developed regions are further restrained from scaling up green innovation efforts [64].

6. Conclusions and Limitations

6.1. Conclusions

This paper selects Chinese small and medium-sized board-listed companies from 2011 to 2022 as samples. By constructing the fixed effect model, intermediary effect model, and regulatory effect model, this paper deeply discusses the effect and mechanism of digital inclusive finance in promoting the quality and quantity of green innovation of enterprises.

The empirical results indicate that digital inclusive finance has a significant incentive effect on the quantity and quality of green innovation in SMEs. It mainly promotes the improvement of the quantity and quality of green innovation by alleviating financing constraints and increasing R&D investment. The results of the moderating effect prove that equity concentration and ESG play a positive moderating role in the process of digital inclusive finance influencing green innovation. In addition, the empirical analysis emphasizes the different impacts of digital financial inclusion on green innovation due to factors such as property rights nature and regional economic conditions. These research conclusions provide valuable insights into how digital finance affects the green innovation of business organizations and also offer action guidelines for financial institutions, policymakers, and corporate leaders.

Financial institutions can use digital technologies to optimize and improve financial products, actively participate in the practice of digital inclusive finance to support green innovation, expand their business areas, and open up new market spaces. Policymakers can formulate more targeted financial and industrial policies based on the nature of enterprises and the level of regional economic development, guiding financial resources towards the field of green innovation and promoting the coordinated development of digital inclusive finance and green innovation. Enterprises can concentrate their limited resources on the most promising and rewarding green innovation projects, improving resource utilization efficiency and achieving a win–win situation for economic and environmental benefits.

6.2. Limitations and Future Research

Despite the contributions of this paper, this study still has some limitations. This paper ignores some other factors that may affect green innovation, such as government incentives, corporate-level digital capabilities, and industry-specific effects. In the future, other influencing factors can be added to the analysis framework to more comprehensively evaluate the impact of digital inclusive finance on corporate green innovation. Moreover, this study selects SMEs among Chinese A-share listed companies, which may have a more significant reference value for listed SMEs, but its reference significance for non-listed SMEs is relatively limited. In the future, the scope of sample selection can be further expanded to draw more generally applicable research conclusions.