Analysis of the Impact of Digital Inclusive Finance on the Development of Green Agriculture

Abstract

Agricultural green development plays a crucial role in addressing the conflict between agricultural production and ecological environmental protection. This study aims to analyze the impact and mechanisms through which digital inclusive finance influences agricultural green development, providing empirical evidence to support the role of digital inclusive finance in promoting sustainable agricultural practices. Using panel data from 30 provinces (cities and districts) in China spanning from 2011 to 2020, this study measures regional agricultural green development levels using the entropy power method. By clarifying the theoretical mechanisms behind the relationship between digital inclusive finance and agricultural green development, this research employs panel fixed effects, mediation effects, and threshold regression models to empirically examine the impact of digital inclusive finance, specifically its coverage breadth, usage depth, and digitization level, on agricultural green development. Furthermore, this study explores the threshold effects of digital inclusive finance on agricultural green development. The findings from this study indicate that digital inclusive finance significantly contributes to agricultural green development, with a double-threshold effect observed as follows: when digital inclusive finance surpasses critical thresholds, its impact on agricultural green development intensifies. This result remains robust after conducting a series of endogeneity and robustness tests. The mechanism analysis reveals that digital inclusive finance drives agricultural green development by fostering technological innovation and increasing farmers’ income. Heterogeneity analysis further demonstrates that digital inclusive finance has a substantial impact on agricultural green development in eastern China compared to the central and western regions. Consequently, this study suggests that digital inclusive financial services should be further expanded and deepened to foster the integration of financial services with agriculture, creating a sustainable development model that promotes shared prosperity and effectively supports the green development of agriculture.

1. Introduction

Agricultural green development (GA) is a crucial initiative aimed at transforming the development model of agriculture and rural areas while addressing the issues of imbalanced and inadequate growth. It plays a key role in increasing farmers’ incomes and achieving common prosperity. Agriculture, as the cornerstone of China’s national development, holds an irreplaceable position in consolidating the overall progress of the country [1]. As a major agricultural powerhouse, China’s agricultural industry is progressing rapidly. However, for an extended period, China’s agricultural production has been marked by a high-pollution, high-consumption, and high-input model, which has intensified the conflict between agricultural production and environmental protection [2]. As a result, the green transformation of agriculture has become essential for the sustainable development of modern agriculture, and the development of green, efficient, and sustainable agriculture cannot be separated from the support of capital and technology. However, traditional financial services fail to meet the developmental needs of agriculture and rural areas due to their high risks, low returns, and limited coverage [3]. This significantly hampers both the economic progress of agriculture and rural areas and the improvement of farmers’ livelihoods. Moreover, the integration of digital financial inclusion (DFI) and digital technology reduces the barriers and financial costs for farmers to access capital, eliminating the inefficiencies inherently found in traditional financial systems, and offers new avenues for the agricultural sector to secure funding, ultimately fostering the growth of the agricultural economy [4]. Therefore, exploring the impact of digital inclusive finance on agricultural green development can help us better understand the formation and evolutionary path of agricultural green development in the context of digital finance.

Agricultural green development has evolved over time, with its meaning shifting along with changes in economic development and conceptual frameworks. From Marxist theory, which predicted that small farmers would eventually be replaced by capitalism and that farmers would become employers [5], to Schultz’s pioneering work on agricultural modernization, which emphasized that new factors of production drive the transformation of traditional agriculture [6], the concept has undergone significant transformations. Scholars have examined agriculture from various perspectives, including the shift from traditional farming practices to agricultural productivity, and more recently, to sustainable agricultural development [7,8]. Recent research on the connotations of green development in agriculture has explored various dimensions. Gao et al. [9] applied the entropy value method to assess agricultural green development across four dimensions: resource utilization, environmental friendliness, production efficiency, and green technology support. They noted that, despite steady improvements, regional disparities remain significant within the Yangtze River Economic Belt. Similarly, Jiang et al. [10] categorized green agricultural development into green production, ecological environment, economic efficiency, and resource conservation dimensions; they also used the entropy value method for measurement. Yu et al. [11] analyzed green agricultural development from the perspectives of environmental friendliness, ecological environment, economic efficiency, and resource saving, highlighting a U-shaped non-linear relationship between rural population aging and green agricultural development. Hong et al. [12] employed the SBM-ML method to measure green agricultural development, identifying the digital economy as a crucial driver of green agricultural growth. In addition, Xu et al. [13] used the SBM-GML index model to measure green agricultural development, revealing spatial spillover effects. Xu et al. [14], in their analysis of provincial data using a fixed effects model, found that environmental regulation and agricultural financial support significantly promote green agricultural development. Overall, the degree of green development varies widely across regions, and each region should strategically direct resources to rural areas according to its unique resource endowments, thereby energizing rural vitality.

By relying on digital technology, the implementation of digital inclusive finance overcomes the limitations of traditional financial systems, offering advantages such as low cost, high efficiency, and ease of accessibility [15]. This enables financial capital to serve the agricultural sector more effectively, addressing the challenges of limited and expensive financing, and thus promoting the development of agriculture and rural areas [16]. Digital inclusive finance plays a dual role: it is both a crucial source of funding for agricultural development and a key driver of agricultural resource revitalization. Currently, China’s rural development funds are insufficient to meet the demands of green agricultural development. Digital inclusive finance, with its inclusive, cost-effective, and convenient features [17], helps overcome these financial constraints, effectively integrating agricultural resources, and stimulates the vitality of green agricultural production. Some scholars argue that the digital nature of finance has transformed agricultural production models, fostering industry development and expanding financing channels for agricultural activities, thus accelerating agricultural progress [3,18]. Additionally, other studies use data at various levels to demonstrate that digital inclusive finance promotes the growth of green agriculture by overcoming information asymmetry and alleviating financial constraints [2,19]. However, some scholars present differing viewpoints, suggesting that the uneven development of digital inclusive finance may lead to financial exclusion, causing funds to flow away from agriculture to other sectors, which results in varied impacts on agricultural development [20]. Furthermore, digital financial inclusion, driven by the digital economy, could exacerbate the urban–rural divide due to the unequal adoption of new technologies [21]. The unpredictable nature of financial markets, coupled with the low-risk resilience of farmers, could lead to severe and irreparable losses if financial risks materializes [22]. Moreover, some studies indicate that digital financial inclusion may exacerbate elite capture effects, disproportionately benefiting regions with faster economic growth, higher technological capabilities, and individuals with better access to information and financial literacy [23,24].

Although the existing literature provides valuable insights into the impact of DFI on GA in depth and lays the foundation for this study, there are still some limitations. Most studies have primarily focused on the facilitative aspects of digital inclusive finance, without sufficiently exploring other potential channels of influence. Additionally, digital inclusive finance operates under a certain digital threshold, which can influence farmers’ access to financing and, consequently, affect the green development of agriculture. To address these gaps, this paper uses panel data from 30 provinces to empirically examine the following questions: Can digital inclusive finance promote green agricultural development? If so, what is the nature of the relationship between the two? What are the key channels of impact? The main contributions of this paper are as follows: First, it expands on the theoretical framework on the impact of digital inclusive finance on agricultural green development, offering a detailed analysis of the transmission pathways through technological innovation and farmers’ income. This enhances the understanding of digital inclusive finance’s applicability in promoting agricultural green development. Second, the paper identifies the non-linear relationship between digital inclusive finance and agricultural green development. Through empirical testing, it reveals a double-threshold effect, overcoming the limitations of previous studies that assumed a single linear relationship. This provides a fresh perspective on how digital inclusive finance influences agricultural green development under varying conditions.

2. Materials and Methods

2.1. Theoretical Analyses and Research Hypotheses

Digital inclusive finance creates new opportunities for promoting green agriculture by leveraging the benefits of low-cost, convenient, and inclusive digital technologies. It enhances financial access to the agricultural sector, offering financial services that support the greening of agriculture while lowering the barriers posed by high financial capital requirements. The promotion of digital financial inclusion is essential for advancing modern agriculture and rural development.

2.1.1. The Direct Impact of DFI on GA

The agricultural sector has long struggled with financing difficulties, where the lack of financial capital directly hinders the sustainable development of agricultural production activities [18]. Traditional financial services are constrained by geographical, resource, and time limitations, making it challenging to meet the financial needs of agricultural production. However, with the rise of the digital economy, digital inclusive finance overcomes the constraints of traditional financial services, optimizes the financial capital service model, and provides crucial financial support for the development of the agricultural economy [23], thereby fostering the green development of agriculture. On the one hand, DFI can provide diversified financial services to farmers and lower the threshold of financial services; farmers can also participate in financial credit and market transactions directly through mobile phones at any time [25], which increases the convenience of the service; the increase in farmers’ financial capital also provides financial support for agricultural production activities and the improvement of basic living conditions. On the other hand, the implementation of digital inclusive finance integrates the resource elements of the agricultural sector, improves the efficiency of agricultural production and land use, promotes the integration and optimization of the agricultural industry [26], and provides strong support for the realization of green agricultural development. Therefore, this study proposes Hypothesis 1.

Hypothesis 1:

DFI can contribute to GA.

Digital inclusive finance broadens the scope and deepens the reach of financial services, improving the traditional financial model [27] and facilitating agricultural green development. By leveraging digital technology platforms, digital inclusive finance has enhanced interregional economic connectivity and fostered cooperation among various economic sectors [28], thus supporting regional agricultural green development. However, the relationship between digital inclusive finance and agricultural green development is not linear. In the early stages, when the coverage and depth of digital inclusive finance are limited, both the participation and awareness of agribusinesses and farmers are relatively low. This leads to less efficient use of financial services and, consequently, a smaller impact on agricultural green development. As digital platforms improve and relevant policies are implemented, the inclusive, convenient, and low-barrier characteristics of digital inclusive finance become more apparent. This increases the recognition and reliance of agricultural producers and farmers on digital finance, encouraging more farmers to engage in the process. As a result, the utilization rate and efficient flow of funds improve [24], leading to more effective promotion of agricultural green development. Thus, the impact of digital inclusive finance on agricultural green development follows a non-linear pattern: once a certain threshold is reached, its positive influence on green agricultural development intensifies. This study proposes Hypothesis 2.

Hypothesis 2:

The impact of DFI on GA is subject to a threshold effect.

2.1.2. Indirect Impact of DFI on GA

Digital financial inclusion enhances technological innovation and fosters the green development of agriculture. Technological innovation plays a central role in advancing green agricultural development in China. Traditional agriculture, characterized by high pollution, excessive consumption, and heavy inputs, has long been constrained by factors such as limited capital, technology, and access to information, making it difficult to achieve sustainable agricultural progress [2]. However, digital inclusive finance, combined with digital technologies, addresses key challenges such as financing difficulties and information asymmetry in agriculture. It provides crucial financial support for technological innovation [29], particularly in rural areas and regions underserved by traditional financial services. This, in turn, significantly improves the efficiency of research and development (R&D) and the application of agricultural technologies, positively driving innovation in this sector. As regional innovation capabilities advance, agricultural technologies are developed and applied, further promoting the transformation of regional innovations into practical solutions [30]. These advancements improve agricultural productivity, reduce resource waste, and minimize environmental pollution. Therefore, digital inclusive finance not only meets the financial needs of agricultural stakeholders but also facilitates the R&D and timely adoption of technological innovations, thereby promoting the green development of agriculture. This study proposes Hypothesis 3.

Hypothesis 3:

DFI promotes agricultural green development by enhancing technological innovation.

Digital financial inclusion plays a key role in promoting the greening of agriculture by increasing farmers’ incomes. With its inclusive, efficient, and convenient nature, digital inclusive finance offers a range of financial services and improves farmers’ access to capital. This financial support enables farmers to scale up their agricultural operations and enhance production efficiency, which directly boosts their income [31]. Moreover, by leveraging digital platforms, digital inclusive finance addresses the information asymmetry typically found in traditional financial services. This not only provides financial assistance to farmers, helping them improve production conditions, but also allows them to engage more flexibly in market transactions [32]. Specifically, digital financial inclusion gives farmers diverse sources of capital. With increased financial resources, farmers can enhance agricultural infrastructure, adopt more efficient and environmentally friendly farming technologies, and even invest in advanced agricultural machinery and modern cultivation techniques [33]. As their income levels rise, farmers are better equipped to invest in green technologies, such as organic fertilizers and water-saving irrigation systems like sprinkler and drip irrigation, which conserve resources, protect the environment, and further promote the green development of agriculture. Thus, digital financial inclusion fosters the greening of agriculture by raising farmers’ incomes. Accordingly, this study proposes Hypothesis 4.

Hypothesis 4:

DFI can improve technological innovation and thus contribute to GA.

2.2. Research Design and Description of Variables

2.2.1. Variable Selection

(1) Explained Variables. Agricultural green development (GA) is a composite concept that is difficult to measure with a single indicator. To address this, the study draws extensively on the existing literature on agricultural green development [11,19] and constructs a three-tier evaluation system encompassing resource-saving, environmental friendliness, and output efficiency. This framework is designed with careful consideration of indicator representativeness and data availability. Additionally, the entropy value method is employed, as it effectively eliminates subjective bias and controls for variations in data scales, ensuring the measurement’s consistency, comparability, and scientific rigor. As a result, the entropy value method is used to assess the level of agricultural green development across regions, as presented in

Table 1. Indicator system for greening agriculture.

System	Index	Description
Resource savings	Cultivated land area per capita	Ratio of arable land to total population (%)
	Mechanization level	Total power of agricultural machinery (10,000 kW)
	Irrigation rate of agricultural land	Effective irrigated area (thousands of hectares)
	Electrification level	Rural electricity consumption (billion kWh)
	Cropland recovery index	Ratio of area sown under crops to area under cultivation (%)
Environmentally friendly	Pesticide loading levels	Proportion of pesticide application to total cultivated area (%)
	Plastic film load level	Ratio of agricultural plastic film use to total cultivated area (%)
	Fertilizer load levels	Ratio of fertilizer application to total cultivated area (%)
	Livestock manure resource utilization rate	Ratio of livestock manure utilization to total livestock manure generation (%)
Efficient outputs	Land productivity	Ratio of agricultural output to cultivated area (%)
	Food output rate	Total grain production as a proportion of the area sown to grain (%)
	Income level	Per capita disposable income of rural residents (yuan)
	Agro-processing outputs	Main business income of agricultural product processing enterprises above the specified size (billion yuan)

. As shown in Figure 1, the overall trend of China’s agricultural green development index is increasing during the sample period.

(2) Explanatory Variables. The digital financial inclusion index (DFI) is derived from the digital financial inclusion index published by the Institute of Digital Finance, Peking University, which measures the DFI in terms of the three dimensions of breadth of coverage (cover), depth of use (depth) and degree of digitization (digital) of DFI [34], and has a high degree of accuracy and credibility. As shown in Figure 1, China’s overall DFI is increasing over the sample period.

(3) Mediating Variables. Combined with the previous theoretical analysis, this study selects farmers’ income (income) and the level of technological innovation (tech) as mediating variables. Among them, farmers’ income is represented by the per capita net income of rural residents [35] and technological innovation level is represented by the number of domestic patents granted [36].

(4) Control Variables. To eliminate the influence of other variables on agricultural modernization, the following variables were chosen as control variables. The level of government support (govsup) is taken as the ratio of expenditure on agriculture, forestry, and water management to GDP [35]. The level of industry structure (ins) is measured by the ratio of tertiary industry value added to secondary industry value added [37]. The degree of openness to the outside world (open) is expressed by the total amount of goods imported and exported according to the location of the business unit and is converted from US dollars to RMB. Educational level (edu) is expressed in terms of the average number of years of education of the rural population [36]. Farmers’ investment in fixed assets (invest) is expressed in terms of total investment in fixed assets in rural areas [5].

2.2.2. Model Construction

(1) Baseline Model. The impact of digital inclusive finance on agricultural green development may be influenced by unobservable regional factors. The fixed-effects model, which controls time-invariant individual-specific disturbances, helps mitigate omitted variable bias, thereby enhancing the accuracy of regression results. Consequently, this study employs the fixed-effects model to analyze the relationship between digital inclusive finance and agricultural green development. The equation is as follows:

$$G{A}_{it}=a+\mathsf{\partial }DF{I}_{it}+\beta X_{it}+{\epsilon }_{it}$$

(1)

where GA_it denotes the level of agricultural green development of region i in period t; DFI_it denotes the level of DFI of region i in period t, which includes the DFI with its breadth of coverage, depth of use, and degree of digitization sub-dimensions; X_it denotes a set of control variables that may affect the level of agricultural green development; and ε_it is a random disturbance term.

(2) Mediation Effect Modeling. According to the previous study, DFI can promote the level of greening agriculture. To further explore the transmission mechanism of digital inclusive finance on green agricultural development, this study proposes to test the mechanism from the perspective of technological innovation level and farmers’ income. Based on the mechanism test method by Jiang [38], the mediation effect model was constructed for the test, as shown in Equation (2).

$$M={\phi }_0+{\phi }_{1}DF{I}_{it}+\omega X_{it}+{\epsilon }_{it}$$

(2)

In Equation (2), the meaning of the letters corresponds to Equation (1), where the explanatory variable M represents the mediating variables farmers’ income (income) and the level of technological innovation (tech), and testing the regression coefficients in Equation (2) proves the existence of mediating effects if they are significant.

(3) Panel Threshold Modeling. The relationship between digital inclusive financial services and agricultural green development may be non-linear. The threshold regression model is capable of capturing this non-linearity, as it can identify the threshold variable through model testing and establish a segmented function, which more accurately represents the non-linear dynamics between digital inclusive financial services and agricultural green development. Therefore, this study employs the panel threshold model for testing, using the digital financial inclusion index (DFI) as the threshold variable. The threshold regression model is presented as follows: Equation (3) represents the single-threshold model, while Equation (4) represents the double-threshold model.

$$G{A}_{it}=u_i+{\beta }_{1}DF{I}_{it}\ast I(DF{I}_{it}\le \gamma )+{\beta }_{2}DF{I}_{it}\ast I(DF{I}_{it}>\gamma )+{\epsilon }_{it}$$

(3)

$$G{A}_{it}=u_i+{\beta }_{1}DF{I}_{it}\ast I(DF{I}_{it}\le \gamma )+{\beta }_{2}DF{I}_{it}\ast I({\gamma }_1<DF{I}_{it}\le {\gamma }_2)+{\beta }_{3}DF{I}_{it}\ast I(DF{I}_{it}>{\gamma }_2)+{\epsilon }_{it}$$

(4)

where GA is the explanatory variable; u_it is the individual fixed effect; β is the coefficient of the threshold variable to be estimated; q_it is the threshold variable; γ is the threshold value; I(•) is the indicator function; and ε_it is the disturbance term.

2.2.3. Data Sources

This paper chooses 2011–2020 as the study period, and since the data from Hong Kong, Macao, Taiwan, and Tibet are not easy to obtain, this study uses the panel data of the remaining 30 provinces in China for the study. The data are mainly obtained from the China Statistical Yearbook, China Rural Statistical Yearbook, China Statistical Yearbook on Science and Technology, China Urban and Rural Construction Statistical Yearbook, relevant statistical yearbooks of each province, the EPS database, etc. For some of the missing data, interpolation was used to fill the gaps. To reduce data errors, logarithms were taken for non-ratio variables in the empirical analysis. Descriptive statistics for each variable are presented in

Table 2. Descriptive statistics of main variables.

Variables	Average	Standard Deviation	Min	Max
GA	0.320	0.109	0.107	0.571
DFI	5.219	0.668	2.909	6.068
cover	5.075	0.820	0.673	5.984
depth	5.201	0.648	1.911	6.192
digital	5.510	0.698	2.026	6.136
govsup	0.032	0.021	0.008	0.110
ins	1.219	0.696	0.518	5.297
open	17.253	1.568	12.280	20.428
edu	2.107	0.866	0.546	4.902
invest	5.382	1.142	0.765	6.874

.

As shown in Table 2, the average level of agricultural green development is 0.320, with a maximum value of 0.571 and a minimum value of 0.107. This suggests that the overall level of agricultural green development across the research sample is relatively low, with significant regional disparities in development. The average value of the digital financial inclusion index is 5.219, indicating a generally high level of digital financial inclusion, although substantial regional differences remain. Similarly, the mean values for the breadth of coverage, depth of use, and level of digitization within the digital inclusion index are also relatively high. However, the wide gap between the minimum and maximum values underscores the regional variation in digital inclusion and its sub-dimensions. The mean value for government support is 0.032, reflecting a low overall level of support, with notable variation across regions. The average industrial structure score is 1.219, indicating considerable differences in industrial composition between regions. The average level of openness to the outside world stands at 17.253, suggesting a generally high degree of openness, though this is not uniformly distributed across regions. The mean value for human capital is 2.107, pointing to a relatively low level of education in rural areas. Finally, the mean investment in fixed assets is 5.382, reflecting a generally positive investment level, but with substantial regional variation.

3. Results

3.1. Baseline Regression Results

Table 3. Benchmark regression results.

Variables	(1)	(2)	(3)	(4)
DFI	0.016 ***
	(0.003)
cover		0.012 ***
		(0.002)
depth			0.013 ***
			(0.003)
digital				0.010 ***
				(0.002)
govsup	−0.119	−0.163	−0.028	0.132
	(0.269)	(0.274)	(0.276)	(0.267)
ins	0.057 ***	0.061 ***	0.062 ***	0.064 ***
	(0.006)	(0.005)	(0.006)	(0.006)
open	0.016 ***	0.017 ***	0.021 ***	0.022 ***
	(0.005)	(0.005)	(0.006)	(0.005)
edu	−0.001	−0.001	−0.002	−0.002
	(0.003)	(0.003)	(0.003)	(0.003)
invest	−0.014 **	−0.014 **	−0.012 **	−0.017 ***
	(0.006)	(0.006)	(0.006)	(0.006)
_cons	−0.030	−0.031	−0.116	−0.098
	(0.091)	(0.092)	(0.090)	(0.091)

Note: ***, ** and * denote 1%, 5% and 10% significance levels, respectively, with t-statistics in parentheses, as below.

reports the results of the benchmark regression on the impact of DFI on GA. Through the Hausman test, this study uses a fixed effects model to examine the impact of DFI on GA. First, the impact of DFI on GA is shown in column (1) of Table 3. The impact of DFI on GA is significantly positive at the 1% level, indicating that DFI can significantly promote GA, which verifies Hypothesis 1. On one hand, one possible explanation is that digital inclusive finance, with its low barriers to entry and inclusive nature, has expanded the financing channels for agriculture, reduced the cost of capital access for green agricultural development, and provided essential financial support for the broader economic development of agriculture. On the other hand, DFI provides favorable support for the development of the digital economy, such as the digital financial platform represented by the flower song in Alipay, which not only provides farmers with convenient loans but also improves the financial digital skills of individual farmers, and lays the foundation for realizing the GA.

As shown in columns (2)–(4) of Table 3, the breadth of coverage, digitization, and depth of use all significantly contribute to the level of green agriculture at the 1% level in promoting agricultural greening. Among these, depth of use has the most substantial impact on green agriculture. This may be because a broader coverage of digital financial inclusion reflects its greater inclusivity. Additionally, farmers’ digital literacy directly affects how well digital inclusive finance integrates with green agricultural production. The deeper the level of use, the more farmers can leverage the inclusive and convenient features of digital inclusive finance. Moreover, the digitization of platforms has lowered the financial barriers, allowing farmers to access agricultural funds more easily, thereby increasing agricultural capital availability, stimulating the agricultural economy, and ultimately promoting the green development of agriculture.

3.2. Robustness Tests

3.2.1. Endogenous Problem Solving

Digital financial inclusion has a significant impact on the green development of agriculture. However, it is important to acknowledge that as the level of green agricultural development increases, the application of digital financial inclusion also deepens, creating a mutual causal relationship. This dynamic could lead to endogeneity issues that may affect the regression results. In previous studies, scholars have employed various instrumental variables to address such endogeneity, including the number of landline telephones and postal services per capita in 1984 [39], the distance from the sample to Hangzhou City, and the one-period lag of digital financial inclusion [40]. Drawing on the research methods of Yang and Qiao [36] and Sun et al. [41], this paper uses the one-period lag of the regional financial inclusion index as an instrumental variable to mitigate endogeneity concerns, as explained in

Table 4. Endogeneity test results.

Variables	(1)	(2)	(3)	(4)
L.DFI	0.012 ***
	(0.003)
L.cover		0.009 ***
		(0.002)
L.depth			0.011 ***
			(0.003)
L.digital				0.008 ***
				(0.002)
Control variables	Yes	Yes	Yes	Yes
_cons	−0.023	−0.032	−0.052	−0.055
	(0.094)	(0.095)	(0.094)	(0.095)

. The results indicate that the coefficients of digital financial inclusion and its sub-dimensions, with a one-period lag, remain significantly positive at the 1% level, confirming the robustness of the previous findings.

3.2.2. Other Robustness Tests

To further verify the robustness of our research findings, this paper follows the method proposed by Sun et al. [41] to re-quantify digital financial inclusion. Specifically, this study uses the digital financial inclusion index and its sub-dimensions, each scaled by a factor of 100, as new core explanatory variables in the regression analysis. The results, presented in

Table 5. Endogeneity test results.

Variables	(1)	(2)	(3)	(4)
DFI divided by 100	0.020 ***
	(0.002)
cover divided by 100		0.022 ***
		(0.002)
depth divided by 100			0.016 ***
			(0.002)
digital divided by 100				0.009 ***
				(0.002)
Control variables	Yes	Yes	Yes	Yes
_cons	0.157 *	0.177 **	0.023	0.024
	(0.091)	(0.089)	(0.089)	(0.094)

, show that the impact of both the digital financial inclusion index and its sub-dimensions on agricultural green development remains significantly positive. This is consistent with the previous regression results, further confirming the reliability of our findings.

3.3. Heterogeneity Analysis

Due to significant differences in culture, economic development, and the level of digital financial inclusion across regions, the impact of digital inclusive finance on agricultural green development may vary. To further examine this heterogeneity, this paper divides the sample into three regional groups—east, central, and west—based on the classification from the National Bureau of Statistics of China, and conducts separate regressions for each. The results, presented in

Table 6. Heterogeneity analysis.

Variables	(1) Eastern Region	(2) Central Region	(3) Western Region
DFI	0.017 **	0.012 ***	0.008 *
	(2.33)	(4.45)	(1.83)
govsup	−4.653 ***	−0.874 ***	−0.202
	(−3.70)	(−3.40)	(−0.35)
ins	0.038 ***	0.077 ***	0.029 **
	(3.47)	(10.29)	(2.28)
open	0.047 *	0.018 ***	−0.030 **
	(1.79)	(4.68)	(−2.61)
edu	0.026	−0.006 ***	0.149 ***
	(1.38)	(−2.90)	(5.84)
invest	0.006	−0.014	0.011
	(0.61)	(−1.48)	(1.05)
_cons	−0.720	−0.053	0.468 **
	(−1.54)	(−0.80)	(2.47)

, show that in all three regions, digital inclusive finance significantly promotes agricultural green development, though its impact diminishes as one moves from east to west.

This pattern can be attributed to several factors. The eastern region, being more economically developed, has a more advanced financial market, better digital infrastructure, and a more favorable environmental policy. Additionally, farmers in this region tend to have higher financial literacy, which greatly enhances the adoption and coverage of digital inclusive finance, leading to the strongest impact on agricultural green development. In contrast, while the central region has a moderate level of economic development, the western region lags behind in both economic progress and the development of digital infrastructure and financial services. Moreover, farmers’ financial literacy is lower in these regions, and there are gaps in policy implementation and market conditions. As a result, the impact of digital inclusive finance on agricultural green development is weaker in these areas.

3.4. Mechanism Analysis

According to the previous theoretical analysis, digital inclusive finance affects agricultural green development by increasing technological innovation and farmers’ incomes, so this study uses the mediation effect model for further analysis, and the regression results are shown in

Table 7. Analysis of the mediation effect results.

Variables	(1) FI	(2) TI
DFI	0.139 ***	0.181 ***
	(0.010)	(0.043)
Control variables	Yes	Yes
_cons	6.762 ***	6.082 ***
	(1.293)	(0.375)

. Among them, column (1) is the effect of digital inclusive finance on technological innovation, and it can be seen that digital inclusive finance significantly improves the level of technological innovation at the 1% level, which in turn promotes agricultural green development, verifying Hypothesis 3. A possible explanation is that digital inclusive finance, with digital technology as the core carrier, effectively makes up for the lack of funds in the process of scientific and technological innovation and research and development by providing inclusive financial services, which opens up diversified financing channels for scientific and technological innovation, has opened up diversified financing channels and injected strong impetus. And high innovation capacity not only provides solid technological support for agricultural production and agricultural modernization but also promotes deep integration within the agricultural sector and significantly improves the efficiency of resource allocation. This process helps to reduce the inefficiency of agricultural production and reduce information asymmetry, ultimately promoting the overall development of agriculture and guiding it to a green and sustainable direction. Column (2) shows the effect of digital inclusive finance on farmers’ income, and it can be seen that digital inclusive finance significantly increases farmers’ incomes at the 1% level, which in turn promotes agricultural green development, verifying Hypothesis 4. The reason may be that digital inclusive finance helps farmers solve the problem of lack of funds by providing diversified financial products and services, which reduces the threshold and cost of farmers’ access to funds to some extent. Farmers can then use the acquired funds to expand production, purchase production materials, etc., which improves the scale and efficiency of farmers’ production and increases their incomes. In addition, as their incomes increase, farmers will have more capital to increase the scale and efficiency of agriculture and to invest in green agricultural technologies, thus promoting the development of green agriculture.

3.5. Threshold Effect Test

To explore the non-linear effect of digital inclusive finance on agricultural green development, this study uses a panel threshold regression model for further analysis, and the regression results are shown in

Table 8. Threshold effect test.

Test Parameter	Threshold	F Value	p Value	The Critical Value
				10%	5%	1%
DFI	Threshold1	59.080	0.003	18.721	22.169	32.822
	Threshold2	22.030	0.050	17.244	21.153	26.251
	Threshold3	11.330	0.720	36.293	46.286	57.870

. It can be seen that the digital inclusive finance index has a double-threshold effect on agricultural green development, which verifies Hypothesis 2.

The results of the threshold regression are presented in

Table 9. Threshold regression results.

Threshold Value	Coefficient
≤5.722	0.017 ***
	(0.002)
5.722–5.882	0.021 ***
	(0.003)
>5.882	0.025 ***
	(0.003)
_cons	0.097
	(0.082)

. The results show that the digital inclusive finance index has a significant positive impact on agricultural green development in both phases, and its facilitating effect is enhanced when the DFI exceeds the threshold. That is, with the promotion and application of digital inclusive finance, the scope and scale of its services continue to expand, so that financial resources can be allocated to rural areas more efficiently, benefiting more farmers and promoting agricultural development. And when the digital inclusive finance index crosses the threshold value, the scale effect will gradually appear, and the efficiency of financial resource allocation and the level of marketisation will gradually increase, improving production efficiency and thus promoting the green development of agriculture.

4. Discussion

Previous studies have mainly focused on the level and drivers of agricultural green development in different regions [9,19], with few examining the relationship between digital inclusive finance and agricultural green development [2]. This study departs from previous research [25] by analyzing the impact of digital inclusive finance on agricultural green development across three dimensions: the breadth of coverage, depth of use, and level of digitization. Our results suggest that the depth of use has the most significant impact on agricultural green development. In contrast, Guo et al. [18] concluded that the breadth of coverage of digital inclusive finance has the greatest impact. The discrepancies between these findings may be due to differences in time periods and research samples, which affect the extent of adoption and use of digital inclusive finance in different regions. In addition, differences in the evaluation frameworks used to measure green agriculture may also contribute to the different results.

This study found that digital inclusive finance contributes to green agricultural development by increasing technological innovation and farmers’ income, which is consistent with previous studies. For example, the study by Qian et al. [42] showed that digital financial development promotes technological innovation. Further analysis by Guo et al. [31] and others found that digital inclusive finance can promote agricultural technological innovation by easing financing constraints, which in turn improves the level of agricultural development. The findings of Yin et al. [32] are similar to those of this paper, which pointed out that the development of financial science and technology can increase the income of agricultural households and thus improve the well-being of rural households. At the same time, this study further analyses the non-linear relationship between digital financial inclusion and agricultural green development based on the previous study [12], which found that digital inclusive finance has a double-threshold effect on agricultural green development. When the digital inclusive finance index crosses the threshold, the impact of digital inclusive finance on agricultural green development gradually increases, enriching the linear relationship between the two [14].

Despite the value of the current findings, there are still some limitations that require further research. First, there is still relatively little data on agricultural green development, and the measurement methods are not universally agreed upon and can only be explored in conjunction with previous relevant studies, so there may be some variability in the level of agricultural green development. In the future, we will continue to explore the specific values and key elements of agricultural green development. Second, considering the availability of data, the sample of this study uses data at the provincial level in China, and there are differences in resource endowment and economic conditions among regions, so the results of this study may not be the same as those at other levels, and further studies should extrapolate from the results of this study according to local conditions. In the future, we will further reduce the scale of this study to explore the impact of digital inclusive finance on the agricultural green development.

5. Conclusions

China’s agricultural green production suffers from the problems of unbalanced development, low production efficiency, and lack of capital chain, which not only restricts the improvement of agricultural competitiveness but also affects the development of GA. With the rapid development of digital economy, the development of DFI is used to stimulate the vitality of rural areas, which enables agricultural production to achieve deep development from quantitative to qualitative changes. This study empirically analyses the impact and mechanism of DFI on green development of agriculture using panel data of 30 provinces from 2011 to 2020, using the fixed effect model, the mediation effect model, and the threshold effect model, and draws the conclusions listed below.

First, DFI has a facilitating effect on GA; and BC, DU, and DD all have a positive effect on GA, with DD having the most obvious facilitating effect. In addition, DFI, BC, DU, and DD all have a threshold effect on GA. Second, DFI not only directly promotes GA but also effectively promotes GA through the three channels of increasing IFA, improving FI, and promoting TI. Third, the impact of DFI on GA is regionally heterogeneous, with the driving effect being more significant for the eastern region than for the central and western regions.

Based on these findings, this study makes the following policy recommendations for the early realization of GA by fully exploiting the advantages of DFI.

First, vigorously promote the role of DFI in promoting GA. DFI is a new idea for the development of agricultural modernization, and efforts should be made to promote the breadth and depth of DFI’s services to the agricultural sector, make full use of the characteristics of DD’s strong penetration, actively build digital technology facilities in rural areas, explore financial service models that suit the characteristics of rural agriculture and farm households, and promote the development of the agricultural economy and the modernization process in accordance with local conditions.

Second, tap the facilitating role of DFI, BC, DU, and DD for GA. At the same time, strengthen the financial support policy for agriculture, promote the level of investment in rural fixed assets, focus on supporting the level of technological innovation in public financial inputs, accelerate the cultivation of high and new technologies adapted to the development of agriculture and rural areas, and provide a focus point to promote GA.

Third, communication and cooperation should be strengthened in various regions to promote the development of DFI. As there are differences in the level of agricultural development and economic growth between the eastern and central and western regions, each region should promote GA based on the full use of DFI technology. At the same time, the central and western regions should actively learn from the experiences of the eastern regions, strengthen communication and cooperation, and build an integrated digital inclusive finance platform with shared resources, ultimately promoting GA.