Empowering Forestry Management and Farmers’ Income Growth Through the Digital Economy—Empirical Evidence from Guizhou Province, China

Abstract

Facilitating the sustained and stable growth of farmers’ income is crucial for achieving sustainable development in forest regions. As an emerging driving force, the digital economy has demonstrated substantial potential in enhancing farmers’ income and promoting regional economic prosperity in forest areas. Based on survey data from 1043 households across 10 counties in Guizhou Province, China, this study empirically examined the direct and indirect effects of digital economy participation on income growth among farmers in forest regions. The findings revealed that, first, participation in the digital economy significantly contributed to income growth for these households. This effect remained robust across various estimation methods, restricted sample tests, and when replacing dependent variables. Second, forestry management and its diversification played a mediating role in the relationship between digital economy participation and farmers’ income. Participation in the digital economy indirectly influenced income growth by fostering forestry management activities and their diversification. Third, the heterogeneity analysis indicated that digital economy participation had a significant positive impact on the income growth of pure farming households, part-time farming households, and households that had previously escaped poverty. This discovery underscored the unique role of the digital economy in alleviating poverty and preventing its recurrence. The conclusions of this study provide essential theoretical and practical guidance for empowering forestry development through the digital economy and advancing the digital transformation of the forestry industry. More critically, this research presents a novel pathway for the deep integration of the digital economy with forestry, jointly fostering income growth for farmers in forest regions, which holds significant implications for achieving rural sustainable development.

1. Introduction

The digital economy, characterized by advanced digital technologies such as big data, artificial intelligence, and cloud computing, represents a new economic paradigm. Its development is widely regarded as a key driver for global economic restructuring and enhancing international competitiveness [1]. Over the years, China has maintained the second-largest digital economy globally, with a growth rate significantly surpassing the average levels of developed countries, highlighting its robust developmental momentum [2]. Simultaneously, the Chinese government has placed significant emphasis on the construction of digital villages, issuing a series of policy documents to support this initiative. However, the overall level of rural digitalization in China remains in its early stages. This is particularly evident in remote and economically disadvantaged forest and mountainous regions, where farmers often lack the necessary skills to effectively utilize digital technologies. As a result, the development of the rural digital economy has yet to substantially improve the livelihoods of most farmers in these areas. Therefore, exploring whether the digital economy can effectively promote income growth for farmers in forest regions, and analyzing the specific mechanisms through which it integrates with forestry, holds profound significance for enhancing farmers’ income levels and advancing the sustainable development of forest regions.

For a long time, increasing farmers’ income has been a core issue of great concern in the academic community. With the growing prominence of the digital economy in the national economy, scholars have increasingly focused on analyzing the potential impact of the digital economy on farmers’ income and its underlying mechanisms. Most researchers argue that the development of the digital economy has a significant and positive impact on rural revitalization and income growth for farmers [3]. Specifically, the digital economy has effectively raised farmers’ income levels by stimulating entrepreneurial vitality and expanding non-agricultural employment opportunities [4,5]. The widespread adoption of internet technology, the emergence of e-commerce platforms, and the development of digital financial services have fostered the integration of the rural digital economy with the real economy, providing new opportunities for non-agricultural employment and entrepreneurship for farmers [6]. This convergence, in turn, has contributed to increased household income in rural areas [7]. Some studies, approaching the issue from the perspective of agricultural production and its input factors, suggest that the digital economy has enhanced agricultural productivity by reducing production costs. Leveraging the internet, farmers can quickly access product and market information, allowing them to effectively adjust resource allocation, thereby improving management efficiency and lowering operational costs in agricultural production [8,9]. Simultaneously, the digital economy has driven the transformation of traditional agricultural production methods, enhancing total factor productivity and significantly improving economic returns [10,11]. Other research has delved into the heterogeneous effects of the digital economy on farmers’ income, highlighting significant individual and spatial disparities in its impact [12,13]. However, some scholars hold differing views, arguing that the development of the digital economy reduces the dispersion of agricultural product prices and diminishes producer surplus. Simultaneously, it weakens the competitiveness of the seller’s market while strengthening the bargaining power of buyers. This, in turn, may limit the potential of the digital economy to directly boost rural income growth [14]. In summary, the current body of research on the impact of the rural digital economy on farmers’ income remains inconclusive.

Forestry, which serves as a key industry with economic, social, and ecological benefits, has played a significant role in driving the green transformation of rural economies and enhancing farmers’ green income [15]. Engaging in forestry activities not only improves farmers’ economic conditions but also promotes the conservation of natural resources and ecosystems, thereby enhancing the stability and sustainability of rural economies [16]. According to estimates by the National Forestry and Grassland Administration of China, forestry has become one of the primary income sources for impoverished populations in mountainous regions, contributing more than 50% to the income of specific forestry-focused areas. In this context, the deep integration of the digital economy and forestry has emerged as a critical driving force for the green transformation and high-quality development of the forestry industry [10]. The rural digital economy has not only opened new pathways for forestry development but also significantly revitalized the economies of forest and mountainous regions. For instance, the digital economy has empowered the entire forestry production and operation process through applications such as mobile 5G forest ranger technologies, the establishment of forestry e-commerce platforms, and digital banking services for forestry [17]. Additionally, research has revealed that while the digital economy promotes high-quality development in forestry, it also generates significant positive spatial spillover effects, positively influencing the economic development of surrounding areas [18]. However, most current studies have focused on the macro-level impact of the digital economy in enabling forestry industry development. At the micro-level, there remains a lack of in-depth exploration of how the digital economy specifically affects forestry management and household income.

In summary, existing literature has analyzed the impact of various specific aspects of the digital economy on farmers’ income growth. However, most of these studies have failed to comprehensively integrate different segments of the agricultural and forestry production, sales, financial, and information service value chains, thereby neglecting a systematic examination of the complex relationship between digital economy participation, its modes of engagement, and farmers’ income growth. Notably, while forestry management serves as a significant income source for farmers in forest regions, empirical analyses on how the digital economy empowers forestry management and subsequently promotes income growth for these farmers remain scarce. Furthermore, exploring the heterogeneous effects of digital economy participation on income growth across different farmer groups constitutes a critical research topic. This research holds substantial value for informing the formulation and implementation of more targeted and effective digital economy development strategies in China.

Guizhou Province, as China’s first comprehensive big data pilot zone and one of the key collective forest areas in southern China, holds unique research value. Accordingly, this study selects Guizhou Province as a representative research area and collects data from 1043 forest households. Using the Ordinary Least Squares (OLS) method, a baseline regression model is constructed to empirically analyze the impact of digital economy participation and its three forms—digitalized production, digitalized sales, and digitalized services—on the income of forest households. Furthermore, by constructing a mediation effect model, this study reveals the internal mechanisms through which forestry operations and diversification of operations mediate the relationship between digital economy participation and forest household income. Lastly, this study examines the heterogeneity of income growth effects among different groups of households participating in the digital economy. This research provides significant academic insight and practical guidance for advancing the development of digital rural areas and achieving sustainable income growth for rural households.

The remainder of this paper is organized as follows: Section 2 reviews the relevant literature on the relationship between the digital economy, forestry management, and farmers’ income and formulates the research hypotheses for the subsequent empirical analysis. Section 3 presents the research design, including data sources, variable definitions, and model specifications. Section 4 reports the empirical results and provides a discussion of the findings. Finally, Section 5 outlines the conclusions and policy recommendations.

2. Theoretical Analysis

2.1. The Direct Impact of Digital Economy Participation on Farmers’ Income in Forest Regions

As a new engine for national economic growth, the digital economy has also provided significant momentum for the development of forest region economies. The construction of digital villages has promoted the digital transformation of the entire agricultural and forestry production value chain, encompassing production, sales, finance, and information services. Farmers’ participation in the digital economy involves integrating information, technology, and capital into various stages of agricultural and forestry production. This process helps reduce production costs, improve efficiency, and lower the expenses associated with the distribution and marketing of agricultural and forestry products, thereby increasing farmers’ income levels [19]. Studies have shown that the development of the rural digital economy, particularly in areas such as digitalized production, digitalized supply chains, digitalized marketing, and digitalized finance, can significantly enhance farmers’ income growth [20,21].

First, the digitalization of production has deconstructed and reshaped traditional agricultural and forestry production models, significantly enhancing their efficiency. By adopting cutting-edge digital technologies such as the Internet of Things (IoT), big data, and artificial intelligence (AI), farmers have substantially improved the precision and intelligence of agricultural and forestry production, accelerating the deep integration of the digital economy with these sectors [22,23]. Second, digital marketing strategies have markedly improved the accuracy of supply–demand matching for agricultural and forestry products, boosting market transaction efficiency and ensuring effective coordination across the sales process. Leveraging big data analytics, suppliers—farmers—can open sales channels promptly, while consumers can be precisely matched with the products they need, enhancing the overall efficiency of economic activities. Studies have indicated that the rise of e-commerce and live-streaming sales has not only expanded the marketing avenues for agricultural and forestry products but also reduced intermediary distribution links and marketing costs, directly contributing to increased farmers’ income [24]. Lastly, digital services, in the dimensions of financial services and information services, have provided robust support for farmers’ income growth. Digital financial services, offering online payment, credit, investment, and other diverse financial products, have effectively supported the sustainable development of agriculture and forestry and facilitated the improvement of farmers’ income levels [25]. Additionally, digital financial services have reduced the customer search costs for financial institutions and the financing costs for farmers, while broadening farmers’ access to financing through technological innovation, alleviating financial constraints in agricultural and forestry production [26,27]. In terms of information services, digital platforms have made agricultural and forestry-related information, such as land transfer, socialized services, and skills training, more accessible and widely disseminated. The rapid circulation of such information has significantly reduced farmers’ information search costs and improved the decision-making efficiency and scientific basis of agricultural and forestry management.

Based on the above analysis, the following hypotheses are proposed:

H1: 

Participation in the digital economy positively affects the income of farmers in forest regions.

H1a: 

Participation in digital production positively affects the income of farmers in forest regions.

H1b: 

Participation in digital marketing positively affects the income of farmers in forest regions.

H1c: 

Participation in digital services positively affects the income of farmers in forest regions.

2.2. Indirect Impact of Digital Economy Participation on Farmers’ Income in Forest Regions

Participation in the digital economy not only directly contributes to the increase in farmers’ income in forest regions but also potentially exerts indirect positive effects through the promotion of forestry management and diversification. First, forestry management activities and diversification strategies are regarded as key drivers of income growth for forest households. As a critical component of natural endowment, forest resources hold immense value in alleviating poverty, improving farmers’ livelihoods, increasing forestry household income, and driving the overall socio-economic development of forest regions [28,29]. Specifically, farmers improve their household economic conditions by engaging in industries such as timber forestry, bamboo cultivation, economic forestry, under-forest economy, and forest tourism, selling timber, bamboo, forest fruits, medicinal plants, and other forestry products [30]. The emergence of new forestry management entities and the deepening of socialized forestry services have further promoted green income growth for farmers [31,32]. Additionally, diversified forestry management strategies enrich the livelihood structure of households, enhancing the diversity and stability of their income sources. In the face of market fluctuations, natural disaster risks, and other uncertainties, diversified forestry income structures provide stronger buffering mechanisms, increasing the resilience and sustainability of income growth. By implementing diversified forestry management, farmers can utilize forestry resources more efficiently, optimizing their allocation in terms of time, space, and functionality, thus improving resource utilization efficiency and increasing output value, leading to a stable rise in farmers’ income.

Furthermore, participation in the digital economy has profoundly reshaped farmers’ forestry management behaviors and diversification strategies, accelerating the digital transformation of the forestry industry and indirectly contributing to significant income growth for forest households. The integrated development of the digital economy and forestry has expanded the pathways for forestry development, influencing forestry production, management, and industry [33]. In particular, in forestry management, farmers’ participation in the digital economy significantly affected household production decisions and resource allocation efficiency [33]. Studies have shown that the digital economy provides intelligent, networked, and digitalized technological support for forestry production and management, enhancing the scientific and automated levels of forestry management. This transformation has facilitated the shift from traditional extensive operations to modernized practices, thereby improving the efficiency of forestry production and management [10,34]. From the perspective of forestry product sales, digital transformation has effectively reduced management and transaction costs, minimized efficiency losses due to information asymmetry, and enabled forestry production to more accurately match market demand. This has allowed for personalized production and diversified marketing of forestry products. Additionally, through data integration, the digital economy has enhanced the flow of forestry-related information, promoting green development and innovation within the forestry industry chain and optimizing its overall operational efficiency [35].

In summary, participation in the digital economy has indirectly contributed to farmers’ income growth by promoting forestry management and diversification. Based on this analysis, the following hypotheses are proposed:

H2: 

Participation in the digital economy indirectly increases farmers’ income in forest regions by promoting their engagement in forestry management activities.

H3: 

Participation in the digital economy indirectly increases farmers’ income in forest regions by promoting the diversification of their forestry management activities.

3. Materials and Methods

3.1. Data

This study selects Guizhou Province, China, as the research area. Located in the heart of southwest China, Guizhou lies between 103°36′ to 109°35′ E and 24°37′ to 29°13′ N, stretching approximately 595 km from east to west and 509 km from north to south, covering a total area of 176,167 square kilometers, which accounts for 1.8% of China’s total land area. Guizhou comprises six prefecture-level cities and three autonomous prefectures, with a permanent population of 38.65 million. Situated on the Yunnan–Guizhou Plateau, Guizhou’s terrain is characterized by higher elevations in the west and lower elevations in the east, sloping from the central region towards the north, east, and south, with an average altitude of around 1100 m. The province is predominantly mountainous, often described as “eight parts mountain, one part water, and one part farmland”. Its landforms can be classified into three basic types: plateau mountains, hills, and basins, with 92.5% of its area comprising mountains and hills. Guizhou boasts a forest coverage rate of 62.81%, making it the only province in China without plains [36]. Capitalizing on its rich forest resources, Guizhou has actively developed its forestry economy, forming an industrial structure centered on bamboo, tea-oil trees, Sichuan pepper, and soapberry.

As China’s first comprehensive big data pilot zone, Guizhou places significant importance on the development of the digital economy, recognizing it as a core driver of regional economic growth. In recent years, Guizhou has achieved full 5G network coverage at the county level and widespread 4G coverage in villages, while actively promoting the deep integration of the digital economy with forestry. In the production stage, Guizhou has applied digital information technology to transform its forestry industry. In the sales stage, the province has utilized e-commerce platforms, live-streaming platforms, and social media to integrate small and micro forestry enterprises, enabling scaled operations. Additionally, the creation of the e-commerce platform has increased the market visibility of forestry products and promoted the efficient allocation of resources. Overall, Guizhou’s remarkable achievements in digital economy development, combined with its unique mountainous geographical features, provide a representative case study area for this research.

In May 2023, a preliminary survey was conducted in Chishui City and Bijiang District of Guizhou Province. Based on the feedback obtained from this pilot study, adjustments and refinements were made to the household questionnaire, culminating in the final version used for the main survey. The formal research commenced in July 2023, involving a research team composed of over 40 members, including professors, lecturers, graduate students, and undergraduates. The primary respondents were household heads or key family members involved in production decision-making.

A combination of stratified sampling and random sampling methods was employed for sample selection. Stratified sampling was primarily utilized in the selection of research areas. To ensure representativeness, this study selected 1–2 counties from each of the nine prefecture-level cities in Guizhou Province and determined the number of villages to be sampled based on the population distribution within each county. Following extensive communication and discussions with personnel from the forestry department of Guizhou Province, the research team selected ten representative counties (districts/cities) for field research, considering the geographical, cultural, climatic, and forestry development characteristics of Guizhou, as well as the role of forestry in enhancing farmers’ incomes. The selected counties included Chishui City, Bijiang District, Yuping Dong Autonomous County, Jiangkou County, Jianhe County, Shuicheng District, Qianxi County, Pingtang County, Ziyun Miao and Buyi Autonomous County, and Ceheng County (Figure 1). These counties encompass a variety of forestry operations, such as bamboo forests, economic forests, forest product harvesting, understory economy, and forest tourism, each exhibiting significant representativeness.

Subsequently, a random sampling method was applied to select rural household samples from each village, with 10–15 households randomly selected per village. A total of 1054 household questionnaires were collected, of which 1043 were deemed valid, resulting in an effective response rate of 98.96%. The questionnaire covered multiple dimensions, including basic demographic information, household resource status, family production and operational circumstances, development of the understory economy, participation of new forestry business entities, and involvement in the digital economy.

To minimize potential biases arising from respondents’ misunderstandings or misinterpretations, this study conducted face-to-face interviews with adult household members who were knowledgeable and cognitively capable. Prior to the formal interview, the purpose and content of the survey were thoroughly explained to the respondents, with each interview lasting approximately one hour. Upon completion of the survey, participants received an umbrella as a token of appreciation for their involvement.

3.2. Variables

3.2.1. Dependent Variable

This study selected total household income as the dependent variable, specifically referring to the total household income of farmers in 2022, which includes operating income, wage income, property income, and transfer income. Following previous research [29], to reduce heteroscedasticity caused by large income disparities and to ensure data stability, the total household income was logged after adding 1 to the corresponding value to represent household income more appropriately.

3.2.2. Core Independent Variable

This study selected farmers’ participation in the digital economy as the core explanatory variable. Drawing from existing research [33,37] and the actual conditions observed during fieldwork in Guizhou Province, the concept of digital economy participation was defined to encompass the entire agricultural and forestry production value chain. In this study, digital economy participation includes three aspects: digitalized production, digitalized marketing, and digitalized services. If a farmer participates in at least one of these activities, the value is coded as 1; otherwise, it is coded as 0. Specifically, the survey questions related to these three aspects were designed as follows: ① Digitalized production—“Did you purchase production materials via an online platform?” If yes, the value is 1; otherwise, it is 0. ② Digitalized marketing—“Did you sell agricultural or forestry products via an online platform?” If yes, the value is 1; otherwise, it is 0. ③ Digitalized services—“Did you purchase forestry financial products via an online platform/obtain land transfer information via an online platform/search for labor opportunities via an online platform?” If yes, the value is 1; otherwise, it is 0.

3.2.3. Mediating Variables

This study selected forestry management behavior and the degree of diversification of management as mediating variables. Specifically, forestry management behavior refers to whether a household engages in forestry management activities, such as bamboo cultivation, economic forestry, forest product harvesting, under-forest economy, or forest tourism. If a household participates in any of these activities, the value is coded as 1; otherwise, it is 0. The degree of diversification in forestry management refers to the number of types of forestry management in which a household participates. This is a continuous variable, ranging from 0 to 5.

3.2.4. Control Variables

In addition to the core explanatory variable, household income is influenced by multiple factors. Based on existing research [38,39], the control variables in this study are grouped into household head characteristics, household characteristics, and village characteristics. First, household head characteristics include the head’s age, experience as a village official, and whether they have received agricultural or forestry technical training. Second, household characteristics include the number of non-agricultural workers, the education level of the labor force, the number of family members with major illnesses, whether the household has suffered from natural disasters, the amount of arable land, and the status of land transfers. Third, village characteristics include the average total household income per village and the management capacity of the village committee, which represents village-level characteristics.

3.3. Methods

This study aims to empirically examine the impact of digital economy participation on farmers’ income. First, a baseline regression model is established, followed by a robustness check using a propensity score matching (PSM) regression model. Additionally, the mediating role of forestry production and management capacity in the relationship between digital economy participation and farmers’ income is tested using a mediation effect model.

3.3.1. Baseline Regression Model

The analysis begins with an Ordinary Least Squares (OLS) model, where OLS stands for Ordinary Least Squares. Under the conditions of homoscedasticity and uncorrelated error terms, OLS provides a linear, unbiased estimate of the regression parameters with the minimum variance. The model is specified as follows:

$$Ln(Y_i+1)={\beta }_0+{\beta }_1{DE}_i+{\beta }_2{Control}_i+{\epsilon }_i$$

(1)

In Equation (1), $Y_i$ represents the total income of the $i$ household, ${DE}_i$ represents the digital economy participation of the $i$ household, ${Control}_i$ represents the control variables, ${\beta }_0$ represents the random constant term, ${\beta }_1$ and ${\beta }_2$ represent the regression coefficients of the variables to be estimated, and ${\epsilon }_i$ represents the random disturbance term.

3.3.2. Propensity Score Matching (PSM) Model

Farmers’ participation in the digital economy is a self-selection process based on expected benefits and their individual circumstances. In other words, certain factors may simultaneously influence both a farmer’s decision to participate in the digital economy and their income level. These factors could include the development of the village-level digital economy, educational coverage, and the availability of personal resources. Ignoring the potential self-selection problem and directly estimating the model may result in biased estimates. The propensity score matching (PSM) method can effectively control for the consistency of these influencing factors by constructing a “counterfactual” to estimate the impact of digital economy participation on farmers’ income [40]. The research steps of the PSM method are as follows: First, select appropriate covariates (such as whether farmers participate in the digital economy) to divide the sample into two groups. Second, use a Logit model to estimate the probability (propensity score) of farmers’ participation in the digital economy. Third, based on the propensity scores, match each farmer who participates in the digital economy with a similar non-participating farmer. Finally, calculate the Average Treatment Effect on the Treated (ATT) for farmers participating in the digital economy. This study employed three matching methods (nearest neighbor matching, radius matching, and kernel matching) and compared the results. If the results are consistent across the different matching methods, the estimates can be considered robust. The specific expression is as follows:

$$\mathrm{ATT}=E\left(Y_1-Y_0|D=1\right)$$

(2)

In Equation (2), $Y_1$ represents the total household income of farmers who participate in the digital economy, $Y_0$ represents the total household income of farmers who do not participate in the digital economy, $D=1$ indicates farmers’ participation in the digital economy, and $D=0$ represents farmers’ non-participation in the digital economy.

3.3.3. Mediating Effect Model

To further examine how participation in the digital economy influences farmers’ income by promoting forestry management behavior and the degree of diversification, a mediation effect model is constructed based on existing research [41,42] and the baseline model (1). The specific expressions for models (3) and (4) are as follows:

$$M_i={\beta }_0+{\beta }_1{DE}_i+{{\alpha }_{i}Control}_i+{\epsilon }_i$$

(3)

$$Ln(Y_i+1)={\beta }_0+{\beta }_1{DE}_i+{\beta }_2{M_i+{\alpha }_{i}Control}_i+{\epsilon }_i$$

(4)

In Equation (3), $M_i$ represents the mediating variables, including farmers’ forestry management behavior and the degree of diversification. ${\beta }_1$ denotes the effect of digital economy participation on forestry management behavior and diversification. In Equation (4), ${\beta }_1$ and ${\beta }_2$ represent the direct effects of digital economy participation, forestry management behavior, or diversification on farmers’ income. The mediation effect of forestry management capacity is the product of coefficient ${\beta }_1$ in Equation (3) and coefficient ${\beta }_2$ in Equation (4).

4. Results and Discussion

4.1. Descriptive Statistics

4.1.1. Farmers’ Participation in the Digital Economy

Out of the 1043 surveyed households, 536 households participated in the digital economy, accounting for approximately 51.39% of the respondents (

Table 1. Overview of farmers’ participation in the digital economy.

Forms of Digital Economy Participation	Specific Indicator	Number of Households	Percentage (%)
Digitalized Production	Purchasing production materials via online platforms	425	40.75
Digitalized Sales	Selling agricultural or forestry products via online platforms	120	11.51
Digitalized Services	Purchasing forestry financial products via online platforms	27	2.59
	Searching for land transfer or labor information via online platforms	253	24.26

). First, 425 households participated in digitalized production, indicating that the primary economic activity for farmers involved in the digital economy is as producers, which reflects the fundamental role of agricultural and forestry production. Second, 120 households engaged in digitalized sales, representing 28.39% of those involved in digitalized production. This suggests that the majority of farmers participating in the digital economy opt not to handle product sales themselves. On one hand, the technological threshold for adopting digital sales methods is relatively high, making it difficult for many farmers to master the necessary skills. On the other hand, small-scale farming operations often cannot meet the requirements for independently accessing platforms for direct sales. As a result, most farmers choose to sell their products to businesses or distributors. Field surveys revealed that many farmers who engage in digital sales use social media platforms like WeChat for small-scale transactions, rather than using e-commerce or “live-streaming” platforms for sales. The primary reasons for this are a lack of technology, insufficient capital, and small-scale operations. Lastly, 285 households participated in digitalized services, with only 27 using digital financial services. This indicates that while the development of the digital economy has greatly enhanced the convenience of financial services, the demand for financial services related to forestry production and management remains relatively low. Additionally, 278 households utilized digital information services, demonstrating that digital development provides farmers with abundant and valuable information related to production and management.

4.1.2. Descriptive Statistics of Variables

The descriptive statistics of the main variables are shown in

Table 2. Variable definitions and descriptive statistics.

Variable	Indicator Level	Explanation and Meaning	Mean	Standard Deviation
Dependent Variable	Household Income	Total household income in 2022 (in RMB), log-transformed after adding 1	10.378	3.315
Core Independent Variable	Digital Economy Participation	Whether the household participates in the digital economy; No = 0, Yes = 1	0.514	0.500
	Mode of Digital Economy Participation	Whether the household participates in digitalized production; No = 0, Yes = 1	0.407	0.492
		Whether the household participates in digitalized sales; No = 0, Yes = 1	0.115	0.319
		Whether the household participates in digitalized services; No = 0, Yes = 1	0.273	0.446
Mediating Variable	Forestry Management	Whether the household participates in forestry management; No = 0, Yes = 1	0.395	0.489
	Forestry Diversification	Number of types of forestry management the household participates in, continuous variable ranging from 0 to 5	0.490	0.689
Control Variable	Age	Age of the household head (years)	52.405	10.328
	Village Official Experience	Whether the household head has been elected as a village official; No = 0, Yes = 1	0.169	0.375
	Skills Training	Whether the household head has participated in agricultural and forestry technical training; No = 0, Yes = 1	0.363	0.481
	Non-agricultural Employment	Actual number of non-agricultural workers in the household	1.256	1.238
	Education Level of Labor Force	Average years of education per household laborer (years)	7.925	3.223
	Number of Serious Illnesses in Household	Number of family members with serious illness or disability	0.221	0.481
	Natural Disaster	Whether the household experienced flooding in the past year; Yes = 1, No = 0	0.198	0.399
	Cultivated Land Area	Current cultivated land area managed by the household (mu)	6.420	40.639
	Forest Land Transfer	Whether forest land was transferred in or out; Yes = 1, No = 0	0.135	0.342
	Village Committee Governance Capacity	Very poor = 1, Poor = 2, Average = 3, Good = 4, Excellent = 5	4.023	0.881
	Average Household Income in Village	Log-transformed average household income in the village after adding 1	11.411	0.654

below. The average age of farmers participating in the digital economy is 51.17 years, which is 1.24 years younger than the overall sample average of 52.41 years. The average years of education for farmers participating in the digital economy is 7.87 years, 0.66 years higher than the overall sample average of 7.21 years. The average experience of farmers serving as village officials among digital economy participants is 0.19, higher than the overall sample average of 0.17. The average participation in agricultural and forestry technical training for farmers in the digital economy is 0.40, compared to the overall sample average of 0.36. Additionally, the average cultivated land area for farmers participating in the digital economy is 7.54 mu, 1.12 mu more than the overall sample average of 6.42 mu. Overall, farmers participating in the digital economy tend to have superior livelihood capital compared to those who do not participate.

4.2. Baseline Regression Results

To ensure the consistency and unbiasedness of the regression results, this study conducted a multicollinearity test using the Variance Inflation Factor (VIF) prior to model estimation. The results showed that the VIF values for all variables were below 10, indicating no multicollinearity among the variables, allowing for further analysis. The baseline regression results are presented in

Table 3. Estimated impact of digital economy participation and its forms on farmers’ income.

Variable	Model (1)	Model (2)	Model (3)	Model (4)
Digital Economy Participation	0.494 ** (0.202)
Participation in Digitalized Production		0.410 ** (0.204)
Participation in Digitalized Sales			0.494 ** (0.202)
Participation in Digitalized Services				0.621 *** (0.223)
Household Head’s Age	0.005 (0.010)	0.004 (0.010)	0.002 (0.010)	0.004 (0.010)
Household Head’s Experience as Village Official	−0.079 (0.271)	−0.103 (0.271)	−0.049 (0.271)	−0.040 (0.271)
Household Head’s Agricultural/Forestry Technical Training	0.611 *** (0.208)	0.632 *** (0.208)	0.642 *** (0.208)	0.600 *** (0.208)
Number of Non-Agricultural Workers	0.527 *** (0.081)	0.529 *** (0.081)	0.522 *** (0.081)	0.520 *** (0.081)
Average Years of Education for Labor Force	0.036 (0.033)	0.041 (0.033)	0.041 (0.032)	0.036 (0.033)
Number of Family Members with Serious Illnesses	−0.488 ** (0.207)	−0.481 ** (0.207)	−0.490 ** (0.207)	−0.485 ** (0.207)
Experienced Natural Disasters	−0.850 *** (0.247)	−0.856 *** (0.248)	−0.801 *** (0.247)	−0.794 *** (0.247)
Cultivated Land Area	0.002 (0.002)	0.002 (0.002)	0.002 (0.002)	0.002 (0.002)
Forest Land Transfer	0.466 (0.287)	0.445 (0.288)	0.484 * (0.288)	0.496 * (0.287)
Village Committee Governance Level	−0.181 (0.113)	−0.171 (0.113)	−0.171 (0.113)	−0.162 (0.113)
Average Household Income in Village	0.568 *** (0.155)	0.572 *** (0.156)	0.595 *** (0.155)	0.594 *** (0.154)
Constant	3.172 * (1.898)	3.162 * (1.901)	3.043 (1.897)	2.875 (1.894)
N	1043	1043	1043	1043
R2	0.104	0.102	0.103	0.106

Note: ***, **, and * represent significance levels at 1%, 5%, and 10%, respectively. Values in parentheses are standard errors.

. Model (1) reflects the impact of overall digital economy participation on total household income, while Models (2) through (4) represent the effects of various dimensions of digital economy participation on total household income.

4.2.1. Impact of Digital Economy Participation on Farmers’ Income

The regression results of Model (1) show that the estimated coefficient for digital economy participation is 0.494, which is significantly positive at the 5% level. This suggests that participation in the digital economy significantly boosted farmers’ income, confirming H1. A possible explanation is that, in the context of rapid digital economy development, farmers enhance their competitiveness in economic activities and improve the efficiency of resource utilization by purchasing production materials, selling agricultural and forestry products, accessing forestry financial products, obtaining land transfer information, and searching for labor opportunities through online platforms, which in turn increases household income. Additionally, control variables such as whether the household head received agricultural or forestry technical training, the number of non-agricultural workers in the household, the number of family members with serious illnesses, whether the household was affected by natural disasters, and the average household income in the village were also significant factors influencing total household income.

4.2.2. Impact of Different Forms of Digital Economy Participation on Farmers’ Income

The regression results of Model (2) indicate that the estimated coefficient for digital production is 0.410, which is significantly positively correlated at the 5% statistical level. This finding suggests that participation in digital production significantly contributes to the increase in farmers’ income, thereby validating Hypothesis H1a. The regression results of Model (3) show that the estimated coefficient for digital sales is 0.699, also demonstrating a significant positive correlation at the 5% statistical level. This indicates that engagement in digital sales significantly enhances farmers’ income, thereby confirming Hypothesis H1b. Furthermore, the regression results of Model (4) reveal that the estimated coefficient for digital services is 0.621, which is significantly positively correlated at the 1% statistical level. This suggests that involvement in digital services significantly promotes farmers’ income growth, thus validating Hypothesis H1c.

First, digital production directly facilitates income growth for farmers by improving the management efficiency of agricultural and forestry resources, reducing production costs, and optimizing decision-making. As dedicated producers, farmers benefit from the digital economy by moving away from traditional production activities or altering conventional production methods, thereby enabling a digital upgrade of the agricultural and forestry sectors. This can involve the use of modern technologies such as the Internet of Things, satellite remote sensing, drones, and artificial intelligence for precise management, resource monitoring, and efficient operations.

Second, digital marketing transcends the limitations of traditional sales channels, expanding market reach and brand influence, which in turn enhances sales revenue. Compared to conventional sales models, digital marketing effectively reduces information asymmetry, eliminates intermediaries, and allows farmers to sell their agricultural and forestry products at market prices, thereby increasing sales volumes.

Finally, digital services provide farmers with convenient financial and informational support. Digital financial services enhance accessibility to production capital, helping farmers address funding shortages and enabling them to better cope with market fluctuations, scale up production, and stabilize income. Additionally, digital information services alleviate the degree of information asymmetry, enabling farmers to quickly access a wealth of labor information, obtain data on land transfers, and receive real-time forecasts on weather, market price fluctuations, and pest alerts. This access to critical information aids them in optimizing their planting and sales decisions.

It is noteworthy that although all dimensions of digital economy participation significantly affect farmers’ income growth, participation in digital sales had the most significant impact on household income. A possible explanation is that digital sales, as a primary form of digital economy engagement, contribute most significantly to household income growth due to its efficient information transmission, access, and favorable market matching for agricultural and forestry products.

4.3. Robustness Checks

Robustness checks were conducted from three perspectives: first, by using an alternative method, applying propensity score matching (PSM) to test the impact of digital economy participation on farmers’ income; second, by restricting the sample, removing households with heads older than 65 years and conducting empirical tests; and third, by replacing the explanatory and dependent variables to examine the effects of digital economy participation on per capita household income and the impact of diversified digital economy participation on farmers’ income.

4.3.1. Alternative Method Check

We employed three common matching methods—nearest neighbor matching, radius matching, and kernel matching—to match samples and calculate the average net effect (ATT) of farmers’ digital economy participation on household income.

Table 4. Robustness check using propensity score matching.

Matching Method	Log of Income for Households Participating in the Digital Economy	Log of Income for Households Not Participating in the Digital Economy	ATT	Standard Error	t-Value
Nearest Neighbor Matching	10.730	9.966	0.764 **	0.314	2.44
Radius Matching	10.730	10.234	0.497 **	0.221	2.24
Kernel Matching	10.730	10.234	0.496 **	0.221	2.24

Note: ** represents significance levels of 5%.

presents the average treatment effects of digital economy participation on household income based on different matching methods. The results consistently indicate that digital economy participation significantly increased total household income at the 5% level. The high degree of consistency in the estimates suggests that the findings of this study are robust and reliable, confirming that participation in the digital economy has a significant income-enhancing effect. To test the matching effectiveness, we conducted a common support assumption test. Figure 2 shows the kernel density distribution before and after one-to-one nearest neighbor matching. The post-matching density distribution of the control and treatment groups is highly similar, indicating that the model satisfies the common support assumption.

4.3.2. Using Restricted Sample

Considering the potential disadvantages that older individuals may face in participating in the digital economy, this study excluded households with heads aged 65 years or older to conduct a robustness check on the impact of digital economy participation on household income. The regression results after excluding these samples show that the effects and significance of digital economy participation and its forms on total household income are almost identical to the baseline regression results (

Table 5. Robustness check using restricted sample.

Variable	Model (1)	Model (2)	Model (3)	Model (4)
Digital Economy Participation	0.545 ** (0.218)
Participation in Digitalized Production		0.456 ** (0.220)
Participation in Digitalized Sales			0.759 ** (0.335)
Participation in Digitalized Services				0.676 *** (0.239)
Control Variable	Controlled
Constant	3.364 * (2.020)	3.336 * (2.023)	3.197 (2.018)	3.014 (2.014)
N	945	945	945	945
R2	0.101	0.099	0.100	0.103

Note: ***, **, and * represent significance levels at 1%, 5%, and 10%, respectively. Values in parentheses are standard errors.

). Therefore, we conclude that the estimation results are robust and credible.

4.3.3. Replacing Explanatory or Dependent Variables

In addition to changing the estimation method and using a restricted sample, we further replaced the explanatory and dependent variables to verify the robustness of the results. The diversity of digital economy participation, measured as the extent of farmers’ engagement in digitalized production, sales, and services (a continuous variable ranging from 0 to 3), was used to replace the original explanatory variable. Additionally, per capita household income was substituted for the original dependent variable. The estimation results are shown in

Table 6. Robustness check using variable replacement.

Variable	Replacing Explanatory Variable	Replacing Dependent Variable
	Total Household Income	Per Capita Household Income
Degree of Diversification in Digital Economy Participation	0.312 *** (0.103)
Digital Economy Participation		0.441 ** (0.178)
Control Variable	Controlled
Constant	3.122 * (1.894)	2.298 (1.671)
N	1043	1043
R2	0.107	0.100

Note: ***, **, and * represent significance levels at 1%, 5%, and 10%, respectively. Values in parentheses are standard errors.

. The results indicate that even after replacing the explanatory or dependent variables, the regression results are consistent with the baseline regression results. The diversity of digital economy participation significantly promotes total household income, and digital economy participation also significantly increases per capita household income, further confirming the robustness of this study’s core conclusions.

Research has demonstrated that farmers’ participation in the digital economy significantly enhances household income, and this conclusion holds true even after a series of robustness checks, supporting the findings of other scholars [3,13,21]. This indicates that the widespread adoption of the digital economy in rural areas has fostered rural economic development. By participating in the digital economy, farmers can effectively break down barriers to the free flow of various factors and optimize their resource allocation efficiency. These factors may include information, labor resources, financial capital, agricultural and forestry products, and land or forest resources.

Empirical results on the different modes of digital economy participation show that digitalized production, sales, and services all significantly increase household income, which aligns with the findings of other scholars [20,25]. Interestingly, the income-boosting effects of different modes of digital economy participation vary. Digitalized sales contribute the most to farmers’ income, followed by digitalized services and production. Notably, the number of farmers participating in digitalized production far exceeds the other two categories. A potential reason for this is that digitalized production is a longer-term process, meaning the income-boosting effects for farmers may experience a time lag. In China, where smallholder farming dominates, agricultural and forestry production is characterized by small-scale operations and fragmented land, leading to relatively low levels of digitization in production. Furthermore, digitalized sales are the most dynamic area in the digital transformation of rural economies. By engaging in digitalized sales, farmers shift from obtaining producer prices to accessing consumer prices, thereby increasing the potential revenue from agricultural and forestry products. At the same time, digitalized information services can significantly reduce the costs of information transmission and search, mitigate information asymmetry, and increase the likelihood of rural labor employment and entrepreneurial activities. Digitalized financial services create opportunities for farmers to access financial loan policies, significantly lowering the cost of capital acquisition and supporting agricultural and forestry production, thus boosting income. Therefore, despite the higher participation rate in digitalized production, its contribution to income growth is smaller. Another possible reason is that compared to digitalized production, digitalized sales and services require higher levels of education, technical proficiency, and digital literacy from farmers. This suggests that there remains substantial room for growth in the digitalization of agricultural and forestry production in China, which could become a potential driver for sustainable development in these sectors.

4.4. Mediation Effect Test

To explore the mediating role of forestry management behavior and the degree of diversification in the impact of digital economy participation on farmers’ income, this study employed the stepwise regression method as part of the mediation effect test based on the baseline regression model. The stepwise regression results, presented in

Table 7. Results of the mediation effect test.

Variable	Forestry Management			Diversification
	Income	Forestry Management	Income	Income	Diversification	Income
Digital Economy Participation	0.494 ** (0.202)	0.150 *** (0.030)	0.399 * (0.204)	0.494 ** (0.202)	0.237 *** (0.043)	0.389 * (0.204)
Forestry Management			0.631 *** (0.206)
Degree of Diversification						0.443 *** (0.147)
Control Variable	Controlled
Constant	3.172 * (1.898)	0.349 (0.286)	2.951 (1.892)	3.172 * (1.898)	-0.019 (0.400)	3.180 * (1.891)
N	1043	1043	1043	1043	1043	1043
R2	0.104	0.065	0.112	0.104	0.078	0.112

Note: ***, **, and * represent significance levels at 1%, 5%, and 10%, respectively. Values in parentheses are standard errors.

, indicate that digital economy participation has a significantly positive effect on farmers’ forestry management behavior and degree of diversification. Moreover, forestry management behavior and diversification have a significantly positive effect on farmers’ income. This demonstrates a significant positive mediation effect, suggesting that digital economy participation promotes income growth through the channels of enhancing forestry management behavior and diversification, thereby validating hypotheses H2 and H3.

To verify the robustness of the mediation effect results, we further applied the Bootstrap test method [43] to examine the mediating role of forestry management behavior and diversification in the relationship between digital economy participation and farmers’ income. The number of Bootstrap samples was set to 1000. The Bootstrap test results, shown in

Table 8. Bootstrap robustness test results for the mediation effect.

	Forestry Management Behavior				Forestry Management Diversification
	Observed Coefficient	Standard Error	Confidence Interval		Observed Coefficient	Standard Error	Confidence Interval
			Lower Bound	Upper Bound			Lower Bound	Upper Bound
Direct Effect	0.399 **	0.193	0.020	0.778	0.389 **	0.189	0.018	0.760
Indirect Effect	0.095 ***	0.032	0.032	0.157	0.105 ***	0.031	0.045	0.166

Note: ***, and ** represent significance levels at 1%, and 5%, respectively. Values in parentheses are standard errors.

, reveal that the direct effects of forestry management behavior and diversification are 0.399 and 0.389, respectively, both significant at the 5% statistical level. The indirect effects are 0.095 and 0.105, respectively, significant at the 1% statistical level. The confidence intervals for the indirect effects are (0.032, 0.157) and (0.045, 0.166), neither of which includes zero, indicating that forestry management behavior and diversification mediate the relationship between digital economy participation and farmers’ income.

A possible explanation is that the development of the digital economy, particularly through advancements such as big data and e-commerce, has significantly facilitated the transition of traditional forestry management towards digitalization, enhancing the value-added production of forest products and thus increasing household income. This result aligns with the earlier theoretical analysis, suggesting that the digital economy accelerates the dissemination of advanced forestry technologies, with technological progress leading to improved production efficiency. This, in turn, encourages forestry management behavior among farmers and promotes greater output and income growth.

The research confirms that promoting farmers’ participation in forestry operations is a crucial channel through which the digital economy enhances farmers’ income, supporting findings from other scholars [34]. As the primary agents in traditional forestry operations, farmers’ engagement in the digital economy effectively elevates the level of forestry management, achieving digitization of management, precision in production, intelligent decision-making, and diversification of operations. First, by engaging in experience exchange and resource complementarity on forestry information-sharing platforms, farmers can significantly reduce transaction costs arising from information asymmetry, thereby enhancing their capacity to acquire and convert information. Second, e-commerce platforms (such as Taobao, Pinduoduo, and JD.com) provide farmers with extensive sales channels and create greater exposure opportunities for high-quality forest products. Bamboo products, understory economy products, and ecotourism services are effectively promoted and sold through these platforms, significantly enhancing the visibility and sales volume of forest products. Third, farmers can access training and technical services through the internet, learning to use new forestry management tools and pest control techniques, thereby overcoming limitations of time and location and effectively enhancing their professional skills in forestry operations. Fourth, participation in the digital economy improves the accessibility of financial services. The digital economy offers farmers more convenient loan channels, such as obtaining low-interest loans through online financial platforms, which assists them in investing in advanced forestry equipment or expanding their operational scale. Fifth, farmers can stay informed about national and local forestry policies, such as subsidies and tax incentives, through the internet, providing valuable references for their forestry management decisions. Additionally, the digital economy fosters diversification in forestry operations, enhancing household resilience to risks. Digital platforms can assist farmers in promoting regional resources, facilitating ecotourism, attracting visitors to experience the scenic beauty of forest areas, and encouraging the development of diversified operational models such as understory planting and livestock raising, thereby increasing the comprehensive benefits of forest land. The implementation of these strategies is largely driven by the digital economy, which not only enhances the operational efficiency of farmers in forest areas but also broadens income sources, providing strong momentum for regional economic prosperity.

4.5. Heterogeneity Analysis

The above research has analyzed and validated the effect of digital economy participation on farmers’ income and its underlying mechanisms. However, the income-boosting effects of digital economy participation may vary across different types of farmers. The extent and effectiveness of digital economy participation can differ depending on farmers’ occupational backgrounds, leading to variations in post-participation outcomes. Moreover, previously impoverished households, compared to non-impoverished households, may face disadvantages in terms of financial capital, natural capital, social capital, material capital, and technical capital, resulting in differing impacts after participation. Digital economy participation primarily affects agricultural and forestry production and management, likely contributing most to operating income. Thus, analyzing the income-boosting effects of digital economy participation based on income structure may reveal differences. This section conducts a heterogeneity analysis of the income-boosting benefits across different farmer types and from the perspective of household income structure.

4.5.1. Occupational Heterogeneity Among Farmers

In this study, pure farmers are defined as those whose agricultural income accounts for more than 80% of total household income, part-time farmers are defined as those whose agricultural income constitutes 20%–80% of total income, and non-farmers are those whose agricultural income makes up less than 20% of household income.

Table 9. Heterogeneity analysis based on different occupational types of farmers.

Variable	Model (1)	Model (2)	Model (3)
	Pure Farmers	Part-Time Farmers	Non-Farmers
Digital Economy Participation	1.685 * (1.011)	0.221 ** (0.090)	−0.014 (0.151)
Control Variable	Controlled
Constant	−1.637 (7.400)	7.100 *** (0.956)	5.747 *** (1.479)
N	183	313	547
R2	0.194	0.308	0.124

Note: ***, **, and * represent significance levels at 1%, 5%, and 10%, respectively. Values in parentheses are standard errors.

reports the effects of digital economy participation on total household income across different types of farmers. The regression results show that both pure farmers and part-time farmers significantly increased their household income through digital economy participation, whereas non-farmers did not experience a significant income increase from digital economy participation. This may be because the digital economy participation defined in this study involves the agricultural and forestry value chain, where non-farmers are less involved. Therefore, within the context of this study, digital economy participation did not significantly increase income for non-farmers, though this does not suggest that other forms of digital economy participation have no income-boosting effects for non-farmers.

4.5.2. Heterogeneity Among Previously Impoverished Households

In this study, households that were in poverty before 2020 but escaped poverty through government poverty alleviation policies and their own efforts are referred to as previously impoverished households, while all others are non-impoverished households. The results in

Table 10. Heterogeneity analysis based on different types of farmers.

Variable	Previously Impoverished	Non-Impoverished	Previously Impoverished	Non-Impoverished	Previously Impoverished	Non-Impoverished	Previously Impoverished	Non-Impoverished
Digital Economy Participation	0.592 ** (0.281)	0.418 (0.294)
Participation in Digitalized Production			0.596 ** (0.292)	0.265 (0.290)
Participation in Digitalized Sales					0.491 (0.426)	0.866 * (0.452)
Participation in Digitalized Services							0.655 ** (0.321)	0.634 ** (0.320)
Control Variable	Controlled
Constant	3.568 (2.650)	3.640 (2.787)	3.558 (2.651)	3.670 (2.791)	3.097 (2.653)	3.944 (2.788)	3.307 (2.645)	3.277 (2.786)
N	492	551	492	551	492	551	492	551
R2	0.135	0.090	0.135	0.088	0.129	0.093	0.135	0.093

Note: ** and * represent significance levels at 5% and 10%, respectively. Values in parentheses are standard errors.

indicate that the income-boosting effects of digital economy participation are more significant for previously impoverished households compared to non-impoverished households. This may be because, on the one hand, digital economy participation helps overcome barriers to the flow of resources, reducing gaps between previously impoverished and non-impoverished households in areas such as “information asymmetry” and “non-agricultural employment opportunities”. On the other hand, according to the theory of marginal utility, digital economy participation yields higher marginal returns for previously impoverished households compared to non-impoverished households. Additionally, internet usage breaks down barriers to resource flow, narrowing gaps between low-income and high-income households in terms of information asymmetry and non-agricultural employment options. Marginal utility theory further suggests that the marginal income-boosting effect of internet use is significantly higher for low-income households than for middle- and high-income households.

Notably, regarding the different forms of digital economy participation, both previously impoverished and non-impoverished households experience significant income-boosting effects from participating in digitalized services. However, previously impoverished households saw a more significant income increase from digitalized production, while non-impoverished households saw greater income benefits from digitalized sales. This may be because previously impoverished households are more focused on agricultural and forestry production due to limitations in technology and capital, whereas non-impoverished households focus more on sales.

The income-boosting effects of digital economy participation vary among different types of farmers, consistent with the findings of other scholars [44]. For different occupational types of farmers, both pure farmers and part-time farmers saw significant income increases from digital economy participation. This may be because the digital economy participation discussed in this study is based on agricultural and forestry production, and non-farmers are less involved in the value chain of these sectors. Thus, digital economy participation has a less significant effect on income growth for non-farmers. Participation in the digital economy significantly increased the income of previously impoverished households, highlighting the pro-poor nature of digital economy participation in poverty reduction efforts. Notably, for previously impoverished households, digitalized production activities had a significant positive effect on income. In contrast, digitalized sales activities had a more significant impact on income growth for non-impoverished households. This reflects the fact that previously impoverished households tend to be more traditional agricultural and forestry producers, relying heavily on these sectors for their livelihoods. Non-impoverished households, on the other hand, have more diversified livelihood strategies. Additionally, previously impoverished households face multiple constraints in terms of age structure, education level, and information access, and often lack the skills necessary to engage in digitalized sales. These factors collectively hinder their ability to increase household income through digitalized sales. Therefore, these findings suggest that differentiated strategies should be developed to maximize the pro-poor potential of the digital economy when promoting its growth as part of poverty alleviation efforts.

This study presents three main contributions. First, using micro-level survey data, it empirically analyzes the impact of digital economy participation on farmers’ income. Previous studies have mainly used macro-level data to examine the income-boosting effects of the digital economy, which may lead to a fallacy of composition and limit the micro-level explanatory power of the findings. Second, this study explores the impact of digital economy participation on farmers’ income from the sub-dimensional perspective of the agricultural and forestry value chain. Existing studies have mainly focused on the overall relationship between digital economy participation, rural economic development, and household income, with fewer studies examining the effect of participation in specific segments of the value chain. The development of the rural digital economy is increasingly extending to the entire agricultural and forestry value chain, including production, processing, sales, and services, accelerating the digital transformation of input supply, social services, product processing, distribution, storage, branding, and after-sales services. This, in turn, promotes household income growth and rural economic development. Third, by incorporating forestry management behavior and diversification into the analysis of how digital economy participation affects household income, this study reveals the important role of forestry production and management in the economic development of southern collective forest regions and household income growth. It further demonstrates that digital economy participation promotes more stable and sustainable income growth by enhancing farmers’ forestry management behavior and diversification strategies.

5. Conclusions and Recommendations

Based on survey data from 1043 households in forest regions of Guizhou Province, China, this study empirically examined the impact of farmers’ participation in the digital economy and its modes of participation on household income, and further explored the underlying mechanisms. First, farmers’ participation in the digital economy significantly enhances their household income, with varying income effects associated with different modes of participation. Digital sales contribute the most to household income, followed by digital services and digital production. Second, the level of forestry operations and operational diversification mediates the impact of farmers’ participation in the digital economy on their income. Engagement in the digital economy effectively improves the level of forestry management, achieving digitization of management, precision in production, intelligent decision-making, and diversification of operations, which further promotes household income growth. Lastly, heterogeneity analysis indicates that the income effects of different types of farmers participating in the digital economy vary. The income effect for pure farmers and part-time farmers is more pronounced, and the impact is particularly significant for poverty alleviation households compared to non-poverty alleviation households. This finding underscores the important pro-poor nature of digital economy participation in poverty reduction efforts. Guizhou Province is one of the key areas for collective forestry in southern China and is also a typical region in China’s impoverished mountainous areas. The sustainable development of forestry serves as a vital channel for enhancing farmers’ income and promoting green economic growth in the region. In our survey sample, 79% of farmer households own forest land, and nearly half are engaged in forestry activities, predominantly in bamboo and economic forestry. This demonstrates that in impoverished mountainous regions, the deep integration of digital economy participation with forestry operations contributes significantly to income growth for farmers and local economic development. In future research, our team will continue to closely examine the dynamics of the digital economy and forestry development in Guizhou Province, conducting ongoing surveys and longitudinal studies at the household level to systematically assess the long-term impacts of digital economy participation on farmers’ income.

Based on these findings, the following policy recommendations are proposed:

First, the empirical results of this study indicate that participation in the digital economy, along with various modes of engagement, significantly promotes income growth for farmers, and there remains substantial potential for further development. Therefore, the government should continue to support the development of the rural digital economy, leveraging its role in driving sustainable agricultural and forestry development. Efforts should focus on accelerating the application of digital technologies across the entire agricultural and forestry value chain, from production to sales, financial services, and information services. Specifically, the rural digital economy should be harnessed to expand agricultural and forestry management, reduce production costs, improve financial services, broaden product sales channels, and enhance information services, thereby ensuring steady income growth for farmers. In addition to increasing investment in rural digital infrastructure—such as broadband, 5G base stations, and data centers—the government should also accelerate the digital transformation of rural infrastructure, including roads, water conservancy, and power systems. This would improve access to digital equipment and information systems, lowering barriers to the use of digital services.

Second, given the mediating role of forestry management and diversification, efforts should be made to strengthen forestry infrastructure and encourage farmers to engage in forestry management. Promoting the integration of the digital economy with the forestry industry and enhancing forestry’s digital transformation capabilities are key. The establishment of comprehensive digital platforms for forestry would improve the digitalization and automation of the sector. At the same time, it is important to guide the orderly penetration of digital technologies into the forestry industry in line with local realities. This would ensure the rational allocation of forestry resources and facilitate the digital and intelligent upgrading of the forestry industry, thereby enhancing the income-boosting effects of the digital economy for farmers.

Third, in response to the unique needs of forest areas such as Guizhou Province, the government must develop localized or regional digital solutions. For instance, the development of a sustainable forestry monitoring digital platform is essential, which should include features such as real-time data collection, data analysis and management, sustainability assessment, and a disaster early warning system to ensure the sustainability of forestry operations and to monitor the status of forest resources. Additionally, specialized applications should be provided for smallholder farmers to access market data, with functionalities including obtaining forestry market information, conducting online transactions for forest products, receiving price fluctuation alerts, accessing digital forestry management technologies, market operation training, and an interactive experience exchange platform. Through these localized or regional digital solutions, government authorities can effectively protect the forest ecosystems in areas like Guizhou Province while simultaneously enhancing the productivity and market competitiveness of smallholder farmers, thereby promoting sustainable development and economic prosperity.

Finally, recognizing the heterogeneous effects of digital economy participation on income growth, efforts to promote the digital economy should be targeted and tailored. Improving farmers’ digital literacy and narrowing the secondary digital divide are crucial. The digital economy is a skill-driven technological advancement that requires participants to possess specific skills. However, the generally low levels of education and skills among small-scale farmers hinder their ability to participate in the digital economy and benefit from its opportunities. Therefore, the government should increase its investment in rural digital education, strengthening educational training on the application of digital technologies for farmers, thereby enhancing their capacity to acquire and process information and improving their overall competencies. Moreover, as Guizhou Province is a typical intersection of ethnic minority regions and impoverished mountainous areas, it is crucial to consider the local cultural, economic, and geographical characteristics when implementing digital training programs for smallholder farmers to ensure both effectiveness and sustainability. For instance, developing localized digital training content is essential, ensuring that the training methods align with local cultural practices. Training should begin with basic digital skills, progressing from simple to more advanced levels in a phased manner. For example, the program could start by teaching the basic use of smartphones, followed by instruction on how to utilize digital tools for forestry production and sales. Additionally, the training schedule should be adapted to align with the agricultural seasons, minimizing disruptions to farming activities.