As a central coal resource province, Shanxi Province, with about 25% of China’s resource reserves identified and retained in coal resources, mining development is an essential pillar of local economic construction and social development. Of the 118 counties in the province, up to 94 counties and districts are rich in coal resources, 91 of which have coal mines for mining operations. Long-term high-intensity coal mining activities, while promoting rapid economic growth, have profoundly altered the natural ecological patterns around the mining areas, exacerbating environmental vulnerability (Zhou et al., 2022) and directly threatening the livelihood security of farmers engaged in monoculture farming for subsistence development. Farmers in mining areas face the transformation and restructuring of their livelihoods, and the trend of part-farming and off-farming is becoming increasingly evident (Feng et al., 2024), and those who cannot escape from the countryside fall into a new poverty trap.

The theory of sustainable livelihoods, first proposed by Chambers and Conway (1992), emphasizes the ability of individuals or households to achieve their livelihood goals through diversified activities and asset portfolios without compromising the natural resource base. This theory argues that a rural household’s livelihood depends not only on its income level but also on its livelihood capital, which includes human capital, natural capital, physical capital, financial capital, and social capital (Scoones, 1998). These livelihood capitals are interdependent and mutually transformative, forming the foundation for households to cope with external risks, improve their well-being, and achieve sustainable development (Ellis, 2000). Therefore, livelihood capital has become an important analytical framework for measuring rural household well-being and sustainable development potential, and has been widely used in research fields such as poverty governance, environmental management, and rural revitalization (Tao and Wall, 2009).

Within this theoretical framework, the advancement of digital transformation provides a new path for enhancing farmers’ livelihood capital. The implementation of digital rural development, through the introduction of technologies such as the internet, mobile communications, and digital platforms, has significantly improved farmers’ ability to access information, allocate resources, and participate in the market. Taking internet technology, a widely used digital technology, as an example, statistics released by the China Internet Network Information Center (CNNIC) show that by December 2023, the number of rural internet users in my country had reached 326 million, and the internet penetration rate in rural areas was 66.5%, an increase of 4.6 percentage points from 2022. This trend indicates that digital technology is becoming a key driving force for promoting sustainable livelihood transformation in rural areas. Research has found that the application of digital technology can effectively alleviate information asymmetry among farmers in agricultural production and marketing, optimize resource allocation (Harou et al., 2022), reduce household poverty vulnerability (Zhao et al., 2025), and encourage farmers to adopt non-agricultural livelihood strategies (Liu and Su, 2022), thereby increasing household income (Liu and Liao, 2024). However, the diffusion of digital technology is uneven. Schradie (2018) points out that due to the existence of a “digital divide,” significant differences exist between different social groups in their ability to access and apply technology. Farmers and small-scale miners, as vulnerable labor groups in many parts of the developing world (Del Prado-Lu, 2017), have varying degrees of digital literacy, and their ability to utilize digital technologies and benefit from digital achievements is weaker than that of other groups (Ding, 2025). To a certain extent, they are marginalized and face difficulties in participation, further widening the gap between them and more advantaged groups (Yi et al., 2025). Therefore, the specific manifestations of the uplifting effects of digital technology among farmers in mining areas require further investigation.

For mining households, the growth of livelihood capital depends on the environmental changes in resources and various social resources formed by ties of geography, kinship, and blood, which rely on multiple collaborations of micro-farmers voluntarily (Ostrom, 2000). Collective action is an effective mechanism to address the plight of “smallholders” in developing countries in relation to “big markets” and to enhance the market performance of smallholders (Ochieng et al., 2018). For example, collective action has helped farmers in Central America to reduce the cost of entry into the relevant market (Hellin et al., 2009); smallholder farmers in the Philippines have significantly improved their economic returns and productivity by sharing risks and reducing costs through collective action (Oakeshott, 2016); and villagers in the village of Xiawu, Tibet, have participated in collective action spontaneously, forming a system of “digging three, repairing two, and taking turns to work,” which has brought tremendous vitality to the livelihoods of farmers (Sun et al., 2024). According to the 2021 Analysis Report on the Development of New Agricultural Management Subjects in China, the number of farmers’ cooperatives in China at the end of 2020 was 5.9 times that of a decade ago, with the overall number stabilizing at more than 2.2 million. However, in terms of growth rate, the growth rate of the number of farmers’ cooperatives has narrowed to 1.4%, and there are still a certain number of “empty shell” cooperatives, so the development of rural collective action is not optimistic (Hu et al., 2017). The reasons for this are found to be: under the background of “big country, small farmers,” namely, a large population, high pressure from agricultural market demand, small scale of operation of farmers, strong agricultural self-sufficiency, low commodity rate, the lack of per capita resource endowment, severe fragmentation of arable land and the continuous fragmentation of farm households have led to the decline of collective action in rural China (Wang and Zang, 2020); and the existence of cognitive bias, stakeholder demand challenges and the crisis of trust among participants in the marketization of rural land in China have all contributed to the enormous difficulties of collective action (Zhou et al., 2024). Mining farmers have also significantly reduced their collective action capacity due to resource depletion and labor outflow (Wang et al., 2016).

The emergence of digital technologies has provided necessary information support and technological underpinnings for mining farmers to carry out collective action, not only allowing organizers to operate purposefully (Cardoso et al., 2019) but also influencing the lifestyles of the participants (Young et al., 2019), with more and more farmers being willing to engage in information sharing, technological learning, risk-sharing, and resource integration (Birner et al., 2021). Agriculture is inherently risky, dependent on natural ecosystems, including temperature, rainfall, pollution, crop diseases, and price variations due to market imperfections (Singla and Sagar, 2012). As farmers’ experience of disaster increases, their risk perception is stronger (Li and Lin, 2023). The study found that farmers’ perception of health risks is the strongest and that most farmers in China have shifted their thinking from “having enough to eat” to “pursuing a better and healthier life” (Zhou et al., 2022). The perception of climate and market risks also significantly impacts farmers’ adaptive behavior (Yarong and Minpeng, 2021). In the face of risk, people try to avoid it by strengthening interpersonal relationships and seeking knowledge of best practices available to others in similar situations. So, does the level of risk perception impact the relationship between digital technology and collective action?

Existing studies provide valuable references for enhancing farmers’ livelihood capital. However, there is still room for expansion as follows: (1) Past studies have examined the effects of digital technology use and collective action on farmers’ livelihoods, and seldom have included all three in the same theoretical analytical framework, comprehensively exploring the logical role of farmers’ collective action in the relationship between the paths of digital technology use in influencing the livelihoods of farmers, as well as the role that the perception of risk plays in it. (2) There is a lack of targeted research on using digital technology, collective action participation, and livelihood capital of farm households in China’s mining areas. Compared with other regions, there are significant gaps in economic and social development, digital infrastructure, and livelihoods of farm households in mining areas, making it even more arduous to promote the construction of digital villages and rural revitalization. (3) Most of the studies on the impact of digital technology use and collective action on the livelihoods of farm households are centered on the adoption of production technology, farm household livelihood income, or farm household livelihood mode, and few studies have examined the impact on the livelihood capital of farm households. Given this, this paper constructs a moderated mediation model based on the perspective of sustainable livelihoods, measures the level of livelihood capital and collective action using entropy value method and factor analysis, respectively, and takes the mining farmers as the survey object, and applies hierarchical regression analysis and Bootstrap test to explore further the differences in the constraints on livelihood capital and the impacts of digital technology, to provide scientific evidence and decision-making references to promote the level of livelihood capital of farm households in the underdeveloped areas.

As a country rich in mineral resources, China’s mining areas are widely distributed in rural areas with fragile livelihood conditions and lagging levels of economic development, and these areas are often closely linked to the living spaces of farming households (Niu et al., 2022). Therefore, finding a balance between mineral resource development, the sustainable livelihoods of farmers, and ecological protection has become a significant challenge for sustainable rural development around mining areas. This paper takes Shanxi Province, the wealthiest province in coal resources, as the study area, and finds that although the farming households in the mining areas have escaped from poverty, the accumulation of livelihood capital is still weak. It is easy for them to return to poverty again. There are several reasons for this: first, the destruction of ecosystems has caused large-scale damage to farmland and arable land, and although the high density of ecological restoration policies in mining areas has produced significant effects, it takes time to re-establish vegetation and improve the soil (Hu et al., 2020). Second, for provinces still highly dependent on resource exports, such as Shanxi, a strong link between the rural and mining economies is critical to maintaining sustainable livelihoods for residents. Given the relatively limited number of livelihood strategies people can rely on, a green transition is unlikely to happen overnight.

Existing literature has examined the impact of mining development on the livelihood capacity of farming households from multiple perspectives. First, some studies have focused on how mining influences farmers’ choices of livelihood strategies in mining areas. In their research in Sierra Leone, Maconachie and Binns (2007) revealed a dynamic equilibrium mechanism between the mining and agricultural sectors. They found that farming households, constrained by limited resources, adopted a combined strategy of seasonal crop rotation and temporary mine employment to reallocate their livelihood capital spatially, thereby contributing to the rebuilding of the rural economy. Similarly, Balasha and Peša (2023), in their research on the Katanga Copper Belt, observed that mining activities spurred spontaneous population agglomerations around mining zones, subsequently giving rise to informal trade networks. These networks enabled farmers to buffer the loss of arable land caused by mining through short-chain marketing strategies that directly supplied goods to miners. Second, other studies approach the issue from the perspective of the conflict between resource extraction and ecological conservation. For example, Gokalp and Mohammed (2019) identified mining as a major source of heavy metal contamination in soil. This contamination leads to land degradation and poses serious risks to human health through the food chain and other pathways. Muimba-Kankolongo et al. (2022) similarly highlighted the destructive effects of mining on the agro-ecological environment, noting that extractive industries impact both soil and water sources used by farmers. Crops grown in contaminated soil experience reduced growth, performance, and yield, resulting in poor harvests and food shortages. Third, scholars have also explored the competitive relationship between mining and agriculture regarding resource allocation. Binns (1982), using a case study of diamond mining regions in Africa, found that the labor-siphoning effect triggered by the mining boom disrupted local agricultural systems, as the outmigration of young laborers weakened the region’s capacity for food self-sufficiency. Worlanyo et al. (2022), studying gold mining in Ghana, pointed out that the spatial expansion of mining land leads to agricultural land compression, posing a significant threat to agricultural production and undermining the sustainable income sources of farming households. More recently, research on mining regions in China has shifted toward the adaptive responses of farmers. For instance, Zhou et al. (2022) investigated the mining area of Daye City, Hubei Province, and found that farmland contamination by heavy metals, coupled with limited knowledge and information, hinders farmers from effectively adopting adaptation measures. Building on this, the present study examines how farmers in such contexts can enhance their livelihood capital and whether digital village initiatives can offer new pathways for development.

This paper further explores the delicate balance between farmland cultivation and mining development. Agricultural activities cannot be sacrificed entirely to halt mining operations. Meanwhile, the large-scale relocation of farmers to other regions faces numerous practical obstacles and is mainly unfeasible. In this context, digital technology holds great potential to optimize both the mining process and the development trajectory of farmers. During the pre-mining phase, the specialized application of digital tools demonstrates significant advantages. For instance, 3D radar point cloud imaging can construct real-time three-dimensional surface models of mining areas, providing accurate data for mine planning. This not only minimizes encroachment on farmland but also supports the sustainability of agricultural production (Chen et al., 2024). During mining operations, digital technologies play a vital role in environmental regulation. With their assistance, a data-sharing and monitoring platform for industrial pollution indicators, such as trace heavy metal concentrations, can enable comprehensive and dynamic pollution monitoring, thereby reducing environmental risks (Lv and Ding, 2024). In the post-mining phase, digital ecological restoration technologies have matured considerably. Precision soil remediation enables tailored treatment plans based on contamination levels; water ecosystem restoration aims to recover aquatic systems’ self-purification capacity and ecological balance; and scientifically planned vegetation restoration accelerates ecosystem recovery. Integrating these technologies can effectively repair environmental damage caused by mining activities and mitigate long-term losses (Sun, 2024). Farmers, as primary users of farmland and key actors in food security, play a crucial role in the agroecosystem. Understanding their perceptions of mining-related pollution is essential for addressing environmental challenges such as heavy metal contamination (Zhou et al., 2022). Digital technologies offer accessible platforms for farmers to acquire knowledge, stay informed about the ecological impacts of mining, and learn relevant countermeasures. Furthermore, these technologies facilitate communication between farmers and government authorities, enabling farmers to voice their concerns and contribute suggestions. Such interactions foster cooperation and jointly promote the sustainable development of mining regions.

The popularization and application of digital technology often require specific cost inputs, including the purchase of equipment and skills training, etc., which constitutes a greater economic pressure on farmers in mining areas with relatively poor economic conditions. At the same time, farm households cannot master emerging digital technologies due to education, age, and traditional thinking limitations (Odini, 2014). Nonetheless, this paper still argues that in the context of digital village construction, farm households’ development prospects for using digital technologies are broad. By introducing a series of policies and expanding technology coverage, farm households can enjoy a more convenient and low-cost environment for learning and applying digital technology, thus enhancing their livelihood capacity and living conditions. In addition, the frequency and extent of digital technology use positively impact rural residents’ livelihood capital. Although some studies have shown that excessive use of digital technology for entertainment may lead to problems such as online pan-entertainment and Internet addiction (Young et al., 2019), which in turn reduces livelihood capacity, survey data show that the frequency of digital technology use by rural households in China’s mining areas is relatively low and has not reached the level of excessiveness. On the contrary, digital technology has brought practical benefits to rural residents by enabling them to access production and business information, learn skills, receive government services, and participate in government supervision in a timely and convenient manner. For example, the contribution of digital technology to the livelihood capital of rural households is particularly prominent in the older group than in the middle-aged and younger groups. That is, access to the Internet for the “disadvantaged” not only helps to improve the plight of older people who have “nothing to rely on” (Hill et al., 2015) and enrich their recreational life (Siegel and Dorner, 2017), but also encourages them to pursue a proactive approach to their livelihoods (Li et al., 2025). For the majority of mining villagers, the ability of rural residents to use online information should be improved, and rural residents should be differentially guided to actively use digital technology, such as increasing the frequency of digital technology use in business activities.

Next, this paper attempts to contextualize its main research findings within the framework of classical collective action theory:

First, regarding the conclusion that digital technology promotes collective action, according to Olson’s (1971) view of collective action, individuals in large groups often lack the motivation to actively participate in collective action due to high coordination costs and the prevalence of free-riding, unless selective incentive mechanisms are in place. This paper’s results show that the introduction of digital technology reduces the costs of information collection and communication, lowering farmers’ coordination costs in collective procurement, disaster response, and market access, effectively providing a new form of “incentive.” Furthermore, the application of digital technology improves information transparency and process traceability, providing new tools for oversight and regulation. For example, in agricultural input procurement, revenue distribution, or public affairs negotiations, farmers can access real-time information on fund flows, task progress, and member contributions through WeChat groups, online public information platforms, or agricultural apps, thereby reducing information asymmetry and diminishing the scope for opportunistic behavior. This is highly consistent with Olson’s proposition of “reducing the costs of collective action,” demonstrating that digital tools can overcome the traditional dilemma of collective action in resource-constrained settings. From another perspective, Ostrom (1990) argues that public affairs management mechanisms are not limited to government or market approaches, but rather multiple possibilities, critically depending on management effectiveness, benefits, and fairness. This paper finds that, first, digital technology improves information symmetry and communication efficiency, reducing farmers’ concerns about “uncertainty” and “free riding” in cooperation, thereby enhancing trust in cooperation and willingness to coordinate. Second, digital platforms demonstrate low cost and high efficiency in resource integration, helping to alleviate the structural barriers of poor transportation and dispersed resources in mining areas and significantly reducing the organizational costs of collective action. Third, as farmers frequently interact on digital platforms, they gradually internalize the values of shared development and win-win cooperation. These mechanisms demonstrate that digital technology can enhance the possibility and sustainability of farmers’ collective action through the three pathways of information, resources, and cognition, thereby confirming the institutional value of technological factors within a multi-faceted governance model.

Second, we examine the role of collective action in enhancing farmers’ livelihood capital in mining areas. On the one hand, through self-organization and rule-based constraints, farmers collectively accumulate and manage natural, social, and human capital, achieving collective outcomes that transcend individual capabilities. This model emphasizes endogenous cooperative motivations and community cohesion, aligning with the concepts of polycentric governance and community self-governance. On the other hand, practices such as using collective credit to enhance the availability of financial capital, promoting scale operations through land transfer, and obtaining dividends through equity cooperation rely more heavily on the external integration of market-oriented logic and institutional arrangements, highlighting efficiency and economic benefits. Therefore, it can be argued that the collective action of farmers in mining areas does not simply remain at the level of community autonomy, but rather lies within a process of intertwined “community co-governance and market-driven development.” Its outcomes embody both the self-governance of public affairs emphasized by Ostrom and the distinct attributes of market-based cooperation.

Third, the role of risk perception in promoting the relationship between digital technology use and collective action may encompass two logics. Initially, farmers in mining areas often adopt a passive approach to risk shocks, relying on digital technology and group collaboration driven by loss aversion and self-protection, thereby passively moving towards prosocial behavior. This aligns with the concept of “risk-driven passive adaptation” emphasized in behavioral economics. However, as interactions increased and trust gradually built, collective action ceased to be simply an emergency response to risk situations and gradually evolved into a cooperative mechanism with endogenous motivations. During this phase, Ostrom’s view that “adversity fosters cohesion” was more fully embodied: risk pressures in turn stimulated higher levels of trust and a sense of community among community members, thereby driving farmers’ sustained participation in collective action. This shift from “passive adaptation” to “active cohesion” reveals the dual role that risk perception can play at different stages of development.

Finally, this paper will conduct an in-depth analysis of the goodness of fit of the constructed models. The R² coefficients of the main models all exceeded 0.13. In the field of socioeconomic research based on household survey data, this level is considered reasonable and has a certain degree of explanatory power. The livelihood capital of farmers in mining areas is influenced by multiple factors, including natural endowments, social networks, family characteristics, and the policy environment. However, this study focuses on the mechanisms through which digital technology use and collective action influence livelihood capital. Because it is difficult to comprehensively cover all influencing factors, the explanatory power of the model is relatively limited. Furthermore, farmers exhibit significant individual differences in digital technology application capabilities, social participation, and risk perception, which to some extent affects the overall goodness of fit of the model. This study focuses on identifying the interaction pathways and mechanisms between variables. The direction and significance of the coefficients of the main variables remain consistent across different model settings, indicating that the conclusion that digital technology use significantly enhances farmers’ livelihood capital through collective action is robust and has realistic explanatory power.

Based on the micro-survey data of 312 farmers in mining areas in Shanxi Province, an underdeveloped area, the livelihood effects of farmers’ use of digital technology in the process of digital village construction were investigated and analyzed. The results show that: first, the use of digital technology helps to improve the level of farmers’ livelihood capital. Compared with farmers who do not use digital technology, the level of farmers who use digital technology is 0.177 units higher. Second, collective action plays a partial mediating role in the path relationship between the use of digital technology and farmers’ livelihood capital, and the mediating effect accounts for 25.08% of the total effect. Third, risk perception can positively regulate the relationship between the use of digital technology and collective action, and affect farmers’ livelihood capital through collective action. Fourth, the sub-sample study found that the use of digital technology has a more significant positive effect on the livelihood capital of farmers in the male group and the elderly group. Fifth, there is a significant positive correlation between the frequency of digital technology use of farmers in mining areas and the accumulation of livelihood capital. Specifically, farmers’ high-frequency participation in online shopping, leisure and entertainment, and online business significantly promoted the appreciation of their livelihood capital, with coefficients of 0.070, 0.068 and 0.071, respectively. Sixth, as the use of digital technology deepens, its positive impact on farmers’ livelihood capital continues to exist, but this promotion effect shows a marginal decreasing trend. On the basis that farmers’ households have access to the Internet, further increasing mobile phone service expenses and online shopping behavior can still bring incremental livelihood capital, but this growth effect is gradually weakening. Based on the research conclusions of this paper, the following insights can be obtained.

First, digital skills training for farmers in mining areas should be carried out according to local conditions to improve their adaptability and management level. The government should work with mining enterprises, agricultural cooperatives and other entities to set up digital skills training courses specifically for farmers in mining areas. The course content covers practical application skills such as e-commerce sales of agricultural products, short video marketing, and digital operation of agricultural equipment. The training schedule should fully consider the characteristics of labor in mining areas, and flexibly adopt forms such as “night school + online + mobile classroom” to improve participation and effectiveness. At the same time, a reward mechanism for “pioneer farmers in digitalization in mining areas” will be established to provide outstanding farmers with financial subsidies, equipment support or tax exemptions to encourage them to take the lead in exploring diversified ways to increase income in resource-based areas. Second, strengthen the organizational and collaborative capabilities of farmers in mining areas and promote “cooperation to resist risks and collective development.” Relying on existing agricultural organizations or new rural collective economic organizations around mining areas, farmers’ understanding and trust in cooperatives and joint planting and breeding models will be enhanced through special training, case publicity, demonstration and guidance. The government should provide financial subsidies, tax exemptions and infrastructure support for agricultural communities formed by farmers in mining areas, encourage farmers in marginal areas of mining areas to share resources and make centralized investments, improve production efficiency and market bargaining power, and break the limitation of fighting alone. Third, build a digital information platform exclusively for farmers in mining areas to achieve resource sharing and intelligent management. The government can take the lead in developing an agricultural information service platform exclusively for farmers in mining areas, providing accurate information such as regional market prices, agricultural meteorological warnings, and pest and disease control. The platform can also set up functions such as “agricultural technical questions and answers” and “farmer mutual assistance circles” to facilitate experience exchange and knowledge sharing. Farmers in mining areas who use intelligent agricultural machinery, drone plant protection systems, and precision planting management software will be given purchase subsidies or technical support to promote the development of agriculture in the direction of intelligence and greening, and improve the quality and benefits of ecological agriculture in mining areas. Fourth, focus on the return of talents in mining areas and build a talent revitalization model of “mining areas + universities + scientific research.” Promote mining areas to carry out co-construction projects with universities and agricultural research institutes, establish a “mining area digital agriculture experimental base,” and attract teachers and students of relevant majors to go deep into mining areas for field services. Establish exclusive incentive mechanisms, such as providing targeted entrepreneurship subsidies, housing security and professional title evaluation benefits for college students and technicians who return to their hometowns to start businesses, to attract more talents to inject intellectual and technical resources into the digital transformation of agriculture in mining areas, and to create a professional, young and long-term stable local talent team. Fifth, open up the “last mile” of digital infrastructure in mining areas to ensure full access for vulnerable groups. In view of the communication characteristics of mining areas with many mountains, tunnels and blind spots, the government should accelerate the construction of communication base stations and fiber optic networks, especially in remote villages in mining areas and areas where laid-off miners live, to strengthen digital infrastructure investment. And coordinate operators to introduce “mining area care packages” to provide preferential network rates for low-income farmers, left-behind elderly people, etc. Simultaneously carry out digital literacy public welfare courses and practical APP operation guidance to improve the information application capabilities of marginal groups and truly achieve “no one left behind” digital inclusion.

This study has several limitations that warrant further exploration in future research. First, due to constraints in data accessibility, the study focuses on mining households in Shanxi Province. While the sample possesses a degree of representativeness, its size remains limited. Given the substantial heterogeneity in farmers’ conditions and digital infrastructure across different regions and resource endowments, future research should aim to expand the sample size and geographic coverage to improve the scientific rigor and generalizability of the findings. Second, although this study centers on the potential of digital technologies to enhance the livelihood capital of mining households, livelihoods are inherently multidimensional, encompassing aspects such as income, strategies, and risk. Therefore, subsequent research should adopt a more comprehensive perspective, exploring these various dimensions to understand the characteristics better and influencing factors of rural livelihoods. Third, the impact of digital technology on livelihood capital is likely influenced by a range of external factors not addressed in this study, such as infrastructure improvements, shifts in economic policy, and the pace of technological advancement. Future studies should incorporate a broader array of mediating and moderating variables to gain deeper insights into the underlying mechanisms, allowing for a more nuanced analysis of how digital technology interacts with various contextual influences. Finally, the reliance on cross-sectional data limits this study’s ability to capture the dynamic evolution of digital technology adoption and its long-term effects on livelihood outcomes. Future research should adopt longitudinal designs to monitor changes over time, offering more robust and actionable insights into how digital transformation can sustainably impact the economic development of mining-area households.