The theory of quality of life posits that an individual’s wellbeing is derived from the positive emotions generated by the continuous satisfaction of their needs. For farmers, essential needs that enhance wellbeing in the context of digital village development include material needs (He, 2017), social security (Yin et al., 2021), social participation (Huo et al., 2021), and basic psychological needs (Liu et al., 2013).

The digitization of the rural economy plays a pivotal role in meeting farmers’ material needs by providing “more satisfactory income” and improving their overall wellbeing. Specifically, the rapid expansion of rural e-commerce, particularly through the “Digital Commerce Promotes Agriculture” project initiated in 2022, has optimized various aspects of rural production, supply, marketing, and service sectors. Digital agricultural production technologies have improved operational efficiency, while digitization of agricultural supply chains and marketing have enhanced the circulation efficiency of agrarian products (Li and Li, 2024). Furthermore, the development of digital inclusive finance provides crucial financial support for farmers’ entrepreneurial activities (Zhang et al., 2019). As a result, digital initiatives in the rural economy enable farmers to achieve substantial gains in agriculture, employment, and entrepreneurial income, thereby continuously fulfilling their material needs.

Digital infrastructure and the digitization of daily life are equally critical in enriching farmers’ social needs and enhancing their SWB. The construction of digital infrastructure, supported by the “Wide China strategy” and the “Broadband Frontier” program, has substantially improved connectivity in rural areas. This advancement reduces information transmission costs, accelerates knowledge creation and sharing, and promotes equal opportunities (Zhan et al., 2024). Enhanced access to government services has further addressed farmers’ developmental needs, including improved education, healthcare, and social security (Pang et al., 2023). Meanwhile, the digitization of daily life supports both social security and participation needs. It includes various aspects of rural life, such as consumption, culture, tourism, education, healthcare, and essential services. Digital consumption improves the convenience of purchasing goods and services for farmers, while developments in digital culture and tourism enrich social participation opportunities. Additionally, digital education and healthcare provide high-quality social security services, making essential resources more accessible. In summary, the development of digital rural areas continuously meets farmers’ social needs—particularly in terms of participation in digital life and leisure tourism (Wei and Jiang, 2023)—thereby significantly enhancing their wellbeing.

The digitization of rural governance plays a critical role in enhancing farmers’ social participation and fulfilling their basic psychological needs, improving their wellbeing. On the one hand, the promotion of digital rural governance encourages farmers to engage more actively in rural revitalization and governance, enriching their social interactions in areas such as community relations and public affairs (Chen and Yang, 2024). On the other hand, the advancement of digital rural governance addresses farmers’ basic psychological needs, such as social identity and a sense of belonging (Liu et al., 2013; Zhao and Fu, 2017). The development of rural e-government improves the transparency and accessibility of village government information, which, in turn, bolsters farmers’ trust and fosters a sense of collective identity and belonging to their local governance (Zhang and Yi, 2023).

Based on these findings, this article proposes research:

Based on the capability approach, Binder (2014) proposes a theoretical framework for wellbeing, suggesting that wellbeing operates as a function of individual capability. “Capability” refers to the opportunities and freedoms individuals have to achieve their desired functions, encompassing all the potential functions they can choose from. “Functioning” represents aspects of life that people value and have reason to cherish, emphasizing the pursuit of happiness (Sen, 1985). Farmers utilize internal and external conversion factors to transform their available income, goods, and services into functions related to wellbeing. Internal conversion factors include personal attributes such as age, gender, and education level, while external transformation factors encompass social policies, social norms, cultural contexts, and natural environments. In this sense, capability emerges as a critical determinant of individual wellbeing (Yeung and Breheny, 2016). Empirical studies have confirmed that access to educational opportunities, safety (Casini et al., 2021), respect, leisure, and social connections (Steckermeier, 2021) significantly enhance individual wellbeing.

From a macro perspective, social policies also play a significant role in shaping wellbeing. Increased public expenditure by governments aimed at improving healthcare and medical insurance (Tran et al., 2017), as well as the implementation of reasonable secondary distribution policies, can contribute to enhancing wellbeing (Deaton, 2018). Conversely, adverse social conditions, such as income inequality (Cheung, 2018) and government corruption (Li and An, 2020), can significantly diminish wellbeing. Furthermore, some studies have explored the role of social transformation factors in influencing the relationship between capability and wellbeing. Broader contextual elements—such as the country’s economic conditions, civil liberties, and societal values —affect key capabilities, including financial security, leisure, and social recognition, thereby influencing overall wellbeing (Steckermeier, 2021).

In this study, digitalization of rural governance and digitalization of rural life are conceptualized as two critical dimensions of digital village development. They are incorporated into the existing capability framework as social conversion factors to examine how they influence farmers’ subjective wellbeing through two core capabilities: belonging capability and self-development capability.

On the one hand, traditional rural governance has long been constrained by problems such as information asymmetry, fragmented public services, and limited villager participation. To overcome these challenges and advance rural governance modernization, governments have actively promoted digital governance transformation. For instance, platforms such as WeChat, Alipay, Alibaba Rural Ding, and China Telecom’s Village-to-Village Service have been increasingly embedded into grassroots party building, farmer training, village affairs disclosure, financial management, and democratic supervision. These initiatives alleviate information asymmetry, diversify villager participation channels (Su et al., 2023), and enhance villagers’ sense of identity and emotional attachment (Ding, 2021), thereby strengthening their belonging capability. In addition, provincial-level digital governance platforms streamline public service processes and improve service efficiency, thereby expanding farmers’ access to high-quality education, healthcare, and elderly care services, which in turn enhances their self-development capability.

On the other hand, digitalized rural life at the regional level encompasses digital consumption, digital education, digital tourism, digital culture, and digital healthcare. As an important social conversion factor, it exerts a significant impact on farmers’ core capabilities. Specifically, digitalized rural life broadens channels for social participation and interaction (e.g., online communities, digital cultural and tourism activities), fostering social integration and emotional belonging. Furthermore, digital education enhances farmers’ knowledge and skills, while digital consumption and digital tourism expand their horizons and information access, thereby empowering them to achieve personal development. Overall, these environmental factors reinforce farmers’ development capability and foster a stronger sense of accomplishment in self-development.

Consequently, we propose the following research:

The relative deprivation theory posits that environments characterized by income inequality can intensify the sense of “relative deprivation” among groups, thereby diminishing individuals’ overall wellbeing. Helsper (2017) argues that citizens assess their social equality status based on their emotional perception of relative disadvantage within a group. When individuals perceive their living conditions to be inferior to those of their peers and believe they deserve a better quality of life, they are likely to experience negative emotions such as anger and resentment (Smith and Pettigrew, 2014). One perspective suggests that digital rural construction can help alleviate relative deprivation in rural areas. Digital infrastructure and a digital economy can foster non-agricultural employment, enhance agricultural productivity, improve information accessibility, and develop human capital (Si and Wang, 2024; Wang et al., 2024). Conversely, another viewpoint argues that such digital initiatives may fail to reduce relative deprivation. The marginal agglomeration effect, combined with the concentration of digital technologies and financialization, enables high-income individuals to access financial products and digital technologies at lower costs, thereby widening income inequality (Daud et al., 2021). Furthermore, the digital divide and the expansion of social networks may further exacerbate relative deprivation further (Liu and Luo, 2023; Antinyan et al., 2019; Zhang and Sun, 2024).

This article argues that digital infrastructure and economic digitization can contribute to the Matthew effect of welfare inequality. While these advancements can raise incomes for farmers overall, resource-rich individuals are more likely to benefit disproportionately from the digital dividend, resulting in a faster rate of income growth compared to the less affluent groups. At the same time, the widening digital divide is expected to exacerbate the income gap between resource-rich and resource-poor farmers. Therefore, we propose the following research:

This study employed two matched datasets. The first dataset comprises county-level digital village development data, sourced from the County Digital Village Index (2018 and 2019), jointly released by the Peking University New Rural Development Research Institute and the Alibaba Research Institute. This dataset provides insights into the status of digital countryside initiatives across various Chinese counties.

The second dataset is derived from a rural household survey conducted by the research team between July and August 2020. The survey employs a stratified random sampling method. Provinces were first selected based on regional economic development levels. Within each selected province, five counties (or districts) were chosen, followed by one village per county. Finally, households were randomly sampled within each village. After integrating Peking University’s Digital Rural Dataset, a total of 666 valid farmer responses from 24 counties across five provinces were retained for analysis. The distribution of the matched samples is presented in Figure 2, and the main characteristics of the three surveyed regions are summarized in Table 1. In eastern China, Zhejiang Province represents the relatively developed region, with a per capita GDP exceeding 100,000 RMB and rural per capita disposable income of 29,876 RMB in 2019. A total of 187 valid samples were collected from six sites: Xianju, Yuyao, Jingning, Jiangshan, Pan’an, and Suichang. In central China, Hubei and Jiangxi illustrate moderate development, with per capita GDPs of 77,387 RMB and 53,164 RMB, and rural incomes about 54% of Zhejiang’s level. Together, 224 samples were collected, including 107 from Hubei and 117 from Jiangxi. In western China, Guizhou and Yunnan are less developed, with per capita GDP levels below 50,000 RMB and rural incomes just above 10,000 RMB—roughly 36–40% of Zhejiang’s level. A total of 255 samples were obtained, 125 from Guizhou and 130 from Yunnan.

Notably, within Zhejiang Province, Jingning and Pan’an are relatively underdeveloped, with rural per capita incomes of 20,005 RMB and 19,461 RMB, comparable to those of several central counties such as Tianmen (18,138 RMB), Yicheng (20,154 RMB), Gong’an (19,784 RMB), and Wuning (18,086 RMB). A total of 60 samples were collected from these two sites. Considering this, the overall sample size is broadly consistent with the proportional distribution of the rural population across the surveyed regions.

Specifically, rural digital infrastructure serves as the material foundation for digital village construction, encompassing five key aspects: information infrastructure, digital financial infrastructure, digital commercial hubs, and essential digital resources. The digitization of the rural economy forms the core of digital village construction, including four dimensions: digital production, digital supply chains, digital marketing, and digital finance. The digitization of rural governance functions as a crucial mechanism for advancing digital government and rural transformation, comprising four governance indicators. Lastly, the digitization of rural life represents the fundamental objective of digital village construction, covering six essential areas: digital consumption, digital culture, digital education, digital healthcare, digital tourism, and digital life services.

First, digital village development enhances social belonging by providing farmers with more excellent opportunities for social participation, strengthening their social connections, and reshaping their identity with traditional cultural concepts. These aspects of social participation and connections are critical in fostering a sense of social belonging (Nussbaum, 2009). The social belonging profoundly influences farmers’ sense of identity and belonging (Suppa, 2021), significantly contributing to overall life satisfaction (González et al., 2021). In this study, social belonging is measured by indicators such as mutual assistance between neighbors and the practice of daily social gifts.

Sen (2012) emphasizes that social opportunities are crucial for achieving freedom, which includes government-provided public services such as education, healthcare, and elderly care. In this context, the construction of digital rural areas enhances the quality and accessibility of these essential public services through digital governance and improved digital living conditions. Such development is instrumental in strengthening farmers’ self-development abilities. A more substantial development capability correlates with a heightened sense of personal advancement among farmers, which in turn increases their wellbeing (Zhu and Li, 2024).

Sen (2012) emphasizes the importance of social opportunities in achieving freedom, which encompasses access to government-provided public services such as education, healthcare, and elderly care. In this regard, the construction of digital rural areas significantly improves the quality and accessibility of these essential services through digital governance and enhanced digital living conditions. Such developments are crucial for strengthening the farmers’ self-development abilities. A more substantial development capability fosters a heightened sense of personal advancement among farmers, leading to increased wellbeing (Zhu and Li, 2024). To empirically capture farmers’ development capabilities, this study analyses their subjective evaluations of several key areas, including major illness medical security, elderly care security, children’s education, village public facilities, and community safety.

To explore the equality mechanism through which digital rural construction impacts farmers’ wellbeing, this study draws on relevant research (Wu and Li, 2024; Hua and Pan, 2024) and employs the Kakwani inequality index to assess the relative sense of deprivation experienced by farmers. This index is introduced as a mediating variable. If digital rural development successfully reduces relative deprivation among farmers, it enhances social (income) equality, leading to higher wellbeing. Conversely, if digital rural development exacerbates relative deprivation, it diminishes social (income) equity, resulting in lower wellbeing.

To investigate the impact of digital village development on the wellbeing of farmers, this study constructs a benchmark linear regression model.

In Equation 1, Wellbeingi represents the dependent variable, representing the wellbeing of the i-th farmer. Digital_villagei denotes the core explanatory variable, indicating the level and dimensions of digital village development in the county where the ith farmer resides. Correspondingly, β1 represents the estimated parameter value of the explanatory variable. CVi refers to a series of control variables that capture various personal and family characteristics of the respondent, with βm indicating the estimated parameter values associated with these control variables. Finally, εi represents the random error term.

Given that the traditional three-step mediation method is unable to confirm causal relationships (Wen and Ye, 2014), Jiang (2022) proposed a mediation approach that assesses the significance of the relationship between the core explanatory variable and the mechanism variable, as well as the significance of the relationship between the mechanism variable and the dependent variable (Jiang, 2024). This method has gained increasing attention in the field (Sun and Han, 2024; Song and Fu, 2025). In this article, we construct a mechanism model represented by equations 2, 3. Here, Mechanismi denotes the mechanism variable, which includes factors such as the income inequality of the i-th farmer, as well as the social belonging ability and self-development ability of the i-th farmer. α1 represents the impact of digital rural development on the mechanism variables, while γ1 denotes the effect of the mechanism variable on farmers’ wellbeing. If both α1 and γ1 are found to be significant in equations 2, 3, this indicates the presence of significant mechanistic effects.

For the robustness checks, the ordered probit model is employed as an alternative estimation method. In addition, two-stage least squares regression with instrumental variables (IV-2SLS) is applied to address potential endogeneity. To further examine the mechanism of income inequality, quantile regression is conducted to capture the heterogeneous income effects of digital rural development across different quantiles of rural households.

Table 3 reports the benchmark regression results on the impact of county-level digital rural development on farmers’ wellbeing. Columns (1)–(5) present the OLS estimates for the core explanatory variables, while Columns (6)–(10) report the results with control variables. To ensure model validity, diagnostic tests are conducted. For example, in Model 6, the VIF values of the 17 explanatory variables range from 1.07 to 2.44, all below the conventional threshold of 5, indicating no multicollinearity. The Breusch–Pagan (BP) test yields an LM statistic of 0.87 (p = 0.351), suggesting no linear-form heteroscedasticity, whereas the White test reports an LM statistic of 209.87 (p = 0.01), confirming heteroscedasticity. Models 7–10 produce similar results. Accordingly, OLS estimation with robust standard errors is employed (see Table 3).

The findings indicate that, after accounting for individual, family, and regional characteristics, the estimated coefficient for digital rural development at the county level is significantly positive, indicating that digital rural initiatives enhance farmers’ wellbeing. From a dimensional perspective, the coefficients for rural digital infrastructure, the digitization of rural economies, rural governance digitization, and the digitization of rural life are all significantly positive, suggesting that each of these four dimensions of digital rural construction contributes to improving farmers’ wellbeing. Overall, the results confirm a positive correlation between the development of digital rural areas and farmers’ wellbeing, underscoring the role of digital rural construction in enhancing farmers’ happiness.

From the perspective of control variables, the coefficients for the health status of the household head, land area, social capital, social security, and political trust are all significantly positive. This indicates that higher levels of education among household heads are associated with greater wellbeing. Additionally, families with more material and social capital, along with better access to medical and educational security, and greater trust in community officials, tend to report higher wellbeing. Furthermore, the coefficient for Internet usage is significantly positive, suggesting that households with Internet access can utilize it to enhance farmers’ wellbeing. On the other hand, the coefficient for debt is significantly negative, indicating that substantial financial obligations, such as housing loans and educational loans for children, can significantly reduce life satisfaction. At the county level, government financial support demonstrates a significantly positive impact, suggesting that increased economic assistance from these authorities can effectively enhance farmers’ subjective wellbeing. In some models, the coefficients for the county-level industrial structure and healthcare levels are also significantly positive; however, these results lack robustness.

In conclusion, digital rural development and its four dimensions exert a significant positive impact on farmers’ subjective wellbeing. This suggests that, as a key strategy for rural revitalization, digital rural development can serve as an effective means of enhancing farmers’ subjective happiness.

To test the robustness of the regression results, alternative dependent variables, key explanatory variables, and estimation methods are employed. Table 4 reports the ordered probit estimates. Columns (1)–(5) present the results based on life satisfaction, measured on a 1–5 scale. The Pseudo R² values of Models 1–5 all exceed 0.1, indicating acceptable model fit. Compared with regressions excluding household- and county-level factors, the Akaike Information Criterion (AIC) and Bayesian Information Criterion (BIC) for Models 1–5 decline substantially, suggesting improved model performance. For example, the AIC and BIC of Model 1 are 1,437.835 and 1,532.362, respectively, which are markedly lower than the corresponding values (1,595.743 and 1,631.753) when household- and county-level factors are omitted.

Columns (6)–(10) report the regression results when the dependent variable is replaced with confidence in daily communication, while columns (11)–(15) present the estimated results after substituting the core explanatory variable with Peking University’s 2019 Digital Village Index. Across these robustness tests, the coefficients of digital rural development and its four dimensions remain significantly positive, providing further evidence that digital village development improves farmers’ life satisfaction. These findings are consistent with the baseline regression, thereby confirming the robustness of the study’s conclusions.

When analyzing the wellbeing of farmers from the perspective of county-level digital rural development, the main challenge lies in potential endogeneity arising from omitted variables and measurement errors. To mitigate these issues, this study employs instrumental variable methods and sensitivity analysis, thereby enhancing the robustness and credibility of the empirical results.

Drawing on Wei’s (2022) research and considering data availability, this study selects the number of fixed telephone users in each county and district in 2011 as the first instrumental variable for digital rural development. The number of fixed telephone users is closely linked to the development of digital rural areas but does not directly impact the farmers’ wellbeing. Additionally, following Zhang et al.’s (2019) approach, the spherical distance between the city where rural households reside, and Hangzhou is used as the second instrumental variable for digital rural development. The underlying data for the Digital Village Index primarily originates from Ant Group in Hangzhou, a city that is highly advanced in digital economy. The diffusion and spillover effects of digital technology from Hangzhou are influenced by geographical distance, making it strongly correlated with digital village development across regions. Importantly, this distance does not directly affect the subjective wellbeing of farmers (Hu et al., 2024).

Table 5 presents the results from the two-stage regression using instrumental variables. In the first-stage, both the under-identification test (LM) and the weak instrument test (Wald F) decisively reject the null hypothesis for both instrumental variables, confirming their validity. In the second stage, the coefficient for digital rural development at the county level remains significantly positive, thereby providing further support for the robustness of the benchmark regression results.

The sensitivity analysis method proposed by Cinelli et al. (2020) effectively addresses concerns related to potential omitted variables. Lin and Zhu (2022) adapted this approach to assess the minimum strength an omitted variable must have to invalidate previous research conclusions and conducted sensitivity tests for omitted variables. Table 6 presents the estimated results, with Parts A and B showing the findings without and with included comparison variables, respectively. The results for the four dimensions of the index similarly demonstrate robustness, though these are not included here due to space constraints.

The analysis revealed that when omitted variables with a strength three times greater than the original were introduced, most variables exhibited R²dz values and R²yz dx values smaller than the robustness thresholds RV_q = 0.1364 and RV_qa = 0.0669, respectively. This indicates that the conclusions would only be overturned if the omitted variables were more than three times stronger than all included control variables. Therefore, this study concludes that the risk of significant bias from omitted variables in the model is extremely low, further reinforcing the reliability of the original model results.

In conclusion, the results derived from the instrumental variable approach and sensitivity analysis confirm that the finding—namely, that digital villages significantly improve farmers’ subjective wellbeing—is highly robust.

Due to substantial variations in the level of digital rural development and differences in household resources (such as income and social capital) across the surveyed regions, the happiness effects of digital rural development may vary significantly. Table 10 presents the results of the heterogeneity analysis. First, regarding the heterogeneity of digital rural development, counties were divided into more developed and less developed groups based on the average digital rural index. The results reveal that, compared to less developed counties, the impact of digital rural development on farmers’ SWB is greater and more significant in counties with more developed digital villages. This finding contrasts with the conclusions of Liu and Lai (2024), who suggested that digital rural development benefits farmers in underdeveloped areas more. Instead, the current results indicate that digital rural development has widened the welfare gap between farmers in developed and underdeveloped areas. Second, with respect to income heterogeneity, the sample was divided into high-income and low-income groups based on farmers’ average annual income. The results show that, compared to the low-income group, digital rural development exerts a more significant impact on the SWB of the high-income group. This suggests that digital rural development has not substantially improved the subjective wellbeing of low-income farmers but has had a more pronounced effect on high-income farmers. Third, regarding social capital heterogeneity, the sample was divided into high-level and low-level social capital groups based on farmers’ average social capital. The results demonstrate that digital rural development significantly impacts the SWB of both groups. However, the effect is markedly stronger for farmers with high social capital (1.045) than for those with low social capital (0.428).

In summary, resource conditions play a critical role in determining whether digital rural development can enhance farmers’ wellbeing. Higher levels of digital rural development, higher income, and higher social capital significantly increase the likelihood that farmers will derive tangible benefits from digital rural development, thereby achieving higher levels of subjective wellbeing.