The digital economy, characterized by data as a key production factor and digital technologies as enabling tools, has been widely recognized as a fundamental driver of efficiency improvement and structural transformation across multiple sectors (Guo et al., 2023; Liu et al., 2025). Existing studies consistently show that the digital economy reshapes traditional production and governance systems by reducing information asymmetry, optimizing resource allocation, and accelerating technological innovation, thereby exerting broad economic, environmental, and social impacts (Wang J. et al., 2022; Wang X. et al., 2022; Wang L. et al., 2023; Wang Z. et al., 2023). Similar cross-sectoral effects of the digital economy have also been documented in the tourism sector, where digitalization promotes growth through online attention allocation and transportation infrastructure, as well as in rural human settlement improvement, highlighting its role in enhancing living environments beyond purely economic outcomes (Liu and Huang, 2025; Liu et al., 2023; Liu et al., 2024b).

From the perspective of environmental sustainability, a growing body of literature demonstrates that digital economy development significantly contributes to green outcomes. Empirical evidence from China indicates that the digital economy reduces carbon emissions and enhances carbon efficiency by promoting green technological innovation and upgrading industrial structure, with these effects exhibiting clear nonlinear characteristics and regional heterogeneity. Similar findings are reported in studies on energy systems, where digitalization alleviates energy poverty primarily through innovation-driven channels, and the positive effects only emerge after surpassing certain development thresholds (Lyu et al., 2023). These results highlight that the environmental dividends of the digital economy are neither automatic nor uniform, but depend critically on development stages and structural conditions.

Beyond environmental outcomes, the digital economy has also been shown to exert profound influences on social governance and public service provision. Recent studies focusing on rural China reveal that digital economy development improves public security by increasing non-agricultural employment opportunities and household income, thereby indirectly enhancing social stability (Wang et al., 2025). In the public health domain, interprovincial evidence suggests that the digital economy significantly raises medical and health service levels through improved resource allocation efficiency and coordinated regional development, while generating notable spatial spillover effects (Wei et al., 2025). Together, these findings underscore that the digital economy functions as a systemic force that reshapes factor mobility, institutional performance, and governance capacity across sectors.

Moreover, quasi-natural experimental studies further strengthen the causal interpretation of these relationships. Using the establishment of big data comprehensive pilot zones as an identification strategy, recent research confirms that digital economy development promotes high-quality economic growth through human capital accumulation and green technological innovation, with pronounced heterogeneity across regions, city sizes, and development stages (Guo et al., 2023). This strand of literature emphasizes that the impacts of the digital economy are mediated by internal mechanisms and conditioned by structural environments, rather than operating through simple linear effects.

Overall, the existing evidence suggests that the digital economy is not merely a sector-specific technological upgrade, but a multidimensional development system capable of influencing economic performance, environmental sustainability, and social outcomes through resource reallocation, innovation diffusion, and institutional transformation. These cross-sectoral mechanisms, including resource reallocation, innovation diffusion, and institutional upgrading, are also highly relevant to agriculture, where factor mobility and technology adoption are central to green productivity improvements (Qian et al., 2022; Qing and Chen, 2024; Zhang and Liu, 2023).

Agricultural green total factor productivity (AGTFP) has been widely used as a comprehensive indicator to assess the quality and sustainability of agricultural development, as it simultaneously incorporates desirable outputs and undesirable environmental impacts such as carbon emissions and non-point source pollution (Chen et al., 2022; Zhu et al., 2022). A growing body of literature has examined the determinants of AGTFP, indicating that agricultural green productivity is jointly shaped by factor endowments and allocation efficiency, technological progress and diffusion, as well as institutional and environmental conditions.

First, factor endowment structure and allocation efficiency constitute fundamental determinants of AGTFP. Existing studies suggest that labor input quality, capital accumulation, and land use efficiency are closely associated with agricultural green productivity (Han et al., 2018; Zhong et al., 2021). In particular, distortions in factor allocation and barriers to factor mobility tend to reduce production efficiency while intensifying environmental pressures, thereby constraining improvements in AGTFP (Liu et al., 2021). Optimizing the allocation of labor and capital across regions and sectors has therefore been identified as a critical pathway for enhancing agricultural green productivity.

Second, technological progress and the diffusion of agricultural technologies represent another key driver of AGTFP. Technological innovation not only improves production efficiency but also reduces resource consumption and pollution emissions through cleaner production methods and environmentally friendly input use (Zhou et al., 2023). Empirical evidence indicates that green technological innovation, mechanization, and the adoption of sustainable farming practices significantly enhance AGTFP; however, the effectiveness of technological progress depends critically on the speed and scope of technology diffusion, which varies across regions and stages of development (Fu and Zhang, 2022; Mei et al., 2022).

Third, institutional arrangements, environmental regulations, and external conditions also exert significant influences on AGTFP. Climate change, environmental constraints, and policy interventions affect agricultural green productivity through both direct and indirect channels. For example, adverse climate conditions—such as rising temperatures and increased precipitation variability—have been shown to exert nonlinear and regionally heterogeneous effects on China’s AGTFP (Song et al., 2022). At the same time, appropriate environmental regulations and agricultural support policies can incentivize cleaner production behavior and stimulate green technological innovation, thereby contributing to improvements in agricultural green productivity (Zhong et al., 2021; Zhou et al., 2023).

Overall, the existing literature indicates that AGTFP is the outcome of multiple interacting forces, including factor allocation efficiency, technological progress and diffusion, and institutional and environmental conditions. This body of research provides a solid foundation for examining how new development forces may further influence agricultural green productivity and underscores the importance of exploring the mechanisms through which such influences operate.

Building on the established determinants of AGTFP discussed above, recent studies have begun to explore digital rural development as a new development force that reshapes these traditional drivers rather than replacing them (Jingtao et al., 2024; Liu and Liu, 2024b). As an extension of the digital economy into agriculture and rural areas, digital rural development integrates digital infrastructure, digital services, and supportive institutional environments, thereby transforming agricultural production conditions and rural factor allocation systems.

Existing empirical evidence suggests that digitalization-related factors can significantly enhance AGTFP through multiple channels (Li et al., 2023; Zhao et al., 2022; Zhang et al., 2024; Wan et al., 2023; Ye, 2025). In particular, studies on digital inclusive finance show that improved access to digital financial services alleviates farmers’ credit constraints, promotes investment in green technologies, and enhances production efficiency, ultimately contributing to higher AGTFP (Gao et al., 2022). Similarly, recent research demonstrates that digital finance promotes AGTFP indirectly by fostering digital village development, highlighting the mediating role of rural digital infrastructure and service capacity in transmitting digital dividends to agricultural green productivity (Jiang et al., 2024).

Beyond financial digitalization, a growing strand of literature emphasizes the importance of comprehensive digital rural development in improving agricultural green productivity. Digital infrastructure, such as broadband networks and information platforms, reduces information asymmetry and transaction costs, thereby facilitating the more efficient allocation of labor, capital, and land across agricultural activities (Liu and Zhou, 2023; Chun et al., 2023; Zhai et al., 2023). Meanwhile, enhanced digital service capacity improves farmers’ access to market information, technical guidance, and public services, which in turn promotes organizational upgrading and the adoption of environmentally friendly production practices.

Digital rural development also plays a crucial role in accelerating the diffusion of green technologies and knowledge spillovers in agriculture. Through digital platforms and information networks, advanced agricultural technologies and sustainable farming practices can be disseminated more rapidly across regions, shortening the lag between innovation and adoption. This process not only enhances productivity but also mitigates environmental pressures by encouraging cleaner production methods and more efficient input use.

Taken together, existing studies suggest that digital rural development affects AGTFP primarily by reshaping factor allocation, facilitating organizational upgrading, and accelerating the diffusion of green technologies, although the strength and timing of these effects depend on regional development conditions and infrastructure endowments. In regions with insufficient digital infrastructure or limited human capital, the positive impacts of digitalization on agricultural green productivity may be constrained or delayed, implying the presence of nonlinear characteristics and regional heterogeneity.

Despite these contributions, existing studies exhibit several important limitations. First, much of the literature focuses on single dimensions of digitalization, such as digital finance, rather than conceptualizing digital rural development as an integrated system encompassing infrastructure, service capacity, and the broader development environment. Second, the internal mechanisms through which digital rural development influences AGTFP, particularly factor allocation, organizational upgrading, and technology diffusion, remain insufficiently examined. In addition, most existing studies rely on linear specifications and thus fail to capture potential threshold effects or stage-dependent dynamics, which may obscure the conditional nature of digital rural development’s impact on agricultural green productivity. In particular, when digital infrastructure and service capacity remain below critical levels, the enabling effects of digital rural development may not fully materialize. Accordingly, this study conceptualizes digital rural development as a systemic and conditional driver of agricultural green transformation rather than a technological input. By constructing a multidimensional digital rural development index and empirically examining its effects on AGTFP through mechanism analysis, threshold modeling, and regional heterogeneity analysis, this study seeks to extend the existing literature and provide more nuanced evidence on how digital transformation contributes to sustainable agricultural development.

Digital rural development—anchored in digital infrastructure, service capacity, and the development environment—facilitates the transformation and upgrading of agricultural production through multidimensional coordination, thereby enhancing agricultural green total factor productivity (AGTFP). The construction of digital infrastructure enhances information acquisition and transmission efficiency, reduces transaction costs in production and circulation, and improves resource utilization (Cen et al., 2022; Du et al., 2022). Enhanced digital service capacity allows farmers easier access to markets and policy resources, strengthening operational flexibility and supporting the transition toward precision and green production (Gao and Lyu, 2023). Meanwhile, a supportive digital development environment—characterized by stable energy supply, efficient transport, and industrial linkages—creates favorable conditions for applying digital tools and disseminating green practices (Zhang Q. et al., 2023; Zhang Z. et al., 2023).

Digital rural development encompasses multiple dimensions—digital infrastructure, service capacity, and development environment—and variations across these dimensions may generate nonlinear or stage-dependent effects. At early stages, weak infrastructure constrains information flow and factor mobility; once a threshold is reached, network externalities and scale effects amplify the benefits for AGTFP. By contrast, the effect of digital service capacity may display diminishing marginal returns—strong at initial stages when information barriers are reduced but weaker once systems mature. Finally, the role of the digital development environment is conditional: under weak institutional or physical conditions, improvements yield substantial gains in green efficiency, but the marginal impact declines as supporting conditions strengthen.

Agricultural Green Total Factor Productivity (AGTFP) serves as the core indicator of green agricultural efficiency in this study. Following Zheng et al. (2023), AGTFP is measured using the super-efficiency slack-based measure (SBM) model, which accounts for both desirable outputs (agricultural output value) and undesirable outputs (agricultural carbon emissions). The corresponding indicator system is summarized in Table 1.

Digital rural development is conceptualized as a multidimensional system integrating digital infrastructure, digital service capacity, and the digital development environment. Digital infrastructure provides the physical foundation for information transmission and factor mobility, encompassing broadband networks, mobile communication facilities, and related equipment. Digital service capacity reflects the degree to which digital tools, such as e-commerce, digital finance, and e-governance, translate infrastructure into productive capacity and enhance operational efficiency. In this study, digital rural governance is understood as an integral component of digital service capacity and the broader digital development environment, reflecting the application of digital technologies in public service delivery, information disclosure, and administrative coordination in rural areas. Due to data availability and the focus on development capacity rather than institutional performance, governance-related aspects are therefore not separately extracted as an independent dimension. The digital development environment captures the broader enabling conditions, including energy supply stability, logistics systems, data processing and collection capacity, and the regional digital industrial ecosystem, which together support technology diffusion and ensure the long-term sustainability of digital transformation. It should be noted that this dimension does not directly represent agricultural production activities; rather, it reflects the external infrastructural and industrial conditions under which digital technologies can effectively operate and be sustained in rural contexts. In this context, indicators reflecting logistics development and regional digital industry foundations are included to proxy the enabling environment, rather than to measure agricultural digitalization itself. Based on the above conceptual framework, the digital rural development index was constructed following a structured, three-step procedure. First, 13 indicators were selected to correspond to the three core dimensions (see Table 2). Second, all indicators were standardized to ensure comparability across provinces and over time. Third, the entropy-weight method was employed to objectively determine indicator weights based on the degree of information dispersion, thereby reducing subjectivity and redundancy in index construction. In addition, variance inflation factor (VIF) tests indicate that multicollinearity is not a concern (mean VIF = 1.63, all VIF values < 3). Regarding indicator direction, all indicators are intentionally defined as positive to capture the enabling capacity and development potential of digital rural development, consistent with mainstream practice in the literature (Pan et al., 2024; Zhu and Zhou, 2025). Potential imbalances or inclusiveness issues were therefore examined through heterogeneity, mechanism, and threshold analyses, rather than embedded directly into the index construction. The weighted indicators were then aggregated to obtain a composite index of digital rural development. In particular, indicators such as the added value of transportation, storage, and postal services, as well as the added value of information transmission, software, and information technology services, are used to capture the maturity of logistics systems and digital industry support, which are essential for the effective functioning of rural digitalization.

In this study, digital rural development is operationally defined as a composite, enabling-capacity index that captures the extent to which digital infrastructure, digital services, and supportive development environments jointly facilitate the application of digital technologies in rural areas. This operational definition is grounded in the theoretical perspective of digital empowerment and rural modernization, and is closely aligned with China’s policy framework on digital village construction, which emphasizes infrastructure upgrading, service integration, and institutional support as core pillars of rural digital transformation.

To identify the transmission mechanisms through which digital rural development influences agricultural green total factor productivity (AGTFP), this study draws on established theoretical frameworks on productivity and green development and identifies three primary transmission channels: factor reallocation, organizational upgrading, and technology diffusion.

Factor allocation efficiency is proxied by the land transfer rate and labor outflow. Existing studies indicate that land transfer plays a critical role in optimizing land factor allocation and promoting scale management, thereby improving agricultural productivity and efficiency (Deininger and Jin, 2005; Chen et al., 2011). Meanwhile, the reallocation of surplus rural labor has been shown to alleviate labor redundancy and enhance factor use efficiency through land-labor substitution and capital deepening (Benjamin, 1995; Yang and An, 2002). Accordingly, this study adopts these two indicators to capture the factor reallocation channel through which digitalization reduces transaction costs and information asymmetry, thereby facilitating large-scale and more efficient agricultural production.

Organizational structure is proxied by the level of large-scale operation and the number of agricultural cooperatives. Existing studies indicate that large-scale operation contributes to agricultural productivity and efficiency improvements by enhancing resource integration, reducing coordination and agency costs, and facilitating the adoption of modern production technologies (Foster and Rosenzweig, 2011; Wang et al., 2020). Meanwhile, agricultural cooperatives play an important role in organizational upgrading and collaborative production by improving farmers’ access to information, markets, and shared services, thereby strengthening collective action and economies of scale (Ito et al., 2012; Ma and Abdulai, 2016). Accordingly, this study adopts these two indicators to capture the organizational upgrading channel through which digitalization reduces coordination costs and improves organizational efficiency, thereby promoting more efficient and coordinated agricultural production.

Technology diffusion is proxied by the comprehensive mechanization rate of grain cultivation and the intensity of agricultural R&D investment. Existing studies indicate that agricultural mechanization plays a critical role in facilitating the adoption of modern and resource-saving production technologies, thereby improving production efficiency and promoting green agricultural development in China (Peng et al., 2022; Lu F. et al., 2024; Lu S. et al., 2024). Meanwhile, agricultural R&D investment has been shown to be a key driver of technological progress and technology diffusion in agriculture, enhancing productivity and sustainability through continuous innovation and knowledge spillovers (Jin et al., 2002; Lee et al., 2017). Accordingly, this study adopts these two indicators to capture the technology diffusion channel through which digital rural development accelerates the dissemination of advanced agricultural technologies and the adoption of green innovations, ultimately contributing to agricultural green total factor productivity.

To control for potential confounding influences of macroeconomic, social, and industrial factors, this study follows established empirical literature and incorporates a set of control variables (Wang L. et al., 2023; Wang Z. et al., 2023).

Openness to the outside world is measured by the ratio of foreign direct investment (FDI) to gross domestic product (GDP). Foreign direct investment can improve resource allocation efficiency by alleviating capital constraints, introducing advanced technologies, and generating technology spillover effects, thereby influencing agricultural productivity and green development (Wang J. et al., 2022; Wang X. et al., 2022).

Level of economic development is represented by GDP per capita, which reflects regional income levels and production capacity. Higher economic development provides better resource endowments and infrastructure conditions, forming an important foundation for productivity improvement and green transformation in agriculture (Zhang Q. et al., 2023; Zhang Z. et al., 2023).

Rural education level is calculated using a weighted education index of the rural population aged six and above, which serves as a common proxy for human capital accumulation. Improvements in rural education enhance farmers’ ability to adopt advanced technologies and environmentally friendly practices, contributing to higher agricultural efficiency and sustainability (Bai and Lei, 2020).

Agricultural planting structure is expressed as the ratio of grain-sown area to total crop-sown area, indicating the orientation of agricultural production structure. Planting structure affects factor allocation and resource use efficiency in agriculture and has important implications for green total factor productivity (Li et al., 2025).

Industrial structure is measured by the ratio of the value added of the secondary industry to that of the tertiary industry. Changes in industrial structure may indirectly influence agricultural green development by reshaping factor allocation and generating structural spillover effects across sectors (Zhao et al., 2025).

Incorporating these control variables helps isolate the net effect of digital rural development on agricultural green total factor productivity (AGTFP) by minimizing the influence of external economic, social, and structural heterogeneity.

To examine the mediating mechanism through factor allocation, Table 5 reports the estimated effects of digital rural development on land transfer and labor mobility, as well as their relationships with AGTFP. As shown in column (1) of Table 5, digital rural development significantly increases land transfer. Column (2) further shows that land transfer is negatively associated with AGTFP. Complementary evidence based on a land misallocation index (following Lei et al., 2022) indicates that rising land transfers have not yet alleviated land misallocation, which is consistent with institutional frictions that may generate excessive or inefficient transfers. As shown in column (3) of Table 5, digital rural development significantly promotes labor outflow, while column (4) indicates that labor outflow is positively associated with AGTFP, suggesting that digitalization facilitates more efficient labor reallocation and reduces redundant on-farm labor. Taken together, the inclusion of land transfer and labor outflow leads to a partial attenuation of the estimated effect of digital rural development on AGTFP relative to the baseline result reported in column (3) of Table 3. Consistent with the two-step mediation framework (Jiang, 2022), these results provide mechanism-consistent evidence that factor reallocation may serve as an important transmission channel, although institutional frictions in land markets may weaken its green efficiency effects.

To examine the mediating mechanism through organizational upgrading, Table 6 reports the estimated effects of digital rural development on the scale of agricultural operations and the number of rural cooperatives. As shown in column (1) of Table 6, digital rural development significantly increases the scale of agricultural operations, while Column (2) shows that the expansion of operational scale is positively associated with AGTFP. As shown in column (3) of Table 6, digital rural development significantly promotes the formation of rural cooperatives, whereas column (4) indicates that cooperative development is positively associated with AGTFP. Taken together, the inclusion of organizational variables leads to a partial attenuation of the estimated effect of digital rural development on AGTFP relative to the baseline result reported in column (3) of Table 3. Following Jiang (2022), this attenuation is interpreted as mechanism-consistent evidence rather than formal mediation identification. These findings support the view that digitalization facilitates organizational upgrading, resource pooling, and coordinated production, thereby enhancing the efficiency and sustainability of agricultural systems.

To examine the mediating mechanism through technology diffusion, Table 7 reports the estimated effects of digital rural development on the comprehensive mechanization rate and agricultural R&D intensity. As shown in column (1) of Table 7, digital rural development significantly increases the comprehensive mechanization rate, while column (2) indicates that mechanization is positively and statistically associated with AGTFP. As shown in column (3) of Table 7, digital rural development significantly promotes agricultural R&D intensity, whereas column (4) shows that agricultural R&D investment is positively associated with AGTFP. Taken together, the inclusion of mechanization and agricultural R&D variables leads to a partial attenuation of the estimated effect of digital rural development on AGTFP relative to the baseline result reported in column (3) of Table 3. Consistent with the two-step mediation framework, these results are interpreted as mechanism-consistent evidence of an indirect technology diffusion channel rather than a definitive causal decomposition. These findings highlight that digitalization enhances green agricultural productivity by promoting technological progress, innovation diffusion, and the adoption of modern agricultural practices.