In the Chinese context, rural entrepreneurship extends beyond traditional agriculture to encompass a broad range of industrial and commercial activities. Typically organized around family-based units, rural entrepreneurship is characterized by the mobilization of entrepreneurial capital and capabilities to identify opportunities, reallocate resources, innovate business models, and pioneer emerging sectors (Xiong and Huang, 2022; Li X. et al., 2022). This dynamic not only enhances economic diversification but also expands employment opportunities and increases rural incomes. Effective rural entrepreneurship thus plays a pivotal role in driving local economic transformation and unlocking the developmental potential of rural regions (Wei et al., 2023).

Digital technologies have increasingly become pivotal in transforming rural entrepreneurial activities, offering profound impacts across various dimensions, including opportunities, costs, and risks associated with entrepreneurship (Zhao et al., 2023; Samudra, 2023). Notably, digitalization facilitates the emergence of innovative business models and operations (Ajide and Osinubi, 2023; Li, 2023; Li Y. et al., 2022), exemplified by initiatives like rural Taobao and direct-to-consumer sales models. This evolution, coupled with the integration of online and offline channels, broadens the horizon of entrepreneurial ventures, unlocking new opportunities and bolstering entrepreneurial engagement in rural areas. On the dimension of costs and risks (Wanming et al., 2023), advancements in rural infrastructure alongside the pervasive use of internet and mobile platforms significantly streamline transactional processes and information searches for financial services, thereby reducing associated costs (Ozgen, 2022). Big data platforms enhance transparency around enterprise credit information, mitigating risks linked to information asymmetry. Furthermore, the application of big data analytics substantially lowers the barriers to risk assessment in rural entrepreneurial ventures, rendering a more conducive environment for business operations. Consequently, digital technologies not only furnish a robust platform for production and sales but also curtail entrepreneurial costs and risks, thereby energizing rural entrepreneurship. The dynamism in rural entrepreneurship, fueled by increased opportunities and reduced costs and risks, catalyzes employment creation, fosters integration across agricultural and other industrial sectors, and augments rural incomes. This synergy cultivates a self-reinforcing cycle that propels the rural economy forward (Li X. et al., 2022; Xiong and Huang, 2022). Based on this analysis, we posit our first hypothesis:

Agricultural innovation is inherently multidimensional, with existing scholarship identifying five key domains: the restructuring of agricultural practices, novel recombination within agricultural systems, synergistic effects arising from the transformation of production factors, systemic innovations in agricultural management, and challenge-driven innovation in response to sector-specific constraints (Zhao et al., 2020).

The infusion of innovative factors and the stimulation of external market demand are pivotal in catalyzing innovation activities. These dynamics foster investments in research and development, elevate the modernization of agriculture, and consequently, facilitate RI (Li et al., 2023). Three primary mechanisms underlie this process: Firstly, the proliferation of digital factors increases the availability of novel inputs and leverages digital technologies to heighten capital allocation efficiency (Cui and Wang, 2023). This enhancement allows rural finances to exert a more substantial impact, augmenting the efficiency of modern agricultural production and fostering a more integrated development of rural tertiary industries (Hu et al., 2023). Secondly, the expansion of digital factors, coupled with the adoption of digital technologies, bolsters the level of rural human capital and innovation efficiency. This dynamic propels technological advancements in agriculture, contributing to the deeper development of RI (Wu et al., 2021). Thirdly, the escalation in digital factor inputs diversifies consumption patterns, sparking the emergence and adoption of new consumption methods and technological innovations. These innovations stimulate activities that upgrade the industrial structure and amplify the level of RI (Yang and Wang, 2022). Based on these observations, we articulate our second hypothesis:

This study focuses on 30 provincial-level administrative regions in mainland China from 2013 to 2021. Tibet, Hong Kong, Macao, and Taiwan are excluded due to limitations in data availability and institutional comparability. Specifically, Tibet’s unique geographical and agricultural conditions, along with differences in statistical standards, impede data consistency. Hong Kong and Macao, as special administrative regions with minimal agricultural contributions to GDP and distinct statistical systems, are not included. Data from Taiwan remain inaccessible due to institutional constraints.

The research sample encompasses key actors in RI: agricultural enterprises, professional cooperatives, family farms, newly emerging professional farmers, and smallholder farmers. These entities represent the principal agents driving the convergence of agriculture with secondary and tertiary industries.

Agricultural enterprises serve as core coordinators, functioning simultaneously as organizers of the industrial chain, enhancers of the value chain, and architects of the benefit-sharing mechanism. Professional cooperatives facilitate centralized production and modern agricultural management, playing a pivotal role in industrial organization. Family farms, as the smallest operational units, bridge the upstream demands of enterprises and cooperatives with the heterogeneity of individual households. New professional farmers act as catalysts of innovation, embedding technological capabilities into rural industrial systems and accelerating industrial upgrading. Smallholder farmers constitute the foundational base upon which all integrative processes rely.

The operationalization of RI, the core variable of our study, draws upon the comprehensive framework proposed by Li and Xu (2020). This framework, informed by an extensive review of the literature and a nuanced understanding of RI’s multifaceted nature (Hao et al., 2023; Cui et al., 2022; Li and Liu, 2021), constructs an indicator system that captures the essence of RI from several dimensions. These dimensions include: Firstly, integration subjects. This category encompasses entities such as agricultural professional cooperatives, family farms, new professional farmers, agricultural enterprises, and ordinary farmers, highlighting the diverse actors involved in RI. Secondly, the integration environment. This dimension reflects the contextual factors that facilitate or hinder RI, including policy and financial environments, infrastructure quality, and market demand. Thirdly, integration methods. It looks at the strategies employed to achieve RI, focusing on industrial chain extension and the promotion of service industry leadership. Fourthly, integration dynamics: This aspect examines the forces driving RI forward, such as technological innovation, diffusion, and industrial restructuring. Finally, integration effects. The outcomes of RI, including improvements in agricultural efficiency, rural prosperity, and farmer income, are assessed under this dimension.

Digital factors, representing a unique category of production resources, are pivotal in driving productivity advancements (Li and Liu, 2021). Recognizing their role as distinct production factor forms, this study constructs an indicator system for digital factors based on their multifaceted impact on the digital economy. This system is informed by the prevailing definitions and classifications within core digital economy sectors, encompassing a broad spectrum of dimensions: Firstly, digital infrastructure captures the foundational elements of digitalization, including hardware, software, information network facilities, and the level of logistics infrastructure, all of which are essential for enabling digital activities (Kai et al., 2022). Secondly, digital technology encompasses the adoption and application of advanced information networks, information and communication technologies (ICTs), and internet finance solutions, highlighting the technological underpinnings of digital transformation. Thirdly, digital efficiency evaluates the effectiveness of digitalization in agriculture, measuring the extent of agricultural digitization, digital transactions of agricultural products, and digital investments in the agricultural sector. Finally, digital product services assess the availability and consumption of digital services, including the expertise of digital professionals, consumption patterns of digital products, and the provision of digital technology services.

To quantitatively assess rural entrepreneurship, this study adopts a multifaceted measurement framework encompassing key indicators of entrepreneurial scope and success. These include enterprise performance, land transfers exceeding 10 acres, annual agricultural incomes surpassing CNY 50,000 (Xiong and Huang, 2022), participation in individual or private enterprises (Li and Xu, 2020), and operational scales exceeding four times the per capita arable land (Liu and Qian, 2022). Recognizing the predominance of individual businesses and private enterprises in China’s rural entrepreneurial landscape—and the limited availability of disaggregated data on private investors—we use employment figures from these entities as a proxy for entrepreneurial activity. Specifically, we define rural entrepreneurship as the ratio of employment in private enterprises and individual businesses to the total rural population (Ji et al., 2023), offering a novel and practical metric for capturing entrepreneurial dynamics. While this approach reflects broad patterns of formal entrepreneurship, it does not capture informal or unregistered entrepreneurial activities, which remain significant but are absent from official statistics.

To quantify agricultural innovation, prior studies have employed a range of indicators, including innovation efficiency, total factor productivity in agriculture (Wu et al., 2021), investment in agricultural R&D, the number of agricultural invention patents granted, and the introduction of new crop varieties (Yang and Wang, 2022). These metrics capture diverse dimensions of the agricultural innovation landscape—from tangible outputs such as patents and new varieties to broader systemic measures such as productivity and efficiency. While informative, each indicator reflects only a subset of the innovation process and may not fully capture the scope and complexity of innovation activities across the sector. To address this, our analysis employs agricultural patents as a proxy for the level of agricultural innovation, guided by two primary considerations. First, the utilization of patent data offers a high degree of comparability across different regions within the same country due to uniform patent authorization regulations. This consistency ensures that the comparison of patent activities across domestic regions is both valid and robust. Second, despite the diversity of patent types, including invention patents and utility models, and the absence of a specific international classification standard exclusively for agricultural patents, the international patent classification system’s Class A provides a relevant aggregation of patents in agriculture, forestry, animal husbandry, and related fields. Therefore, this study quantifies agricultural innovation by analyzing the proportion of invention and utility model patents granted in the agricultural sector relative to the total number of patents issued. Agricultural innovation is driven by a diverse set of actors, and not all forms of innovation are captured through formal metrics such as patent filings. Process innovations, organizational improvements, and non-technological advances often fall outside the scope of patent systems. Moreover, innovations arising from small-scale entrepreneurs and individual farmers are typically informal and seldom documented in official statistics, leading to a systematic underestimation of innovation activity within the sector.

The study covers the period from 2013 to 2021. Data on RI were drawn from a broad array of official sources, including the Rural Business Management Statistical Yearbook (2014–2018), the China Rural Cooperative Economy Statistical Yearbook (2019–2021), and other specialized yearbooks and sectoral reports that capture diverse dimensions of China’s rural economy—from policy frameworks to financial indicators and intellectual property. Supplementary data were obtained from the National Bureau of Statistics and provincial statistical yearbooks to ensure comprehensive coverage.

Digitalization indicators for the same period were primarily derived from the Peking University Digital Inclusive Finance Index, the China Statistical Yearbook, and regional statistical publications. Where necessary, gaps were filled using data from the National Bureau of Statistics, the Ministry of Commerce, and relevant research reports. Data on rural entrepreneurship were compiled from official statistics released by the National Bureau of Statistics, while agricultural innovation metrics were sourced from the China National Intellectual Property Administration’s patent database.

The operationalization and measurement of all variables are detailed in Table 1, reflecting the study’s emphasis on methodological transparency and its commitment to accurately capturing the complexity of rural transformation processes.

This study investigates the mechanisms by which digital factors facilitate the integration of RI, with particular attention to the pathways through which digitalization enables this process. To elucidate these mechanisms, we adopt a mediation analysis framework, which allows us to examine not merely whether digital factors exert an influence, but how and why they do so. Unlike traditional regression models that focus on establishing the presence or magnitude of an effect, mediation models decompose the total effect of an independent variable (X) on a dependent variable (Y) into direct and indirect components. The indirect pathway captures the influence of X on Y via an intermediary variable (M), thereby uncovering latent mechanisms that may be obscured in standard causal inference. This analytical approach helps mitigate the risk of misattribution by explicitly modeling the processes underlying observed relationships. In this context, we identify rural entrepreneurship and agricultural innovation as key mediating variables linking digitalization to RI.

This modeling framework quantitatively captures both the direct and indirect effects of digital factors on RI, offering a structured approach to disentangle the complex interactions among digitalization, rural entrepreneurship, and agricultural innovation. The analytical model is specified through Equations 1–3 as follows:

In this model, we delineate the relationships between digital factors, rural entrepreneurship, agricultural innovation, and RI through a series of equations. Each equation is designed to test a specific aspect of our hypothesis: Equation 1 assesses the direct impact of digital factors on RI, aiming to establish a foundational understanding of how digitalization contributes to rural industrial development. Equation 2 examines the influence of digital factors on rural entrepreneurship and agricultural innovation. This step is crucial for understanding the mechanisms through which digitalization fosters an environment conducive to entrepreneurial and innovative activities in the rural sector. Equation 3 evaluates the effects of rural entrepreneurship and agricultural innovation on RI, thereby elucidating the mediating role these variables play in the digital transformation of rural industries.

In these equations: i denotes the province, t denotes the year; RI , DF , RE and AI correspond to the constructs detailed in Table 1, ε represents a stochastic disturbance term to account for unobserved heterogeneity and random shocks.

Pearson correlation analysis indicates minimal risk of multicollinearity, with all coefficients below 0.5—except for a strong correlation (r = 0.859) between digital infrastructure and RI. Using data from 2013 to 2021 and Equation 1, the national-level results (Table 2, column 1) suggest that digitalization significantly promotes the integration of rural industries. To address potential endogeneity, we apply a GMM estimator, using rural broadband access as an instrumental variable. The results show no evidence of significant endogeneity.

Applying Equations 2, 3, we further find that digital infrastructure plays a critical role in fostering rural entrepreneurship (Table 2, columns 2 and 3), reinforcing the view that digitalization is a key driver of rural economic vitality. Two primary mechanisms emerge from this analysis: Firstly, the infusion of rural entrepreneurship catalyzes the creation of new industries, business models, and formats, thereby amplifying the impact of digitalization on RI. Secondly, rural entrepreneurship contributes to the enhancement of human capital in rural areas, further facilitating RI development.

Column 4 of Table 2 reveals that digital factors significantly influence agricultural innovation, suggesting that advancements in digital technology are pivotal in fostering innovation within the agricultural sector. This influence is characterized by the facilitation and development of agricultural innovation, underscoring the integral role of digitalization in modernizing rural industries. Further examination in column 5 of Table 2 demonstrates that agricultural innovation significantly contributes to the enhancement of RI through digital factors. The impact of digitalization on agricultural innovation and RI unfolds in two critical dimensions: firstly, by accelerating innovation through the dissemination of digital knowledge, thereby expediting the pace of industrial development. Secondly, the synergy between digital advancements and innovation optimizes resource allocation, significantly promoting the integration and advancement of rural industries (Kim, 2023; Vasilchenko et al., 2023; Steinke et al., 2021).

To examine regional heterogeneity in the impact of digitalization on the integrated development of rural industries, we adopt a tripartite regional framework based on the official classification introduced in the 1986 “Seventh Five-Year Plan.” This division—eastern, central, and western China—captures both geographic and economic distinctions and has been reinforced through major national strategies, including the Western Development Strategy and the Rise of Central China Plan. It thus provides a robust basis for comparative analysis. In parallel, we consider the categorization of “main grain-producing areas,” defined administratively through a multidimensional assessment of resource endowment, production scale, commercialization level, export capacity, and strategic relevance to national food security. This policy-oriented classification encompasses 13 provinces: Hebei, Inner Mongolia, Liaoning, Jilin, Heilongjiang, Jiangsu, Anhui, Jiangxi, Shandong, Henan, Hubei, Hunan, and Sichuan, serving both agro-ecological and strategic planning purposes.

While the preceding section presents an empirical analysis from a regional perspective, the following section shifts focus to a multidimensional framework, dissecting the roles of digital infrastructure, technology level, efficiency, and product services. Our results demonstrate that digital infrastructure, technology level, efficiency, and product services significantly enhance RI, albeit with variations in the magnitude of their impacts (Table 5). Notably, digital infrastructure and product services emerge as the most influential dimensions. This suggests that the foundational aspects of digitalization, alongside the availability and quality of digital products, are crucial drivers of rural industrial development. Further comparative analysis across these dimensions confirms that all four digital factors—infrastructure, technology level, efficiency, and product services—positively affect rural entrepreneurship and agricultural innovation, with statistical significance at the 1% level. This underscores the comprehensive role of digitalization in fostering an environment conducive to entrepreneurial and innovative activities within rural settings.

Incorporating these digital factors alongside rural entrepreneurship and agricultural innovation into a unified model reveals a persistent positive impact on RI, although the magnitude of this effect diminishes slightly, as indicated by the comparison with the baseline coefficients in Table 5, column 1. This reduction in coefficients suggests that while digital factors directly influence RI, their impact is further mediated and enhanced through the stimulation of rural entrepreneurship and agricultural innovation. This dimensional analysis underscores the multifaceted influence of digital factors on RI, highlighting the essential role of rural entrepreneurship and agricultural innovation as mediators in this process. It emphasizes the need for a holistic approach to digitalization that considers infrastructure, technology, efficiency, and product services as key drivers of RI.

Our analysis, from the perspective of mediating effects, reveals that agricultural innovation exerts a stronger mediating influence on the relationship between digital factors and RI than rural entrepreneurship. This distinction underscores the pivotal role of innovation in agriculture as a key driver of digital transformation in rural industries. When comparing the mediating effects across various digital dimensions, we observe distinct patterns for rural entrepreneurship and agricultural innovation. For rural entrepreneurship, the mediating effects follow a descending order, starting with digital efficiency, followed by digital product services, digital infrastructure, and finally, digital technology. This sequence highlights the relative importance of digital efficiency in enhancing the impact of rural entrepreneurship on RI. Conversely, the mediating effect of agricultural innovation is characterized by a different order of significance: digital efficiency leads, succeeded by digital infrastructure, digital product services, and digital technology. This order emphasizes the critical role of digital efficiency and infrastructure in amplifying agricultural innovation’s contribution to RI. These findings illuminate the nuanced ways in which rural entrepreneurship and agricultural innovation serve as conduits for the influence of digital factors on RI. The variance in their mediating effects across digital dimensions suggests that targeted strategies might be required to fully leverage digital transformation in rural settings.

In assessing the robustness of our findings, we employ panel quantile regression, a methodology chosen for its several key advantages. Unlike traditional regression models, panel quantile regression does not presuppose a normal distribution of the dependent variable, offering resilience against outliers and eliminating the need to address heteroscedasticity directly. This approach provides detailed insights by outputting regression coefficients and measures of fit, facilitating a comprehensive analysis of the relationship between independent and dependent variables across the distribution. The panel quantile regression serves two primary purposes in our analysis. It allows us to trace the varying impact of independent variables on the dependent variable across different points in the distribution, offering a nuanced view of these relationships. In addition, it enables us to rigorously test the robustness of our regression model (Yang and Chen, 2022), ensuring the reliability of our results. To this end, we focus on quantiles at the 25th, 50th, and 75th percentiles, providing a broad assessment of the model’s performance across the distribution.

The robustness testing, detailed in Table 6, underscores the varying impact of digital factors on RI across different quantiles. Notably, digital factors consistently exert a positive influence on RI, evidencing their pivotal role across the spectrum of rural industrial development stages. For rural entrepreneurship, our analysis reveals a nuanced relationship with RI. At the 25th quantile, rural entrepreneurship significantly promotes RI, with an estimated coefficient of 0.2457. However, as we move towards the 75th quantile, this effect diminishes, becoming non-significant with an estimated coefficient decreasing to 0.0684 and further to 0.0225. This trend suggests that the contribution of rural entrepreneurship to RI is more pronounced in provinces with lower levels of integration, diminishing in significance as the level of RI increases. Conversely, the influence of agricultural innovation on RI presents an opposite trend. Initially, at the 25th quantile, the effect of agricultural innovation is not significant, with an estimated coefficient of 0.0658. Yet, as the quantile progresses to 0.75, the impact becomes significantly positive, with the coefficient increasing to 0.2763. This indicates that agricultural innovation’s role in enhancing RI becomes increasingly significant in provinces with higher levels of industrial integration.

These trends may reflect diminishing marginal returns and path dependence in the process of RI. In the early stages, integration is constrained primarily by shortages of production factors, underdeveloped markets, and fragmented industrial chains. At this point, entrepreneurial activity plays a catalytic role: it fills structural gaps, introduces new market participants and resources, and stimulates industrial dynamism. The contribution of entrepreneurship to integration is thus both direct and substantial.

However, as integration advances and industrial chains mature, the constraints shift. The focus moves from establishing basic linkages to upgrading value chains, enhancing productivity, and fostering competitiveness. Under these conditions, the proliferation of new enterprises alone contributes less to integration. Incumbent firms, fortified by capital and technological advantages, raise entry barriers, limiting opportunities for new entrants. The marginal impact of primary entrepreneurial activity diminishes, while high-quality entrepreneurship—centered on value creation, efficiency gains, and sustainability—becomes increasingly vital. This marks the entry point for agricultural innovation.

Yet innovation is often subject to delayed returns. In early integration stages, weak infrastructure, limited human capital, underdeveloped institutional frameworks, and restricted market demand hinder the adoption and diffusion of innovation. Consequently, its impact remains muted. Once a higher level of integration is achieved, however, the situation changes. More robust industrial systems, stronger factor markets, and increased absorptive capacity enable innovation to be scaled through multiplier effects, yielding increasing marginal returns.

In sum, rural entrepreneurship serves as a “disruptor” and “growth catalyst” during the nascent phases of integration, accelerating progress through enhanced market connectivity and resource mobilization. As integration deepens and competition intensifies, its marginal returns naturally wane. Conversely, agricultural innovation—initially constrained by systemic limitations—emerges as a “quality and efficiency enhancer” and “value creator” in more advanced stages, delivering transformative gains and exhibiting significant marginal contributions.

The robustness testing across the eastern, central, and western regions, as presented in Table 7. In the eastern region, the results closely mirror national trends, affirming the general pattern. In the central and western regions, however, the effect of rural entrepreneurship on RI shifts from non-significant to strongly positive across higher quantiles, indicating its growing relevance in driving industrial development in these areas.

By contrast, the role of agricultural innovation diverges across regions. In the central region, its influence on RI diminishes—significant at lower quantiles but fading at higher ones—suggesting an early-stage contribution that weakens as integration progresses. In the western region, agricultural innovation shows no significant effect across all quantiles, underscoring the limited role of innovation in shaping industrial integration in this context. These findings highlight distinct regional trajectories in how entrepreneurship and innovation contribute to rural industrial transformation.

In the eastern region, high labor and land costs have eroded the profitability of labor-intensive entrepreneurship, accelerating a shift towards high-tech, innovation-driven models. The region’s mature market division of labor further limits the viability of new, homogeneous ventures, thereby reducing their marginal contribution to RI. However, the eastern region’s advanced technological base facilitates closer alignment between research and application—what might be termed a seamless integration between the “laboratory” and the “farmland.” High-income consumer segments continue to pay premiums for quality agricultural products, incentivizing agribusinesses to invest in technological innovation and amplifying the region’s integration momentum.

In contrast, the central and western regions have increasingly absorbed agricultural processing activities from the east, gradually strengthening local industrial chains. Infrastructure initiatives, such as universal broadband access, have lowered barriers to entry and catalyzed the rapid growth of rural e-commerce and other entrepreneurial models, thereby enhancing marginal returns. Nevertheless, agricultural innovation in these regions has underperformed. Limited capacity for technology transfer and application has constrained its impact, resulting in only modest contributions to rural industrial development.

Our analysis, from the perspective of main versus non-main grain-producing areas, reveals a progressive increase in the impact of digital factors on the integrated development of rural industries as we move across higher quantiles. This trend underscores a consistent and growing influence of digitalization on enhancing RI, highlighting the pivotal role of digital advancements in driving the economic development of rural areas, irrespective of their agricultural production focus.

As shown in Table 8, our analysis within main grain-producing areas reveals a nuanced relationship between rural entrepreneurship, agricultural innovation, and the integrated development of rural industries across varying levels of RI. As we examine higher quantiles, the influence of rural entrepreneurship on RI demonstrates a notable increase, with the estimated coefficient rising from 0.0247 to 0.0907. Initially non-significant, the impact becomes positively significant at higher quantiles, highlighting that rural entrepreneurship exerts a stronger effect in provinces characterized by advanced levels of RI. Conversely, the trajectory of agricultural innovation’s influence on RI presents a contrasting pattern. Starting with a high estimated coefficient of 0.3446, indicating a strong positive impact on RI, the coefficient decreases to 0.1716 at higher quantiles. This trend suggests that while agricultural innovation significantly contributes to the development of RI, its relative impact diminishes in provinces with more developed rural industries.

Analyzing the impact within non-main grain-producing regions, we observe a distinct pattern in the relationship between rural entrepreneurship, agricultural innovation, and the development of RI across various quantiles. The estimated coefficient for rural entrepreneurship on RI decreases from 0.4401 to 0.1211 as we progress to higher quantiles. Initially significant, this influence transitions to non-significant, underscoring that rural entrepreneurship is particularly impactful in provinces with less advanced levels of RI. In contrast, agricultural innovation consistently exerts a significant positive effect on RI across all quantiles, although the magnitude of this impact diminishes from 0.2666 to 0.1806. This decreasing trend in the estimated coefficient suggests that while agricultural innovation remains a critical driver for RI, its relative influence is more pronounced in provinces with lower levels of RI, decreasing as the level of integration advances.

In major grain-producing regions, the core objective of industrial integration is primarily to reduce costs, enhance efficiency, ensure stable supply, and bolster the competitiveness of the grain sector—rather than to foster diversification or novel business formats. Entrusted with the responsibility of national food security, these regions exhibit a relatively rigid agricultural production structure. As a result, industrial chain extension and business model innovation are constrained by stronger policy boundaries and path dependence. Demand for high value-added innovation within large-scale grain systems remains limited.

At advanced stages of integration, readily attainable technological innovations in grain production may approach saturation or encounter bottlenecks, reducing their marginal returns. Breakthrough innovations, while potentially transformative, are hampered by long development cycles, high risk, and delayed payoff. Furthermore, these regions are typically characterized by high land turnover rates and the dominance of large-scale farming enterprises. In such contexts, entrepreneurship that enhances service provision, logistics, and cost-efficiency aligns more closely with regional needs, offering immediate and practical gains.

In contrast, non-major grain-producing areas are marked by more diversified agricultural structures and lower barriers to entry for new ventures. In early integration stages, this fosters a surge in low-threshold, homogeneous entrepreneurship. However, as integration deepens, market saturation, intensified competition, and diminishing returns set in. At higher levels of integration, the marginal contribution of conventional entrepreneurship declines sharply.

Enhancing competitiveness in these regions thus hinges on innovation-led differentiation and value addition. The observed decline in marginal gains at high percentiles likely reflects the exhaustion of easily diffused innovations and the slow maturation of more advanced ones. While higher-level innovations demand greater investment and longer time horizons, their absolute contributions to rural industrial transformation remain substantial and irreplaceable.

As displayed in Table 9, our analysis across different digital dimensions reveals nuanced effects on the development of RI. Digital infrastructure emerges as a consistent positive driver for RI across varying quantiles, indicating its foundational role in facilitating rural economic development. The impact of rural entrepreneurship on RI demonstrates a significant to non-significant shift as quantiles increase, with the estimated coefficient depicting a downward trend. This pattern is observed not only in the context of digital infrastructure but also regarding the level of digital technology, suggesting that while rural entrepreneurship initially contributes to RI, its influence wanes at higher levels of integration. Conversely, from the perspectives of digital efficiency and digital product services, rural entrepreneurship exhibits a pronounced positive effect on RI, underscoring the importance of these digital aspects in empowering rural entrepreneurs. Regarding digital infrastructure and product services, agricultural innovation significantly propels RI forward, with the estimated coefficient indicating an upward trend as quantiles increase. This highlights the critical role of innovation in agriculture, particularly supported by robust digital infrastructure and services. Similarly, the influence of agricultural innovation on RI, viewed through the lenses of digital technology and efficiency, transitions from non-significant to significantly positive, with a gradually increasing coefficient. This suggests the growing importance of agricultural innovation in enhancing RI, especially as digital technology and efficiency improve.

Our comparative analysis of quantile results across different regions and digital dimensions consistently underscores the significant positive influence of digital factors on the development of RI. This analysis reveals that while the impact of rural entrepreneurship on RI exhibits variability across quantiles, it consistently contributes positively to RI development. Similarly, agricultural innovation demonstrates a significant forward-direction influence on RI, underscoring its pivotal role in enhancing rural industries. The variability in the degree of impact across quantiles and contexts notwithstanding, both rural entrepreneurship and agricultural innovation exhibit clear positive effects on RI. These findings affirm the robustness of our results, indicating that digital factors, rural entrepreneurship, and agricultural innovation are integral to the advancement of RI, irrespective of regional or dimensional specificities.

At early stages of RI, investments in digital infrastructure can rapidly yield returns, catalyzing entrepreneurial activity and market participation. However, as integration deepens, the marginal stimulus provided by additional infrastructure diminishes. Reliance solely on infrastructure and general-purpose technologies risks entrenching low-level, undifferentiated competition, insufficient to meet the demands for value-added differentiation characteristic of advanced integration stages.

High-tech entrepreneurship—such as intelligent breeding or drone-enabled crop protection—requires substantial fixed capital investment (e.g., agricultural IoT systems, satellite remote sensing) and specialized expertise, posing significant barriers for most rural entrepreneurs. Consequently, technological reserves often fail to translate into entrepreneurial competitiveness. At low levels of integration, these advanced innovations remain largely unviable due to limited supporting infrastructure. In contrast, at high levels of integration, improvements in digital infrastructure—such as widespread 5G coverage and the deployment of agricultural big data platforms—amplify the marginal returns to innovation, as reflected in increasing coefficients.

Digital technologies directly reduce transaction costs and inventory risks, while digital services offer accessible pathways to value creation. Industrial service innovations generate new consumption scenarios and unlock untapped markets. Platform-based tools further lower technological thresholds, enabling ordinary entrepreneurs to engage with high-value ecosystems. However, realizing the full potential of digital innovation requires advanced absorptive capacity, such as data analytics expertise for deploying digital twins. At low integration levels, the scarcity of human capital limits such capabilities. At higher levels, the accumulation of skills and experience—facilitated by “learning-by-doing” dynamics—overcomes these barriers, resulting in a significant and positive impact on entrepreneurial performance.

This study, leveraging data from 2013 to 2021, examines the role of digital factors in enhancing RI, with a specific focus on the mechanisms of rural entrepreneurship and agricultural innovation. Our comprehensive analysis spans national, regional, and various digital dimensions, employing intermediary effect models and quantile regression methods to elucidate the dynamics at play.

Firstly, our analysis identifies digital factors as key enhancers of rural entrepreneurship and agricultural innovation, with the magnitude of their influence exhibiting significant regional and sectoral variations. Specifically, the impact of digital factors on rural entrepreneurship and innovation is notably strongest in the eastern regions, reflecting a more pronounced digital transformation in these areas. In contrast, the central and western regions experience a relatively weaker influence, suggesting a need for targeted digital infrastructure and policy support to bridge the regional digital divide. Within the agricultural sector, the influence of digital factors on rural entrepreneurship shows minor differences between main and non-main grain-producing areas, with non-main areas experiencing slightly greater benefits. This indicates a more uniform digital penetration in the entrepreneurial domain across agricultural contexts. However, when it comes to rural innovation, main grain-producing areas enjoy a more significant boost from digital factors, highlighting the role of digital technologies in enhancing innovation in regions with intensive agricultural production.

Secondly, our findings elucidate the mediating roles of rural entrepreneurship and agricultural innovation in the process through which digital factors facilitate RI. The impact of these mediating factors varies across different regions and agricultural contexts. In terms of the mediation by rural entrepreneurship, we observe a decreasing trend from the eastern regions to the western and central regions. This indicates a stronger mediating effect in the eastern regions, diminishing progressively westward. Conversely, the mediation by agricultural innovation shows a different pattern, being most pronounced in the central regions, followed by the eastern and western regions. This suggests regional disparities in how agricultural innovation influences the digital transformation of rural industries. When contrasting main grain-producing areas with non-main grain-producing areas, the intermediary effect of rural entrepreneurship is more significant in non-main grain-producing regions. This points to a greater role of entrepreneurship in mediating digital impacts in areas with less intensive grain production. On the other hand, the mediating effect of agricultural innovation is more substantial in main grain-producing areas, highlighting the importance of innovation in agriculture for digital empowerment in regions focused on grain production.

Thirdly, our robustness testing further substantiates the significant positive influence of digital factors on the development of RI. While rural entrepreneurship consistently demonstrates a positive influence on RI, this impact exhibits variability across different quantiles. Despite these fluctuations, rural entrepreneurship maintains a significant forward-moving impact on RI development. Agricultural innovation, akin to rural entrepreneurship, plays a substantial role in enhancing RI. Our findings indicate that its impact, much like that of rural entrepreneurship, varies across quantiles, reinforcing the complexity and multifaceted nature of these influences. The robustness tests across different regions reveal divergent trends in the influence of rural entrepreneurship and agricultural innovation on RI. As quantiles increase, these influences show varying trajectories, with some regions even exhibiting non-significant impacts. This highlights the heterogeneity in the effectiveness of these factors across different regional and sectoral contexts.

Fourthly, our analysis across different digital dimensions reveals that digital infrastructure, technology level, efficiency, and product services all substantially foster RI in a positive manner. Each digital dimension—infrastructure, technology level, efficiency, and product services—demonstrates a significant and positive role in promoting RI. This underscores the broad and multifaceted influence of digitalization in facilitating rural economic development. Additionally, these digital factors exert a significant positive impact on both rural entrepreneurship and agricultural innovation. Their enhancement of these areas further contributes to the advancement of RI, highlighting the interconnected nature of digital factors and rural economic activities. Importantly, rural entrepreneurship and agricultural innovation act as intermediary forces in the digital empowerment process. Their mediating roles vary depending on the specific digital dimension in question. For rural entrepreneurship, the order of mediating effects descends from digital efficiency to digital technology. For agricultural innovation, the sequence is similarly characterized, starting with digital efficiency and ending with digital technology. This dimensional analysis reveals that while all digital factors positively impact RI, the extent of their influence through rural entrepreneurship and agricultural innovation varies. The nuanced mediating effects across these dimensions suggest that targeted strategies might be required to fully leverage the potential of digitalization in different aspects of rural development.

This study extends existing theoretical frameworks by critically addressing several identified gaps in the literature concerning the dynamics of RI driven by digitalization. First, previous research largely treated digitalization’s impact on rural economies in broad strokes, often neglecting regional and sector-specific heterogeneity (Zhang L. et al., 2023; Zhang Z. P. et al., 2023; Lei et al., 2025). By explicitly examining the nuanced variations across regional contexts—particularly the distinct patterns observed between eastern, central, and western regions—and differentiating the mediating roles of rural entrepreneurship and agricultural innovation, our study provides a more granular theoretical understanding of digital technology’s varied impacts.

Second, the conventional metrics used in prior research—primarily educational attainment or general infrastructure indices—have been critiqued as insufficient for fully capturing the complexities inherent in rural entrepreneurial and innovative capacities (Fahmi and Savira, 2023; Wang X. et al., 2023; Wang R. et al., 2023). Our study advances theory by incorporating a mediation analysis that explicates the causal pathways linking digitalization with RI through the specific mechanisms of entrepreneurship and innovation. This approach not only enriches the theoretical narrative but also underscores the necessity of using more precise indicators that reflect digital technology’s multidimensional influences.

Third, our findings challenge the assumption that digital technology uniformly empowers rural economies by revealing significant variations in the intermediary effects across main and non-main grain-producing areas. The nuanced interplay identified in our study between digital dimensions (infrastructure, technology level, efficiency, and services) and RI further deepens theoretical understanding by highlighting how specific digital components differentially catalyze entrepreneurial and innovative activities (Müller and Korsgaard, 2018; Zhu et al., 2024).

Lastly, this research contributes to the evolving discourse on spatial spillovers in RI by empirically substantiating the critical roles played by both rural entrepreneurship and agricultural innovation as mediators influenced by digital technology. This reinforces the theoretical proposition that regional policies aimed at digital enhancement must carefully consider local entrepreneurial ecosystems and innovation capacities to effectively harness digital potential for rural revitalization (Ye, 2025; Hao et al., 2023).

Collectively, these theoretical contributions underscore the complexity and multifaceted nature of rural digitalization’s impact, urging future research to adopt similarly nuanced analytical approaches to advance the theorization of rural economic transformations.

The integration of advanced digital technologies—such as artificial intelligence and big data—has increasingly permeated the agricultural sector, unlocking new avenues for rural economic transformation. In this era of digitalization, traditional production factors are undergoing a fundamental reconfiguration, with digital technologies emerging as novel and strategic inputs that reshape production relationships. This study highlights the transformative potential of digitalization in rural contexts, demonstrating how digital infrastructure and tools underpin innovative entrepreneurial activity and catalyze the development of modern rural industries.

First, digitalization exerts a significant and positive impact on RI, not only yielding economies of scale but also enhancing marginal returns. As enablers of modernization, digital technologies inject advanced elements into the rural economy, improving resource allocation, streamlining traditional economic structures, and fostering the emergence of new industries. Their adoption drives the evolution of business formats and models, illustrating the capacity of digital tools to reconfigure rural industrial landscapes (Guo et al., 2023; Hellemans et al., 2022; Leng and Tong, 2022). Empirical evidence confirms that digital technologies are foundational to rural industrial development, underpinning productivity gains, economic diversification, and the creation of new industrial paradigms.

Second, digital technologies enhance access to information and capital, providing the infrastructure necessary to support rural entrepreneurship. This digital empowerment plays a pivotal role in attracting a new wave of rural entrepreneurs, typically comprising two groups: returnee migrant workers equipped with financial and operational experience, and highly educated individuals—such as university graduates—who, though less experienced, possess the adaptive capacity and technical fluency to leverage digital tools effectively.

Finally, the interplay between digitalization and agricultural innovation further amplifies rural industrial development. Digital technologies not only facilitate the management of agricultural information and the deployment of digital platforms but also act as catalysts for systemic innovation across the sector (Ge et al., 2022; Cumming et al., 2022; Reza-Gharehbagh et al., 2023). These findings underscore the strategic imperative of integrating digital technologies with agricultural innovation to advance the agenda of sustainable RI.

Drawing from the key findings of our study, we outline the following policy recommendations to optimize the integration of digital factors in rural industrial development.

Firstly, enhancing digital infrastructure for RI. It is crucial to bolster the digital infrastructure foundational to modern agricultural development. This involves accelerating the deployment of technologies like big data, the Internet of Things, and comprehensive information infrastructure, particularly focusing on agricultural production needs. Special attention is required to address challenges in remote areas, such as low population density and high construction costs, which hinder the widespread adoption of basic digital services like mobile internet. Policies should aim for an equitable expansion of digital technology to these areas, ensuring regional, urban–rural, and cross-industry information interconnectivity. In addition, enhancing the capacity of cold chain logistics infrastructure is vital for reducing losses of agricultural products. Developing a robust cold chain data platform that integrates production, temperature control, and product information can provide valuable services to farmers. Such a platform would facilitate informed decision-making, helping farmers avoid misjudgments about supply and demand, ultimately leading to increased income. These recommendations are aimed at leveraging digital advancements to foster sustainable and inclusive growth within rural industries, addressing specific challenges and harnessing the potential of digital technologies for rural development.

Secondly, enhancing agricultural digital transformation to boost RI. Central to this initiative is the role of leading agricultural enterprises in facilitating digital integration throughout the industrial chain. By deepening the application of digital technology, consolidating digital resources, and establishing shared digital platforms, these enterprises can drive both upstream and downstream entities towards comprehensive digital adoption, thereby guiding new agricultural management entities in their digital transformation journey. The digitalization of agriculture relies heavily on advancing key technologies, including information and biotechnology. Enhancing these digital technologies is crucial for a successful agricultural transformation. This involves bolstering the agricultural science and technology innovation system, aiming to utilize digital transformation as a catalyst for improving the quality, efficiency, and dynamics of RI. The digital transformation of agriculture encompasses a wide array of factors: from elevating total factor productivity to ensuring the sustainable development of the entire ecological industry, and addressing challenges in industrial chains. Strengthening advancements in these areas will unleash new potentials and functionalities in the digital transformation process, propelling the agricultural sector towards a more efficient and sustainable future. These policy recommendations are designed to foster a digital ecosystem in agriculture that not only enhances the sector’s productivity and sustainability but also serves as a cornerstone for the broader integration and advancement of rural industries.

Thirdly, advancing RI through enhanced information platforms. Strengthening digital information platforms is paramount. These platforms should be leveraged to provide real-time, accurate data on market dynamics like product prices, sales, and demand. By minimizing agricultural supply inefficiencies and facilitating timely adjustments in agricultural production, these platforms can synchronize demand and supply, refine marketing strategies, and create synergies between production, transactions, and services. The ultimate goal is to maximize the value of agricultural products and ensure a better alignment with market needs. Another critical aspect is the integration of agricultural management information systems with cutting-edge technologies. By making advanced agricultural knowledge and digital tools widely accessible to farming and management entities, we can significantly enhance the convenience and accuracy of information. This approach not only reduces operational risks but also fosters a more cohesive and integrated rural industry sector. The emphasis here is on using digital technology as a tool to bring together diverse strands of agricultural knowledge and practice, thereby streamlining processes and enhancing efficiency within the rural industrial ecosystem. These recommendations aim to harness the power of digital information platforms and integrated systems to catalyze the transformation and upgrading of the agricultural sector, positioning it as a vital component in the broader landscape of RI.

Fourthly, enhancing the ecosystem for rural entrepreneurship and agricultural innovation. Essential to the development of RI is the creation of a supportive environment for rural entrepreneurship. National policies should focus on providing conditions that facilitate ease of business for rural entrepreneurs. This includes support for farmers to directly manage or engage in cooperative ventures within rural industries. Encouraging local and nearby entrepreneurship among farmers can significantly broaden their income opportunities and contribute to the economic vibrancy of rural areas. In addition, innovation in agricultural products and business formats is critical for reshaping the agricultural sector. Strategies aimed at extending the agricultural value chain and introducing novel business models are pivotal. These innovations can lead to a more diversified and sustainable rural industrial landscape, enhancing the overall resilience and dynamism of rural economies. Implementing these policy measures will not only promote the integration and development of rural industries but also ensure their sustainable evolution in line with modern technological and entrepreneurial advancements.

While the use of GMM estimation helps mitigate endogeneity concerns in the main effects, we acknowledge that mediation models may still be subject to reverse causality, particularly between RI and the mediators of rural entrepreneurship and agricultural innovation. Given the complex, potentially bidirectional relationships involved, the causal pathways identified in this study should be interpreted with caution. Future research employing panel structural equation modeling or instrumental variable strategies within the mediation framework may help further isolate these dynamic effects.