Traditional productivity theory views laborers, means of production, and objects of labor as the three fundamental elements constituting productivity (Sayers, 2007). Against the backdrop of multiple intertwining factors—including the accelerated advancement of the information technology revolution, significant shifts in the primary social contradictions, and profound changes in the economic development environment—traditional productivity theory has become increasingly ill-suited to meet the demands of the new era. To provide proper guidance for practical needs under the new circumstances, the concept of new-quality productivity has emerged. While it places greater emphasis on the leading role of scientific and technological innovation, the supporting role of data elements, and the driving role of emerging industries, and also stresses the symbolic significance of total factor productivity as a marker of new-quality productivity development, its fundamental essence remains rooted in these three basic elements (Xie et al., 2025). Therefore, when analyzing the direct impact of DIF on the new-quality productivity of the grain sector, the analysis should still be conducted from the three dimensions of labor force, means of production, and objects of labor.

First, from the perspective of labor force development, DIF promotes the development of new-quality agricultural productivity through two pathways: enhancing labor force quality and optimizing labor force structure. In terms of improving labor force quality, leveraging the inclusive nature of digital technology, DIF overcomes the geographical limitations of traditional financial services, providing financial support to rural “long-tail groups” (Zhang and Yu, 2025). This support manifests itself in two ways: first, by providing funding for professional skill training for grain producers, enabling them to master advanced production technologies and improve grain yield and quality; second, by offering specialized educational financial products to enhance access to educational resources for rural laborers and their descendants, thereby elevating overall human capital levels and laying the foundation for cultivating new-type professional farmers (Lu et al., 2022). In terms of optimizing the labor force structure, the grain production sector has long been constrained by natural conditions, market saturation, and insufficient policy support, coupled with weak rural infrastructure, leading to a single industrial structure and low production efficiency. The application of digital technology has enhanced financial institutions’ data processing capabilities, enabling them to accurately identify market demand and industrial potential. Additionally, data-driven precision services have driven the development of new industries such as rural e-commerce and leisure agriculture, extending the grain industry chain (Luo and Wang, 2022). This diversification of industries has led to an expansion in the scale of labor demand and structural differentiation, thereby achieving dynamic optimization of the labor force structure and sustained improvements in production efficiency.

Second, from the perspective of labor resources, DIF promotes the enhancement of new-quality productive forces in the grain sector through the following three pathways: first, replacing traditional infrastructure. Compared to traditional financial institutions that rely on physical branches and manual services, which face issues such as high costs, low efficiency, and insufficient coverage, DIF leverages digital infrastructure to build online comprehensive service platforms. This not only optimizes approval processes and risk control mechanisms but also breaks through geographical constraints, significantly reducing operational costs while expanding service coverage, thereby achieving cost reduction and efficiency improvements in financial services (Du et al., 2023). Second, it enhances agricultural mechanization levels. DIF utilizes cloud computing and big data technology to integrate multi-dimensional information such as farmers’ transaction records and credit histories, effectively addressing the financing challenges posed by the lack of collateral. Additionally, its online service platform provides efficient loan approval services, ensuring farmers can promptly access funds needed for agricultural machinery purchase and maintenance, thereby providing robust financial support for agricultural production (Wang et al., 2023). Third, it promotes agricultural technological innovation. Given the long cycle and high risk of innovation activities, DIF alleviates information asymmetry issues, relies on multi-dimensional data to accurately assess the credit status of innovation entities, and achieves precise capital allocation. At the same time, it establishes special loans and a capital supervision mechanism to ensure the effective use of innovation funds, providing continuous and stable financial support for agricultural technological innovation (Shen et al., 2024).

Finally, from the perspective of the labor object, DIF promotes the development of new-quality agricultural productivity through two pathways: optimizing land resource allocation and enhancing the modernization level of the industrial chain. In terms of land resource allocation, the accelerated urbanization process has intensified rural population migration, exacerbating land abandonment. However, remaining farmers are constrained by funding limitations, making it difficult to effectively advance land transfers, thereby hindering improvements in land utilization efficiency. DIF leverages big data analysis technology to provide farmers with financial support through unsecured credit financing models, effectively alleviating financing constraints in land transfers and promoting the intensive use of abandoned land, thereby laying a solid foundation of arable land resources for grain production (Shen et al., 2023). In terms of industrial chain upgrading, a financial information platform built on cloud computing, big data, and the Internet of Things integrates information resources across all links of the grain industrial chain, effectively addressing information asymmetry and financial suppression issues within the agricultural industrial chain. This platform-based service model not only achieves seamless integration across grain production, processing, distribution, and sales, driving vertical expansion and value enhancement along the supply chain, but also fosters the development of agricultural leading enterprises through precise financial support, strengthens brand building and market expansion, and promotes the grain industry’s transition toward high-quality, high-value-added development, thereby comprehensively enhancing the industry’s market competitiveness. Based on the above theoretical analysis, the article proposes the following research hypothesis:

The leap in new-quality productive forces relies not only on technological advancement itself but more critically on its deep integration with compatible forms of production organization (Eliasson, 1996). The fragmented, small-scale traditional agricultural production model has become a structural bottleneck constraining the development of new-quality productive forces in food production due to high transaction costs, weak risk resilience, and limited technology adoption capacity (Ntihinyurwa and de Vries, 2021). DIF effectively promotes agricultural intensification, establishing a critical organizational framework and scale foundation for the large-scale, efficient application of new technologies and factors. This process profoundly demonstrates the synergistic complementarity between technological progress and organizational evolution (Xiong et al., 2024). Specifically, DIF drives intensification through two dimensions: At the factor level, it directly alleviates farmers’ liquidity constraints in land consolidation and fixed-asset investment by providing targeted financial products such as land transfer deposit loans, scaled-operation start-up funds, and agricultural machinery leasing. This significantly reduces land fragmentation, promotes the optimized allocation and contiguous management of arable resources, and creates the physical prerequisites for agricultural mechanization and automation (Ruzzante et al., 2021). At the organizational level, it leverages innovative models such as supply chain finance, contract farming financing, and cooperative credit mutual aid to precisely empower new agricultural entities like family farms, farmer cooperatives, and leading agricultural industrialization enterprises. This not only improves their capital structure but also strengthens coordination and contractual stability across the industrial chain through embedded financial services. Essentially, this drives adaptive adjustments in agricultural production relations to align with the application of advanced agricultural technologies (Liu et al., 2025). This finance-driven intensification not only significantly reduces the unit costs of smart irrigation systems, precision fertilization devices, agricultural drones, thereby enhancing the economic viability of technology adoption. Moreover, the standardized production processes and unified management systems inherent in intensive operations greatly facilitate the collection, aggregation, and analysis of data throughout the entire agricultural production process. This provides indispensable application scenarios and a structured foundation for the deep integration and value realization of data as a new factor of production, as well as for AI-based intelligent decision-making and autonomous management (Zhao et al., 2023). Therefore, DIF promotes agricultural intensification by mitigating fragmentation. This not only optimizes the spatial and organizational allocation of traditional factors like land, labor, and capital but, more crucially, establishes an operational environment where technological advancement and data-driven approaches deeply integrate. Consequently, it systematically catalyzes a qualitative leap in food production efficiency, effectiveness, and quality, constituting one of the core mechanisms for developing new-quality productive forces. Based on the above theoretical analysis, this paper proposes the following research hypothesis:

The preceding theoretical analysis has elucidated the linear promotional effect of DIF on new-quality productive forces in agriculture and its mediating pathways. However, the Technology Diffusion Theory (Ghezzi et al., 2013) and the literature on the nonlinear effects of financial development (Botev et al., 2019) indicate that the impacts of emerging technology popularization and financial deepening are not simply linear. Instead, they depend on specific initial conditions or development stages and exhibit threshold characteristics. As a new financial model highly reliant on digital technology penetration and network externalities, the enabling effect of DIF on the new-quality productive forces in the grain sector may show phased changes with the variation of its own development level, regional infrastructure conditions or the capabilities of application subjects, which means a nonlinear relationship exists. From the perspective of technology adoption and network externalities, the effectiveness of DIF services is highly dependent on the penetration rate of digital infrastructure, users’ digital literacy and the network scale of participating entities. In the initial stage of development, limited coverage of infrastructure, low user acceptance and insufficient data accumulation lead to a small service scale and weak network effects. At this stage, DIF can only address the accessibility issue of financial services shifting from non-availability to availability, exerting limited effects on the optimal allocation of productivity factors with an insignificant marginal improvement effect. Once the critical threshold is crossed, with improved infrastructure, a user base reaching the tipping point and the initial formation of a data ecosystem, network externalities will become prominent. By reducing the unit cost of services, enhancing the accuracy of risk pricing and demand matching, and attracting the agglomeration of complementary resources, a positive feedback loop will be formed (Xu et al., 2025). This loop will significantly amplify the role of DIF in advancing technological innovation and promoting intensification, thereby accelerating its empowerment of the new-quality productive forces in the grain sector. From the perspective of factor matching and synergistic evolution, the upgrading of the new-quality productive forces in the grain sector is the result of the synergistic evolution of technological progress, organizational transformation and financial support (Shi et al., 2025). When the level of DIF is relatively low, financial support is fragmented with low standardization, which makes it difficult to systematically support substantive technological innovation and in-depth organizational transformation. Only when DIF develops to a relatively high level, featuring customized product design, embedded service models and intelligent risk control, can it meet the high-level demands of the new-quality productive forces in the grain sector for the precise drip irrigation and systematic empowerment of financial resources, thus unlocking greater potential (Li and Li, 2022). From the perspective of regional heterogeneity and absorptive capacity, differences in the traditional financial foundation, human capital stock and policy support intensity across regions constitute the absorptive capacity threshold (Lin and Peng, 2025). In regions with weak traditional financial foundations, DIF can quickly fill the service gaps, resulting in obvious marginal effects in the initial stage. In contrast, regions with advanced traditional financial systems need DIF to form complementary advantages or carry out differentiated competition. This requires DIF to possess more unique digital advantages or higher integration capabilities; otherwise, its effects may not be significant. Such heterogeneity also implies a nonlinear structure of the overall impact (Figure 1).

In summary, the impact of DIF on the new-quality productive forces in the grain sector is not a constant linear relationship but a threshold effect based on its own development level or external environmental conditions. Before reaching the specific threshold value, the promotion effect is moderate or limited. After crossing the threshold, the promotion effect will be significantly amplified due to the enhanced network effects, synergistic effects and absorptive capacity. Based on the above analysis, this paper proposes the following research hypothesis:

Below a specific threshold, its promotional impact is limited or insignificant; once this threshold is crossed, the promotional effect significantly intensifies.

In order to test the impact of digital financial inclusion on the new quality productivity of food, and at the same time taking into account the differences in geographic conditions, historical accumulation and other factors in different regions, the article adopts a double fixed effect model for parameter estimation. The constructed econometric model is as follows:

In Equation 1, i denotes the province, t denotes the year, Nqpg_i,t is the level of new food quality productivity in province i in year t, DIF is the Digital Inclusion Index, Controls is the set of control variables, η_i is the area fixed effects, δ_t is the time fixed effects, ε_i,t is the random perturbation term, α₀ is the constant term, and α₁ and α₂ are both parameters to be estimated.

In order to test research hypotheses 2 and 3, the article, on the basis of Equation 1, empirically tests the role path and mechanism of DIF affecting the new quality productivity of food, and constructs the following mechanism testing model with reference to Jiang’s (2022) research:

In Equation 2, Medi, s is the mechanism variable, including science and technology innovation (STI) and agricultural decentralization (AD). β0 and γ0 are constant terms, β1 is the parameter to be estimated, and the rest of the variables are defined in the same way as the baseline regression model.

In order to further explore the non-linear marginal effects of digital financial inclusion affecting the new quality productivity of food, the article adopts a threshold effect model to test the non-linear relationship; the model may have single or multiple thresholds, and the article only takes a single threshold as an example here. The constructed panel threshold model is as follows:

In Equation 3, Ι(·) is the schematic function, q_it is the threshold variable, η is the threshold value, φ₁ and φ₂ are the coefficients to be estimated, and the rest of the variables have the same meaning as in model (1).

The explanatory variable of the article is grain new-quality productivity (Nqpg). Referring to the existing studies on the construction of an indicator system to comprehensively measure the new quality productivity of food (Li et al., 2024; Wang and Wang, 2024), the three elements of productivity, namely, laborers, labor objects and labor materials, are used as the first-level indicators to measure the new quality productivity of food. At the same time, considering the availability and desirability of data, the article firstly measures the individual dimension of workers from the level of labour skills, efficiency and awareness; secondly, it measures the labour objects from the level of technological progress, food security, and green development; and finally, it measures the labour materials from the three aspects of infrastructure, energy consumption, and scientific and technological innovation, with reference to the relevant studies on the specific indicator system and its measurement method (Yang et al., 2024; Wang et al., 2025) as shown in Table 1. In order to comprehensively examine the indicators of grain new quality productivity, the article adopts the entropy value method to measure the index of the development level of grain new quality productivity in each province (district and city).

The article adopts Peking University’s DIF Index (DIF) as the core explanatory variable, which is both vertically and horizontally comparable, and can reflect the level of development of DIF in different regions at different times, with high reliability and validity. In order to standardize the scale, the article divides the DIF index by 100, which is used to characterize the final core explanatory variables.

The article selects science and technology innovation (STI) and agricultural decentralization (AD) as mechanism variables. Among them, STI is measured by the ratio of internal expenditure on R&D funding to regional GDP. This indicator is widely used to characterize the overall intensity of regional investment in science and technology innovation (Lettice et al., 2012). The underlying logic is that R&D investment serves as the source of knowledge creation and technological advancement; higher investment intensity typically signifies more active innovation activities and faster technological accumulation. In the agriculture and food industry sectors, increased overall regional R&D investment can drive progress in areas such as agricultural biotechnology, intelligent agricultural machinery, food processing techniques, and supply chain management technologies through channels including knowledge spillovers, human capital enhancement, and industry-academia-research collaboration (Sharif, 2021). Therefore, although this indicator is not exclusive to the agricultural sector, it captures the macro R&D environment and foundational innovation capabilities that underpin technological innovation and transformation across the entire food value chain (from production to processing). It represents a key regional innovation factor driving the development of new-quality productive forces in food production.

Agricultural decentralization (AD) is measured by the ratio of regional per capita primary industry output to national per capita primary industry output. The rationale behind this indicator is as follows: if a region’s per capita agricultural output is significantly below the national average, it typically indicates a more prevalent agricultural production model characterized by small-scale, fragmented operations, with relatively low land use efficiency and production concentration. Conversely, regions where per capita agricultural output approaches or exceeds the national average often exhibit higher levels of scaled-up and intensive farming practices (Gomes et al., 2019). DIF can promote the optimized allocation and concentration of production factors like land by providing financing for land transfers and supporting new types of agricultural operators, thereby enhancing per capita output efficiency. Consequently, changes in this ratio can indirectly reflect the role of DIF in mitigating agricultural fragmentation and advancing intensive farming practices.

To accurately identify the net impact of DIF on new-quality agricultural productivity and control for other potential factors that may influence the results, this study selected the following set of control variables based on existing literature on agricultural productivity and financial development (Lu et al., 2025; Tan et al., 2025; Lu et al., 2025):

This paper employs factor analysis to recalibrate the weights of the new-quality grain productivity index, measuring the comprehensive scores of new-quality grain productivity (denoted as Nqpg_f) for each province (autonomous region, municipality) over the years. This adjusted score is then reintroduced as the explanatory variable into the regression model for parameter estimation. The regression results are presented in Column (1) of Table 4. Based on the results in the table, it can be observed that DIF consistently exerts a positive influence on new-quality grain productivity, passing the 1% significance level test. Therefore, the test results after replacing the core explanatory variable support the research conclusions drawn from the baseline regression analysis presented earlier.

To mitigate potential impacts of extreme values on estimation results, parameter estimation was repeated after applying 1% forward and backward truncation to all variables. As shown in Column (2) of Table 4, the impact of DIF on new-quality agricultural productivity still passed the 1% significance level test, indicating the robustness of the research conclusions.

Due to the truncation feature of the index of the level of new quality productivity of food measured using the entropy method, in order to overcome the possible estimation bias, the article chooses the Tobit model to re-run the regression analysis. The regression results in column (3) of Table 4 show that the promotion effect of digital financial inclusion on new quality productivity of food is still significant at the 1% significance level, a result that is consistent with the baseline regression conclusions and further validates the robustness of the study’s conclusions.

Although the article has controlled for relevant economic variables as well as constructed a two-way fixed effects model for regression analysis as much as possible to avoid biased estimates due to omission of important variables, there may be a causal relationship between DIF and Nqpg, which may lead to endogeneity problems. Therefore, the article selects the cross-multiplier term (Tele) of the number of national fixed-line telephones in 2000 and the number of national Internet users in the previous year as the instrumental variable, and uses the two-stage least squares (2SLS) method for parameter estimation. The instrumental variables are chosen because, on the one hand, the penetration of fixed telephones and the size of Internet users can reflect the level of regional informatization infrastructure, which is correlated with the development of digital financial inclusion and satisfies the correlation condition of instrumental variables; on the other hand, the historical data of the number of fixed telephones and the number of Internet users serve as exogenous variables, which are not subject to the inverse influence of the current period’s development of digital financial inclusion and are difficult to influence the explanatory variables directly through other channels to directly affect the explanatory variables, satisfying the exclusivity constraint condition. Meanwhile, considering that the number of year-end landline telephone subscribers in 2000 is cross-sectional data, which cannot be directly applied to the endogeneity test of the panel model, the article conducts the endogeneity test by constructing its interaction term with the number of national Internet subscribers in the previous year as an instrumental variable. The results are shown in column (4) and column (5) of Table 4. The digital financial inclusion index after considering the endogeneity issue still passes the 1% significance level test, further validating the robustness of the research findings.

Due to the significant spatial heterogeneity among Chinese regions in terms of resource endowment, technological innovation capacity, and policy environment, the development process of DIF and its policy effects may show obvious regional differences. In order to examine this heterogeneous feature in depth, the article divides the sample into two sub-samples of main grain-producing areas and non-main grain-producing areas based on the regional division of labor pattern of grain production and conducts regression analyses separately in order to reveal the regionally differentiated performance of the impact of digital financial inclusion on the productivity of new grain quality. According to the results in Table 8, the promotion effect of DIF on new quality productivity of food exhibits statistical significance in both main grain producing areas and non-main grain producing areas, but there is obvious regional heterogeneity in the intensity of its impact. Specifically, the marginal effect is significantly higher in main food-producing regions than in non-main food-producing regions. This difference may be due to the fact that, first, food-producing regions usually have better agricultural infrastructure and large-scale production conditions, which enable digital financial tools to be more effectively embedded in the whole process of agricultural production. In contrast, the decentralized business model in non-food-producing regions may limit the depth of penetration and breadth of application of digital financial inclusion to some extent. Second, major food-producing regions tend to have a more mature agricultural industrialization system, including a complete industrial chain and stable market channels. This industrial ecology creates favorable conditions for the in-depth integration of digital financial services and agricultural production factors, thus more fully releasing its productivity-enhancing effects. In addition, regional differences in policy support are also important influencing factors. The main food-producing regions usually enjoy greater financial subsidies and financial policy tilts, and this institutional environment creates policy synergies with DIF, further amplifying its role in promoting new food productivity.

Considering that the level of traditional financial development can often effectively measure the degree of regional financial deepening, reflecting the maturity of traditional financial services to the real economy, and providing benchmarks for further analyzing the “complementary effect” or “substitution effect” of digital financial inclusion, identifying the optimal synergy boundary between digital finance and traditional finance. It provides a benchmark for further analyzing the “complementary effect” or “substitution effect” of DIF, and identifies the optimal synergy boundary between digital and traditional finance. Therefore, the article adopts the ratio of the loan balance of financial institutions to GDP to measure the development level of traditional finance in each region. Regions with traditional finance development level higher than its median are categorized into one group, and those lower than its median are categorized into another group, and group regression analysis is carried out. The results in Table 9 show that in regions with high levels of traditional finance, DIF has a dampening effect on the new quality productivity of food, but its estimated coefficient is not statistically significant, whereas in regions with less developed levels of traditional finance, it shows a significant boosting effect and passes the test of 1% significance level. The reason for this phenomenon may lie in the fact that the developed traditional financial regions have formed a perfect credit service system, and the intervention of DIF may lead to the duplication of financial resources and the crowding out effect. When the target groups of the two types of financial services are highly overlapping, it exacerbates excessive competition among financial institutions and reduces the efficiency of resource allocation. On the contrary, in regions with a low level of traditional financial development, due to the lack of traditional services, DIF can break through the geographical limitations, directly fill the service vacuum, more accurately meet the financing needs of marginalized groups, and open up the “last kilometer” of digital financial services.