The escalating global average temperature intensifies extreme weather, agricultural disasters, and biodiversity loss, largely due to greenhouse gas emissions. China's commitment to carbon neutrality by 2060 necessitates addressing agricultural carbon emissions, a significant contributor to its total emissions (16-17%, potentially reaching 30% by 2050). Finance plays a crucial role in agricultural development, influencing resource allocation and potentially impacting carbon emissions. While some research suggests financial development increases emissions by lowering financing costs and expanding production, other studies highlight its potential to suppress emissions through the adoption of environmentally friendly practices and technological advancements. A non-linear relationship, an inverted U-shape, has also been observed. Traditional finance faces limitations in supporting low-carbon agricultural innovation due to factors like scale discrimination and low efficiency. Digital inclusive finance, leveraging digital technology to improve the affordability, comprehensiveness, and sustainability of financial services, offers a novel pathway to reduce agricultural carbon emissions. This study addresses the gap in research by exploring the impact of digital inclusive finance on agricultural carbon emissions, considering the role of rural business entities (particularly agricultural green cooperatives) and the moderating effect of human capital.

Existing research reveals a complex relationship between finance and carbon emissions. Some studies find that financial development stimulates emissions by fueling economic growth and energy consumption. Others show that it can suppress emissions by promoting green production methods and technological progress. A nonlinear, inverted U-shaped relationship has also been noted, potentially due to a decoupling between financial and agricultural development as the financial system matures. Regarding digital finance, studies show its potential to reduce carbon emissions by promoting innovation, restructuring industries, and optimizing resource allocation. However, the research largely focuses on spatial distribution and intermediary effects like entrepreneurship and technological progress, neglecting the role of digital inclusive finance on rural business entities. This study aims to bridge this gap.

This study uses data from 30 Chinese provinces (excluding Tibet, Hong Kong, Macau, and Taiwan) from 2011 to 2021. Agricultural carbon emissions were calculated using the emission coefficient method, incorporating emissions from fertilizers, pesticides, agricultural films, diesel fuel, electricity for irrigation, and soil tillage. The digital inclusive finance index from Peking University's Digital Finance Research Center was used, analyzed across dimensions of coverage breadth, usage depth, and digitization level. A two-way fixed-effects model was employed to examine the relationship between digital inclusive finance and agricultural carbon emissions, including the quadratic term of the digital inclusive finance index to capture nonlinearity. A moderated effects model was used to analyze the moderating role of agricultural green cooperatives. A panel threshold model, utilizing human capital accumulation as the threshold variable, was applied to investigate potential threshold effects. Control variables included urbanization rate, degree of openness, transportation infrastructure, and the proportion of disaster-affected crop area.

China's agricultural carbon emissions exhibited an inverted U-shaped trend from 2011 to 2021, peaking around 2015 and then declining. Spatial analysis revealed higher emissions in the east and north compared to the west and south, with the east-west gap narrowing and the north-south gap widening over time. Regression analysis confirmed a nonlinear, inverted U-shaped relationship between digital inclusive finance and agricultural carbon emissions. The impact is primarily driven by the usage depth and digitization level of digital inclusive finance; when usage and digitization levels are below a threshold, emissions tend to increase, then decrease above it. The analysis revealed a decoupling between the development of digital inclusive finance and agricultural green cooperatives, suggesting no mutually enhancing effect. The threshold regression analysis using human capital accumulation as the threshold variable showed that the suppressive effect of digital inclusive finance on agricultural carbon emissions strengthens significantly as human capital crosses threshold values.

The findings support the Environmental Kuznets Curve hypothesis, showing that initially, digital inclusive finance may boost economic activity and increase emissions before its positive impact on sustainable practices and technological advancements leads to emission reductions. The moderating effect of agricultural green cooperatives highlights the importance of complementary policies supporting both financial inclusion and the development of sustainable agricultural businesses. The threshold effect based on human capital suggests that investments in education and training are essential for maximizing the emission-reduction potential of digital inclusive finance. This underscores the need for targeted policies and interventions aimed at enhancing the capacity of farmers and agricultural businesses to effectively utilize digital financial tools and adopt sustainable practices.

This study provides valuable insights into the complex interplay between digital inclusive finance, agricultural green cooperatives, human capital, and agricultural carbon emissions in China. It highlights the nonlinear impact of digital finance and the importance of considering contextual factors like the development of green cooperatives and human capital accumulation for effective carbon emission reduction strategies. Future research could expand the geographical scope, explore longer time spans, and delve deeper into the mechanisms underlying the interactions between digital finance and sustainable agricultural development.

This study's limitations include the focus on 30 provinces in mainland China from 2011 to 2020, excluding other regions and potentially influencing regional heterogeneity. The analysis of agricultural carbon emissions is relatively narrow, focusing primarily on the planting industry. Future research could benefit from incorporating a broader range of agricultural activities and extending the sample period and geographical scope. The study also relies on existing data for the digital inclusive finance index, without direct measurement of this index's underlying data, which could potentially affect the result. Furthermore, deeper exploration of the mechanisms driving the interaction between digital finance and agricultural green cooperatives is needed.