Unveiling the influence of agricultural mechanization on greenhouse gas emission intensity: Insights from China using causal machine learning model

Context

The ongoing agricultural mechanization in China contributes considerably to increased efficiency and productivity. However, its mitigation potential of greenhouse gas emissions intensity (GHGI) and underlying mechanisms are still unclear. Debate regarding the influence of agricultural mechanization on GHGI is gaining significance to guarantee that agricultural modernization corresponds with sustainable development goals.

Objective

This study attempts to provide quantitative evidence of agricultural mechanization on reducing GHGI from a fresh perspective of factor allocation.

Methods

The village-level dataset, involving 895 farmers across three northeastern provinces in China, was employed to confirm the theoretical analysis. The causal effect was estimated using a newly developed causal inference approach by incorporating double machine learning models. More explicitly, this research delved into the specific mechanism of factor allocation from three channels involving land transfer and operation scale, labor inputs, and agrochemical use intensity.

Results and conclusions

Empirical results show that GHGI generally decreases as agricultural mechanization increases. Among various field operations, the harvesting operation exerts the most significant influence on GHGI. The disparity in impact between mechanized plowing and mechanized harvesting is relatively minor. Further, the results of robustness tests using instrumental variable methods and recursive models strongly confirm the existence of this causal relationship.

Significance

Overall, these findings suggest advancing environmentally friendly mechanization and implementing targeted policies to reduce GHGI in agriculture along with enhanced factor provisions through optimizing allocation.