Abstract:Integrating crop and livestock production is an important agricultural practice that aligned with the goals of green agricultural development and sustainability. However, existing studies often overlook the material exchange between livestock and crop production in evaluation of agricultural eco-efficiency, resulting in an inability to accurately evaluate agricultural eco-efficiency and the mechanisms for its enhancement. This paper constructs an evaluation framework of agricultural eco-efficiency, incorporating the interlinkages of the crop-livestock system through "chain" perspective. By dividing the crop-livestock system into two mutually coupled subsystems, a network Data Envelopment Analysis (DEA) model based on Slack-Based Measure (SBM) is employed to measure the efficiency of each component of the crop-livestock system across 29 provinces and cities in China from 2000 to 2021. The findings indicate that the aggregate eco-efficiency of China's integrated crop-livestock systems exhibits an upward trajectory yet stays at a comparatively low level, averaging 0.5975. This suggests that while progress has been made, significant room for improvement still exists for the eco-efficiency in China's integrated crop-livestock systems. Regionally, the northeastern region exhibits the highest eco-efficiency in integrated crop-livestock systems, followed by the eastern region, while the central and western regions show relatively lower eco-efficiency. Both livestock and crop systems contribute positively to improving the eco-efficiency of the integrated crop-livestock systems, with livestock systems making a relatively larger contribution to the overall efficiency enhancement. Several key factors contribute to the efficiency losses observed in the four major regions of China. The primary reason for the efficiency losses is the redundant labor inputs in grain production, highlighting an urgent need to enhance mechanization in Chinese agriculture as well as promote the transformation and substitution of agricultural input factors. Additionally, redundancies in fertilizer inputs and the emissions of livestock manure pollutants are significant contributors to eco-efficiency losses in integrated crop-livestock systems. Scenario analysis further reveals that the integrated crop-livestock systems can reduce redundant inputs of fertilizers and feed within the crop-livestock cycle, thereby decreasing non-desired outputs and enhancing agricultural eco-efficiency. Specifically, after integration, fertilizer input redundancy decreases from 48.55% to 23.22%, feed input redundancy decreases from 34.93% to 22.94%, and non-desired output redundancy decreases from 53.39% to 29.39%. Given these findings, future efforts should focus on optimizing the layout of crop-livestock industries based on local agricultural resource endowment, enhancing the integration between crops and livestock, and improving the circular industry chain of crop-livestock systems to promote the green transformation and sustainable development of agriculture.