Integrating policy design with agricultural emissions reductionin China: A multi-sector DSGE Approach

China's agricultural sector plays a pivotal role in the nation's emission reduction strategies while government policy design is crucial for guiding industrial abatement to balance environmental and food security goals. This paperaddresses the challenge of mitigating agricultural emissions while sustaining growth by employing a multi-sector Dynamic Stochastic General Equilibrium (DSGE) model to simulate the impacts of various emission reduction policies on agricultural output and emissions. Various solution techniques, such as perfect foresight, stochastic simulations, and rational expectation extended pathare utilized to examinehow policy stability and uncertainty differ in influencing economic agents’expectations, and thereby affecting sectoral production and emissions.China's Input-Output tables and Social Accounting Matrices are integrated to capture inter-sectoral spillovers impacting agriculture, providing a comprehensive analysis of cross-sectoral dynamics. The paper explores multiple policy scenarios, including different policy instruments, implementation approaches (ranging from fully predictable to highly uncertain), and combinations, evaluating their economic and environmental effects from 2025 to 2060. Key findings reveal significant variations in policy effectiveness: regulatory measures like taxes (e.g., emission taxes with agricultural exemptions) yield significant emissions reductions but suppress agricultural output due to increased production costs from carbon-intensive inputs;incentive-based subsidies (e.g., uniform emissions reduction subsidies) achieve the most effective long-term balance between emissions reduction and agricultural growth by incentivizing low-carbon technology investments;and market-based trading (e.g., emissions trading schemes with sector-specific caps) showsweaker results in both emissions reduction and growth, as producers may make irrational choices by avoiding initial investments in abatement technologies, instead relying on market trading to acquire permits or profits. Regardingimplementation approaches, stable and transparent designs outperform uncertain ones by reducing volatility and enhancing long-term outcomes.Combined policies integrating regulatory, incentive, and market-based mechanisms prove most effective, as they minimize economic disruptions and balance growth with emissions reductions, demonstrating the advantage of synergistic, predictable policy designs for agricultural decarbonization.中国农业部门在国家减排战略中扮演着至关重要的角色，而政府政策设计对引导产业减排、兼顾环境与粮食安全目标具有关键意义。本文针对农业减排与维持增长的双重挑战，采用多部门动态随机一般均衡（Dynamic Stochastic General Equilibrium, DSGE）模型，模拟各类减排政策对农业产出与碳排放的影响。本文采用完全预见（perfect foresight）、随机模拟（stochastic simulations）以及理性预期扩展路径（rational expectation extended path）等多种求解技术，探究政策稳定性与不确定性在影响经济主体（economic agents）预期，进而作用于部门生产与碳排放层面的差异机制。本文整合中国投入产出表（Input-Output tables）与社会核算矩阵（Social Accounting Matrices），以捕捉影响农业的部门间溢出效应，从而实现对跨部门动态机制的全面分析。本文探究了多类政策情景，涵盖不同政策工具、实施路径（从完全可预测到高度不确定）以及政策组合，并评估了2025年至2060年间各类情景的经济与环境效应。核心研究结果显示政策有效性存在显著差异：诸如税收类规制措施（regulatory measures，如针对农业豁免的碳排放税（emission taxes with agricultural exemptions））虽可实现显著的减排效果，但会因碳密集型投入品（carbon-intensive inputs）成本上升而抑制农业产出；而激励型补贴（incentive-based subsidies，如统一减排补贴（uniform emissions reduction subsidies））通过激励低碳技术投资（low-carbon technology investments），可在减排与农业增长之间实现最为有效的长期平衡；市场型交易机制（market-based trading，如分行业配额的碳排放权交易体系（emissions trading schemes））在减排与增长两方面的效果均较弱，原因在于生产者可能做出非理性选择：规避减排技术（abatement technologies）的初期投入，转而依靠市场交易获取配额或牟利。就实施路径而言，稳定且透明的政策设计通过降低波动性、优化长期效果，表现优于不确定性较强的政策设计。整合规制、激励与市场三类机制的组合政策被证明最为有效：此类政策可最大限度降低经济扰动，平衡增长与减排目标，凸显了协同化、可预期的政策设计对农业脱碳（agricultural decarbonization）的优势。