关键词: 农业经济高质量发展, 产业升级, 碳减排, 空间马尔科夫链, 面板向量自回归模型

Abstract:

[Objective] Coordinating high-quality agricultural economic development, industrial upgrading, and carbon-emission reduction has become an inherent requirement for promoting the low-carbon transformation of agriculture and advancing agricultural modernization under China’s “dual-carbon” goals. As a fundamental sector that safeguards national food and ecological security, agriculture faces the challenge of balancing economic growth with environmental sustainability. Existing studies have mainly focused on pairwise relationships among economic development, industrial upgrading, and carbon emissions, while overlooking the systematic interactions, feedback mechanisms, and dynamic evolution of these three dimensions. In practice, the coordinated advancement of high-quality agricultural economic development and industrial upgrading may be constrained by carbon-reduction targets, whereas the effect of high-quality agricultural development on carbon-emission reduction may depend on the degree of agricultural industrial upgrading. Neglecting the interdependence among these three dimensions and pursuing the optimization of individual components may result in a “fallacy of composition,” thereby undermining the efficiency and coordination of the overall system. Therefore, this study aimed to investigate the spatiotemporal evolution, transition characteristics, and dynamic interaction mechanisms among high-quality agricultural economic development, agricultural industrial upgrading, and agricultural carbon-emission reduction. [Methods] Based on panel data from prefecture-level cities in China from 2012 to 2023, this study constructed a three-dimensional analytical framework integrating high-quality agricultural economic development, agricultural industrial upgrading, and agricultural carbon-emission reduction. A coupling coordination degree model, traditional and spatial Markov chain models, and a panel vector autoregression (PVAR) model were employed to systematically investigate the spatiotemporal evolution, transition dynamics, and interactive response mechanisms of the three subsystems. Specifically, the coupling coordination degree model was used to characterize temporal trends and spatial differences in coordination among the three systems. The traditional and spatial Markov chain models were applied to identify state-transition characteristics, path dependence, and neighborhood effects in the evolution of coupling coordination. Furthermore, the PVAR model was adopted to examine the dynamic interactions and impulse-response relationships among high-quality agricultural development, industrial upgrading, and carbon-emission reduction. [Results] The coupling coordination level among high-quality agricultural development, industrial upgrading, and carbon-emission reduction increased continuously nationwide during the study period, indicating that synergistic development gradually improved under policy guidance and endogenous transformation pressures. Nevertheless, the overall coordination level remained relatively low, with most regions remaining between the stages of “mild imbalance” and “near imbalance,” suggesting substantial room for improvement. Significant disparities were observed among China's four major regions. Eastern China maintained the highest coupling coordination level because of its advantages in economic development, technological innovation, and factor agglomeration. Northeastern China ranked second, benefiting from large-scale agricultural production and a high level of mechanization. Central China exhibited strong catch-up momentum, driven by industrial transfer and policy support. By contrast, although Western China experienced gradual improvement, it continued to lag because of constraints associated with resource endowments, technological capacity, and its economic foundations. Overall, the spatial pattern of coupling coordination was characterized by higher levels in the east and lower levels in the west. The analysis of transition dynamics further revealed pronounced path dependence and spatial spillover effects in the evolution of coupling coordination. The traditional Markov transition matrix indicated that the coupling coordination states were highly stable over time, suggesting that short-term leapfrogging between development stages was difficult. Transitions occurred mainly between adjacent coordination states, reflecting a gradual evolutionary process rather than abrupt transformation. The spatial Markov chain analysis showed that regional evolution was significantly influenced by the coordination levels of neighboring areas. Regions surrounded by areas with high coordination levels had a greater probability of upward transition, suggesting that technological diffusion, knowledge spillovers, and demonstration effects generated positive externalities that promoted coordinated development in neighboring regions. Conversely, regions adjacent to areas with low coordination levels were more likely to remain trapped in unfavorable states, reflecting a “lock-in effect” associated with persistent structural disadvantages. These spatial interactions ultimately contributed to a Matthew effect, whereby advantaged regions continued to improve while disadvantaged regions progressed more slowly. The PVAR-based dynamic interaction analysis showed that agricultural industrial upgrading promoted agricultural carbon-emission reduction primarily by enhancing high-quality agricultural development, indicating the existence of a transmission-oriented and asymmetric interaction mechanism among the three subsystems. Specifically, agricultural industrial upgrading improved agricultural production efficiency, technological innovation capacity, and resource allocation efficiency, thereby promoting high-quality economic development. Improvements in development quality subsequently accelerated the adoption of green technologies and environmentally friendly production practices, leading to reductions in agricultural carbon emissions. However, the direct carbon-mitigation effect of industrial upgrading was not evident at the current stage of development. The impulse-response analysis further confirmed that these interactions involved gradual transmission and dynamic adjustment rather than simultaneous responses. Shocks originating in one subsystem required time to propagate throughout the entire system, highlighting the complexity and nonlinearity of agricultural green transformation. [Conclusion] Promoting coordinated development among high-quality agricultural development, industrial upgrading, and carbon-emission reduction requires multidimensional policy interventions. First, the three subsystems should be incorporated into an integrated policy framework guided by China’s carbon-neutrality objectives to strengthen policy coherence and enhance synergies among development goals. Greater emphasis should be placed on optimizing agricultural production structures, improving factor allocation efficiency, and accelerating the large-scale adoption of green technologies to facilitate the transition from factor-driven to innovation-driven development. Second, differentiated regional development strategies should be formulated according to regional comparative advantages and resource endowments. Eastern China should continue to leverage its innovation capacity to develop high-value-added and low-carbon agriculture, whereas Northeastern China should strengthen large-scale and environmentally sustainable agricultural production. Central China should capitalize on opportunities arising from industrial transfer to extend agricultural value chains and facilitate technological diffusion, while Western China should prioritize ecological conservation and develop resource-efficient forms of ecological agriculture. At the same time, cross-regional mechanisms for the mobility of technology, capital, and talent should be improved to enhance positive spatial spillovers and reduce regional disparities. Third, the integration of digital technologies with green agricultural technologies should be accelerated to advance smart and precision agriculture. Institutional arrangements, including agricultural carbon-accounting systems, carbon-footprint assessment standards, ecological compensation mechanisms, and market-based incentive policies, should also be strengthened to transform agricultural carbon reduction from a predominantly policy-driven initiative into an endogenous component of agricultural development.

Key words: high-quality agricultural economic development, agricultural industrial upgrading, agricultural carbon emission reduction, spatial Markov chain, panel vector autoregression model