Research on the Incentive Mechanism for Multiagent Collaboration in Investment and Financing of Rural New Infrastructure in China: Based on Evolutionary Game Theory and Simulation

Against the backdrop of the global accelerated transition into intelligent, rural new infrastructure has become a key vehicle for bridging the urban–rural digital divide and promoting comprehensive rural revitalization. However, the practice of investment and financing for rural new infrastructure in China faces systematic challenges, including significant government fiscal pressure, market failures, and difficulties in multiagent collaboration. Existing research predominantly focuses on two‐party games between the government and a single agent, failing to fully reveal the complex interaction mechanisms and cooperative potential among the three core stakeholders: government, enterprises, and rural collective economic organizations (RCEOs). Based on evolutionary game theory, this study constructs a tripartite dynamic game model and employs MATLAB simulations to systematically analyze the evolutionary paths and stability conditions of strategy selection among agents within the rural new infrastructure investment and financing system. The study finds that the system can eventually converge to the ideal equilibrium E 8 (1,1,1), namely, (Strict Supervision, Active Cooperation, Active Cooperation). Furthermore, an efficient evolutionary path centered on “Government–Enterprise Collaboration” is identified, characterized by a two‐stage process: first, forming a stable core of “Government Supervision–Enterprise Cooperation,” which subsequently facilitates the integration of the RCEOs and achieves comprehensive collaboration. Sensitivity analysis indicates that government supervision efficiency (social benefit increase coefficient α/supervision cost coefficient γ), enterprise incentive constraints (additional benefit E q /subsidy reduction ratio β 1 , penalty P 2 ), and RCEO incentive constraints (additional benefit E c /subsidy reduction ratio β 2 , penalty G 2 ) are key parameters affecting system convergence and efficiency. Accordingly, this study proposes phased and differentiated policies, including initial consensus building, strengthening government–enterprise collaboration, and promoting deep integration of RCEOs, supported throughout by digital supervision and credit mechanisms. This research expands the theoretical framework of multiagent collaborative governance and provides mechanism design and decision‐making support for addressing the collaboration dilemmas in rural new infrastructure investment and financing.