Multi-agent behavioral decision analysis for social public resources digital sharing considering government subsidies and penalties

Digitalization is reshaping the allocation and sharing of social public resources, yet sustained social public resource digital sharing (SPRDS) remains hindered by strategic interaction among stakeholders and governance frictions. A tripartite evolutionary game model involving the government, resource suppliers, and demanders is developed, integrating limited regulatory capacity and a dynamic subsidy–penalty mechanism. The stability conditions and evolutionary trajectories of the participants’ strategies are examined through evolutionary stable strategy analysis, and numerical simulations based on healthcare resource digital sharing in Sichuan. Results show that SPRDS exhibits strong path dependence and bi-stability, with higher initial sharing efforts promoting the adoption of positive sharing strategies. Policy tools have a nonlinear threshold effect, with high-intensity subsidies and penalties leading to an S-shaped shift towards positive sharing. Greater reputational loss under passive regulation accelerates the shift of government to active regulation, while excessively high regulatory costs reduce regulatory effectiveness. Synergistic benefits, the distribution ratio, and sharing investment cost jointly determine the stability of positive equilibrium strategies. This study clarifies the effective boundaries of regulatory intervention and provides actionable guidance for designing cost-effective regulation and fair benefit-distribution mechanisms to promote sustainable and high-quality allocation of social public resources.