Abstract: Based on the strategic interaction and behavioral assumptions of financial institutions, investors and market regulators, a tripartite evolutionary game model of the financial innovation is constructed. The study establishes a time-varying return matrix for the game by taking account of the correlation between investor participation and market risk levels, as well as their impacts on participant behavior. With the dynamical system method, the equilibria of the model are solved, and the asymptotic stability conditions of equilibria are investigated. The result shows that the mixed strategies are not evolutionary stable strategies (ESS). The evolutionary dynamics are influenced by many factors such as the risk level of financial innovation, the efficiency of supervision, the innovation costs, and investment costs. In order to achieve the low-risk regulatory goals, regulators need to take measures to reduce regulatory costs, improve regulatory efficiency, and increase input costs of high-risk financial products. The numerical simulations further demonstrate the complexity of market dynamics under different parameter conditions, as well as the impact of regulatory policy on market evolution. When there are multiple equilibria, regulators should take several measures simultaneously by changing the initial market conditions and equilibrium stability conditions. This study enriches the theoretical research on financial innovation and regulation, and provides significant implications for formulating and adjusting regulatory strategies.

Key words: financial innovation, supervision, evolutionary game, stability