Abstract: Risk issues of peer-to-peer (P2P) lending have become increasingly prominent with its rapid development; in order to promote the sustained and healthy development of the P2P lending market, it is necessary to effectively supervise the P2P lending risk evolution mechanism based on its clarity. This paper applies evolutionary game theory by establishing a game model between P2P platform and supervisor in this context. The model is then solved and deduced, and numerical experiments are carried out on this basis to reveal the rules of P2P lending risk evolution. We find that the P2P lending risk is the result of repeated game between the platform and the supervisory under the established conditions, the level of market risk varies with the structure of the behavioral groups of both parties. The increase in the compliant operating income of the P2P platform will decrease the risk level of the platform group; the raise in the strict supervision income and the reduction in the loose regulatory income will lead to a decrease in the risk level of the regulators; the risk level of both parties will decline when the penalty of the platform and the transfer income of the regulator increase, and the relatively low online P2P lending risk level is gradually formed.

Key words: online peer-to-peer lending risk, P2P platform, supervisor, evolutionary game