Evolutionary Game Analysis On The Illegal Behavior Of Managers In The Securities Investment Fund In China

With the rapid development of China’s fund industry, insider trading, transfer of benefits and "rat" shady of fund management companies and fund staff continue to burst. In particular, since 2014, with big data as "rat artifact" used in the fund industry of our country, the outcome of investigation by commission audit department is that more than 40 people are involved in insider trading. And the percentage of illegal operations of fund managers is more than 90% of the total number. Fund managers’ illegal operation behavior has seriously hindered the healthy and sustainable development of China’s fund industry. So it is very important to understand the influence of fund manager’s illegal behavior and find out the reasons for the development of the fund industry. At present, these papers usually adopt the method of qualitative and principal-agent model to research influence and reasons of fund manager’s illegal behavior. In this paper, empirical and evolutionary game analysis methods are used to study the influence and reasons of fund manager’s illegal behavior. This article introduces the cognitive deviation factor of investors and fund managers, and measures income by the subjective perception, from the endogenous order and exogenous system perspectives to analyze the reasons. It has a certain theoretical significance. This article will be carried out from the following aspects.(1) Through the collection of China’s Fund Industry Association and each big news website data, this paper understands the development of fund industry and fund managers illegal operation cases in China in recent years. Then, using the event study method, this paper makes an empirical test of the market impact of the fund manager’s "rat" behavior. First, this paper selects 25 open-end stock funds as samples which are managed by fund managers having rat behavior during 2007 to 2014 years. Then it determines the media exposure as the event date, takes before and after 20 days of the event day as the event window, and takes after 90 days of the event day as the event estimation period. The expected return model is used to obtain excess yield and average excess return. Finally, there is a significance test of the excess yields and average excess return. The empirical results show that "rat" behavior of single fund manager does not have a significant impact to fund performance. But over the years, so many fund managers’ rat behaviors have a negative impact to fund industry.(2) Using evolutionary game analysis method, this paper constructs the evolutionary game model of the endogenous order of investors and fund managers. Firstly, the basic assumptions of the model include limited rationality, evolutionary game groups, cognitive bias and income of the participants in different conditions. Then it constructs the evolutionary game model of the endogenous order in the fund market of investors and fund managers, and writes the profit equation and the dynamic equation of investors and fund managers. By solving evolution equations, the equilibrium phase diagrams of the evolutionary game between investors and fund managers are drawn. Finally, this paper analyses the influence of evolution stable factors. Four directly factors, namely fund managers’ long-term development cost, cognitive deviation coefficient, penalties for violations of the risk and investors “repo vision" and two indirect factors, cost of self-protection consciousness of investors, fund managers violations checked probability, lead to illegal operations of fund managers.(3) Using evolutionary game analysis method, this paper constructs the evolutionary game model of the general fund supervision and strict supervision of manager group. The first is the basic assumptions of the model. Then this paper constructs evolutionary game model of the general regulation and strict supervision case of fund managers group, and writes the income under the strict supervision of fund managers group and replicated dynamic equation. It shows the strategy choices of fund managers under different conditions. Finally, this paper selects the initial value to 0.3 and 0.7 to replicated dynamic equation of numerical simulation. The results show that evolutionary stable state of fund managers’ strategy has nothing to do with the initial value, only with penalties to fund managers’ illegal operations and awards to fund managers’ not illegal operations.