.
.
.
.
.
.
.
.
.
.
.
.
.
.
It is the main reason for the herd behavior, the article carries out theoretical and game analysis on the herd behavior between individual investors and institutional investors, and reveals the causes and effects of this anomaly from another perspective.
Keywords Behavioral finance, Herd effect, Investor behavior
Herd behavior in the financial market is a special kind of irrational behavior, which refers to the fact that investors, in the uncertain information environment, are influenced by other investors, imitate the decisions of others, or rely too much on the decisions of other investors. others' decisions, or rely excessively on public opinion without considering their own information Since herd behavior is a correlated behavior involving multiple investment subjects, it has a great impact on the stability and efficiency of the market, and is also closely related to financial crises. Therefore, the herd behavior has attracted extensive attention from academics, the investment community and financial regulators.
1 Research on the causes of the herd effect
There are several explanations for the formation of herd behavior. Philosophers believe that it is the limited nature of human rationality, psychologists believe that it is the human herd mentality, sociologists believe that it is the human collective unconscious, while economists explain the herd behavior from the perspective of incomplete information, principal-agent and other perspectives, in summary, there are mainly the following views:
1. 1 Class herd effect due to similarity of information
Froot, Scharfstein and Stein (2000), who have been working in the field of financial management for more than 20 years. Scharfstein and Stein (1992) point out that institutional investors are highly homogeneous, they usually focus on the same market information and use similar economic models, information processing techniques, portfolios and hedging strategies. In this case, institutional investors may react similarly to the same external information, such as earnings warnings or securities analysts' recommendations, and exhibit herd behavior in their trading activities.
1.2 The herd effect due to incomplete information
Information can reduce uncertainty, and investors can obtain accurate, timely and effective information, which means that they can obtain high profits or avoid significant economic losses. However, in the real market, the acquisition of information requires the payment of economic costs, different investors have different ways to obtain information and ability, institutional investors have capital, technology, talent, scale advantage, individual investors in the payment of information costs are far from being able to compare with institutional investors. The direct consequence of this is that institutional investors get more effective information than individual investors, and individual investors are in a disadvantageous position in obtaining effective information and obtaining investment returns. Individual investors in order to avoid risk, to obtain more real economic signals, will probably go around to find out the dealer's "inside information", or to relish in the "unwarranted" empty talk, to a greater extent to promote the market tendency to chase the wind.
And in fact even institutional investors, information is not sufficient. In an incomplete and uncertain market environment, it is assumed that every investor has private information about a stock, which may be the result of their own research or obtained through private channels; on the other hand, even if the public information about the stock has been fully disclosed, investors are still not sure about the quality of this information. In this market environment, when investors have no direct access to other people's private information, but can speculate on their private information by observing other people's buying and selling behaviors, herd behavior is likely to occur. Although institutional investors are in a strong information position relative to individual investors, they are in turn more prone to herd behavior than individual investors because institutional investors know more about their peers' buying and selling among themselves and have a higher ability to infer information.
1.3.3 Herd effect based on principal-agent generation
1.3.1 Herd effect based on principal-agent's reputation
Scharfstein (1992) and others provide the theory of reputation-based herd effect of fund managers and analysts. Concerns about reputation arise because of uncertainty about the competence of investment managers.
Agent 1 invests after being signaled that "income is high". Since agent 2 is concerned about his reputation, he will follow the same investment strategy as agent 1, regardless of the signal. This is because if the decision is correct, his reputation will increase; if it is wrong, it shows that either both are stupid or both are smart but got the same wrong signal, which does not harm his reputation. If different decisions are taken, the principal assumes that at least one of them is stupid. So agent 2 will keep applying the herd strategy, regardless of the signaling differences between him and agent 1.
If several investment managers make investment decisions one after the other, each mimics the decision of the first investment manager to make a choice. Ultimately, if the investment is profitable, good signals will prevail. Private information will ultimately not be reflected in investment decisions because all investment managers will follow the first investment manager in making decisions. Thus, this herd effect is ineffective. Moreover, it is fragile because, the investment behavior of the following investment managers will change because of the little information received by the first investment manager.
1.3.2 Herd Effect Based on Agent Compensation
If the compensation of investment managers depends on their investment performance relative to other investment managers, this distorts the incentives of the investment managers and leads to ineffective portfolios selected by the investment managers (Brennan, 1993).
Maug (1996) and others examined risk averse investors whose compensation increases with the relative performance of the investor and decreases with the relative performance of the investor. Both the agent and his benchmark investment manager have imperfect information about stock returns. The benchmark investor invests first and the agent chooses the portfolio after observing the benchmark investor's choices. Based on the previous model of the herd effect with insufficient information, the investment manager's portfolio choice will tend to select portfolios that are similar to those of the benchmark investor. Moreover, the compensation system encourages the investment manager to mimic the choices of the benchmark investor because, if his investment performance is lower than the average investment performance of the market, his compensation will suffer.
2 Game Analysis in the Herd Effect
The herd effect arises from the consideration of individual investors and institutional investors for their personal interests, therefore, with the game theory approach, we can have a deeper understanding of the reasons for the herd effect.
2.1 The game between institutional investors and individual investors
The game between institutional investors and individual investors can actually be seen as a kind of deformation of the game of the wise pig, we assume that institutional investors and individual investors are invested in the stock market, and the institutional investors, due to the large capital, can get 100 benefits if they invest based on the correct information, while individual investors can only get 5 benefits based on the correct information. Both sides can choose to collect and analyze the information, which generates a cost of 20, or simply collect only the other side of the action of the information and follow, which generates a cost of 1, both sides give up the collection of information, generating a utility of zero. If both the institutional investor and the individual investor take the behavior of collecting information and analyzing it, then the institutional investor will get the benefit of (100-20 = 80), and the individual investor will get (5-20 = -15), if the institutional investor collects information and the individual investor follows, the benefit is that the institutional investor (100-20 = 80) and the individual investor (5-20 = -15) will get the benefit of (100-20 = 80). The interests of institutional investors (100-20 = 80), individual investors (5-1 = 4), such as the opposite, then the interests of -15, respectively, 99, which produces the following matrix of interests: with the analysis of information, then the final result is that everyone's interests are zero. And institutional investors to collect and analyze information, although it will allow individual investors to take advantage of, but after all, there is something to gain, so the cumulative strictly optimal solution to this game is, institutional investors to collect and analyze information, individual investors to analyze the behavior of institutional investors and follow. Therefore, it also produces the herd behavior of individual investors to institutional investors.
2.2 The game between managers
The game between managers is more complex, but we can use a simple model to analyze it roughly, assuming that there are two competing managers, for the current market has produced a certain manager's investment behavior, there are two choices, follow and do not follow, we assume that the success rate of this investment strategy P = 0.5 If the success of this investment strategy P = 0.5, if the success will get 10 gains, if the failure, the loss of 10, they can also choose not to follow this investment behavior, the use of their own information to make investment decisions, so that the success rate of P2 = 0.7, the return on the situation remains unchanged. In this way we can calculate the return expectation of each strategy
The return expectation of following I1=10*0.5+-10*0.5=0
The return expectation of not following is I2=10*0.7+-10*0.3=4
Finally, the game obtains an optimal solution, which is also an efficient solution, that is, not following -No Follow, and this is actually based on the rather idealized assumption that for the manager, utility = return. The above matrix of return expectations does not reflect the considerations of reputation and compensation of the manager as described above, and we can conclude that for the manager, the loss of decision-making errors with other investors is different from the loss of decision-making errors alone, and that decision-making errors arising from no-following behaviors, in addition to the monetary loss of the fund, also carry reputational risks, and the possibility of losing one's job if one is perceived to be a stupid investment manager. And professional managers for the reputation and job opportunity concerns, will undoubtedly have an impact on their decision-making position, so it is necessary to use the manager's utility matrix instead of the earnings expectations matrix, for the manager, due to the decision-making errors arising from the non-following, the loss of: book loss + the manager's personal reputation and loss of compensation = 10 + 20 = 30, from which we can derive:
Following the utility of the expected is u1=10*0.5+-10*0.5=0
The utility expectation of not following is u2=10*0.7+-30*0.3=-2
In this case, following-following is the equilibrium solution of the game, which proves that one of the direct reasons for the herd effect is that in many In many cases, a professional manager will abandon his relatively correct information and investment strategy to follow an unknown investment strategy in order to achieve stability in his own career and reputation.
3 Implications of Herd Behavior
(1) Because "herders" tend to abandon their own private information and follow others, this can lead to a disruption in the chain of market information transmission. However, this situation has two effects: first, "herd behavior" due to a certain degree of convergence, thus weakening the role of market fundamentals on future price movements. When investment funds have "herd behavior", many funds will buy and sell the same stocks at the same time, the pressure to buy and sell will exceed the liquidity provided by the market, the excess demand for stocks has an important impact on the stock price changes, when the fund net selling stocks, will make the price of these stocks to a certain extent of the downward leap; when the fund net buying stock when the fund buys stocks, it causes these stocks to rise sharply in the current quarter. This leads to discontinuity and drastic changes in stock prices, which destabilizes the operation of the market. Secondly, if the "herd behavior" is due to the rapid response of investors to the same basic information, in this case, the "herd behavior" of investors to speed up the speed of stock prices to the absorption of information, to promote the market more efficient.
(2) If the "herd behavior" exceeds a certain limit, it will induce another important market phenomenon, the emergence of overreaction. In the rising market (such as the bull market), blindly chasing the rise over the value of the limit, can only create a bubble; in the falling market (such as the bear market), blindly kill the fall, can only be the deepening of the crisis. Investors' "herd behavior" caused by large fluctuations in stock prices, so that the stability of the securities market decline.
(3) The basis of all "herd behavior" is the incompleteness of information. Therefore, once the information state of the market changes, such as the arrival of new information, "herd behavior" will collapse. The excessive rise or fall in stock prices caused by "herd behavior" will then stop, and even return excessively in the opposite direction. This means that "herd behavior" is unstable and fragile. This is also a direct result of the instability and vulnerability of financial market prices.