Information Cascades in the Stock Market
“Herding Behavior in the Stock Market” by Victoria Duff: https://finance.zacks.com/herd-behavior-stock-market-9833.html
This article describes different forms herding behavior that are common in the stock market and explains how herding behavior shapes the nature of the stock market. In the context of the financial world, herding behavior refers to the phenomenon by which investors will base their investment decisions on the decisions of other investors, such that they are “following the herd.” In general, standard trades on the stock market are not the result of herding behavior, in that individuals will buy or sell a given stock based on what they anticipate will be the success of that stock. However, herding behavior arises in the face of uncertainties in the financial market. The article defines three main types of herding behavior based on their causes. Information-based herding occurs when a large group of investors react in similar ways to announced information. Reputation-based herding occurs when investors make trading decisions based on a trading decision made by a respected or major trading organization. Compensation-based herding occurs when an institution that controls a large amount of financial assets, for the purpose of securing profit, sells a large amount of stock in a portion of the financial market that is commonly invested in, resulting in reactions from a large number of other investors. In general, herding behavior is likely to arise when investors notice unusual trading activity, referred to as trade imbalance. Upon noticing the abnormal activity, investors come to believe that those who were part of the unusual activity know something that they do not, and decide to follow the unusual activity. In the case that the amount of a particular stock sold is abnormally high, investors fear that there is some reason why the value of that stock will go down, so they sell their stock to avoid losing money. Similarly, in the case that an unusually large amount of a stock is bought, investors will buy that stock because they think those who initially brought the stock had some reason to believe that its value would go up. The article also mentions stock promoters, who work to make investors buy their companies’ stock with the hope of inducing herding behavior. A company’s ability to attract investments and seek financing, as well as to compensate its employees using company stock, is largely dependent on its stock price, which is why it is in their best interest to maximize their stock price.
The herding behavior discussed in this article strongly relates to the information cascades discussed in class. As was seen in the examples discussed in class, information cascades occur when the information provided by previous decision-makers’ decisions take precedence over a given decision-maker’s private information, such that, after a series of the same initial decisions, all following decision makers make that same decision, regardless of whether or not their private information supports it. In the case of herding in the stock market discussed in this article, an individual decision maker’s private information is their own evaluation of the company in question and how its stock is likely to change. For example, an investor might research and analyze a company and decide that its products will decrease in popularity, such that their private information would lead them to the decision to not sell, or not buy, stock in the company. An information cascade occurs when an investor sees that previous investors have decided to sell a stock. The investor does not have access to their private information, and only has access to his/her own private information, but decides that the previous investors’ private information supported the decision to sell, and that these signals to sell are more powerful than the investor’s own private information (his/her personal evaluation of the company), so the investor decides to sell as well. Now, the next investors that own that stock have even more signals to sell, and will also decide to sell, resulting in a cascade that will result in the value of the stock plummeting. In the same way, if the initial investors had decided to buy, the value of the stock would increase drastically. Large changes in the stock price can result from a small amount of potentially inaccurate initial decision-making analysis. In the stock market, these cascades occur extremely quickly, which makes them eve more likely to continue, because investors needing to make a trading decision have very little time to do research or analysis that would allow them to gain more private information. The tendency of information cascades to occur in the stock market is the basis of stock promotion. To substantially increase the value its stock, a company need not convince a large number of investors that their company will be successful, but rather needs only convince a relatively small number of investors that their company will be successful so as to initiate an information cascade.
The prevalence of herding behavior in the stock market, which results in the ability for information cascades that are disastrous for a company’s stock price to be triggered easily and without warning, makes the stock market dangerously volatile. However, there is a significant difference between the information cascades discussed in class and the information cascades that occur in the stock market that make the stock market, as well as the financial world in general, even more unstable than it otherwise would be. In the stock market, at least in the short term, the value of a stock is largely determined directly by peoples’ decision to buy or sell that stock. For example, at least in the short term, if many people decide to sell a stock, the value of that stock will go down, regardless of the reason that it was sold. This provides investors deciding whether or not to sell the stock more reason to sell the stock. In the examples discussed in class involving bags of marbles, student’s knowledge of previous students’ decisions on if the bag being picked from was majority blue or majority red influenced their determination of which bag was more likely, but which bag was being picked from was fixed at the beginning of the experiment and was unchanged by previous students’ decisions as to which bag to guess. In contrast, in stock market information cascades, whether or not the value of the stock will increase or decrease is determined partially by if investors think the value of the stock will increase or decrease and make the corresponding trading decision. If we sought to model stock market information cascades using the framework from lecture, we may decide to represent previous trading decisions both directly as previous investor’s decisions based on their not fully known private information and within the decision maker’s own private information. This acts as an “accelerator” for the information cascades of the stock market, and makes them especially difficult to stop. The same phenomenon is well represented in “band runs.” In a bank run, an initial group of people think that the band will fail and are concerned that they will not be able to get their money stored in the band, so they withdraw their money. This makes the bank more likely to fail, and leads more people to withdraw their money. The resulting cascade of withdrawals is extremely difficult to stop.
In addition to resulting in drastic drops in stock prices and closing of banks, information cascades in financial markets can result in financial “bubbles” that initially may help the economy, but eventually will collapse and harm the economy. Bubbles arise when an initial group of investors invest in a particular industry, product, or set of companies, and cause more investors that trust that the initial investors made their decisions based on sound private information to invest. As the amount of investment increases, investing, which may in reality be very risky, comes to seem increasingly safe. The bubble collapses when it is finally realized that the large amount of investments made have little genuine basis. An example is the dot-com bubble of the late 1990s, in which a cascade resulted in excessive investment in internet companies that lacked good business models.
While not discussed in this article, which is about how herding behavior effects the values of individual stocks, in discussing how ideas from network theory discussed in class relate to the financial world, it is important to mention the very high degree of interconnectedness in the network representing the financial worlds and how the nature of this network influences the health of the world economy. Failures of even one relatively minor sector of the economy can have major and far reaching effects throughout the entire world economy. In the network representing the financial world, one could define nodes as companies, industries, industries in particular regions, types of financial assets, or another suitable division of the economy. If one sector of the world economy depends on another, the link could be represented as a directed edge. Further, each link should be assigned a strength, much like our defining ties as being strong or weak in class, which represents how important that link is to its endpoint nodes. Using this network, one could gain insight into how the failure of a particular node may cause the failures of other nodes within the network, and potentially result in an economic recession, much like how a contagion which initially infect only one person may have the potential to spread and lead to an epidemic.