Skip to main content



Information Cascades aplenty in Finance Markets and Real Estate

As discussed in class, Information Cascades is the phenomenon wherein when a group of people are connected by a network, it becomes possible for these people to influence one another’s decisions. We discussed the simple herding model where people make decisions sequentially, and each person can observe the choices made by those people who acted earlier. In addition, each person has some private information or signal to help guide their own decision but does not have access to the signals received by other people. Herding can be used to explain the behavior of financial markets as well and in fact in the recent financial crisis, herding has come to be known as more of a negative term because of the volatility it creates in the market. This blog discusses the affects of herding in the world of finance.

Investors are often influenced by the decisions made by other investors and they rationalize this decision-making by assuming that the other investors probably know something more about the return of investment than they themselves do. There are many reasons for this “rational herd behavior” and the paper I have cited at the end of this blog post does well in explaining those. The ones that are the most important are imperfect information concern for reputation and compensation structures.

Going back to making investment decisions, let us consider a situation where there are a bunch of individuals facing the same investment decisions and have some private but imperfect information and the investors make their decisions in sequence. Each investor has either a positive or a negative payoff and receives a good or a bad signal. Applying Bayes rule, we can infer that the posterior probability of a positive payoff after observing a good (G) signal is greater than 0.5 and so Person 1 will follow his own signal. Now person 2 sees his own signal and also knows the decision made by person 1 and also infers that he received a “G” signal. Let us suppose that even he decides to invest. Now the person 3, as can be shown by Bayes rule, will invest in the stock even if he himself receives a bad signal. Person 4,5,6,… are now all in the same position as person 3 and will therefore all begin investing in the same stock. This is how the cascade begins. The type of cascade also depends on the order in which the signals arrive. If it is in the order GGBB, an “Accept” cascade begins. If it is in the order of BBGG, a “Reject” cascade begins. It is therefore “path-dependent.” This type of cascade is also very fragile and can easily be broken by the arrival of better-informed investors, release of new public information and shift in the underlying value of investing as opposed to not investing. This is what leads to volatility in the stock market.

The idea of Reputation-based Herding is that if an investment manager and his employer are uncertain of the manager’s ability to pick the right stocks, then conformity with other investors benefits the manager and if there exist other similar investors in the same position, then herding occurs. This type of herding is inefficient because it is based on the decision made by the first investor who may or may not be correct and everyone else’s private information has no value.

The prices of real estate are also influenced by information cascades. The sudden boom in real estate prices between 1997 and 2004 was driven by an information cascade where people believed that it is only natural for property prices to increase. There was imperfect information and this was carried on to a larger population that led to the real estate cascade and property owner were able to charge exorbitant amounts on their houses.

Information cascade is an interesting phenomenon especially considering the impact it can have in a business or for that matter even in the day-to-day life. There are many other such examples of information cascade and it is something that happens very naturally and is hard to avoid.

http://economistsview.typepad.com/economistsview/2008/10/informational-c.html

http://www.imf.org/external/pubs/ft/wp/2000/wp0048.pdf

 

Comments

2 Responses to “ Information Cascades aplenty in Finance Markets and Real Estate ”

Leave a Reply

Blogging Calendar

November 2012
M T W T F S S
 1234
567891011
12131415161718
19202122232425
2627282930  

Archives