Skip to main content



Information Cascades in Real Estate

http://www.nytimes.com/2008/03/02/business/02view.html

 

This article from The New York Times investigates the recent housing bubble in the United States. The writer attempts to explain why rational people were pursuing the potential of grand returns on investment without properly considering the risks. He claims through the analysis of various economists that information cascades can lead people astray due to that fact that people at times need to rely on the judgment of others to make decisions. An example follows in which each person in a group of individuals possesses information to choose whether to invest in real estate, and each person’s information has a 60% chance of leading to the correct decision. The ideal solution would be to hold a national meeting where everyone can pool together their knowledge and collectively arrive at the right decision. This is obviously impossible, so people make their own independent choices and reveal them through their actions, such as entering the housing market. The author continues the example, supposing that houses have low investment value. We assume the first person makes a wrong decision and reaches the wrong conclusion, which occurs 40% of the time. This person indicates to the community that investment is a good idea by paying a high price for a house. The second person would make the same incorrect decision if his information confirms that of the first person. If there is a conflict, however, the second person would conclude that his information is useless, and make an arbitrary decision. Thus, we could have two people showing that housing is a good investment even when it is not. This pattern would continue, as more and more people would think that the previous buyers’ information outweighs their own.

The article’s discussion is a real-world example of information cascades. When people are connected by a network, they necessarily influence each other’s behavior. An information cascade can potentially occur when people make decisions sequentially, with later people observing the actions of earlier people and inferring something about what the earlier people know. A cascade develops when people abandon their own information in favor of the evidence from these inferences. Cascades, however, do not result from thoughtless decision-making, but rather logical deduction based on limited information. We can take an example with an urn containing three marbles. It is known that there is a 50% chance that the urn contains two red marbles and one blue marble, and a 50% chance the urn contains two blue marbles and one red marble. Each person draws a marble from the urn, looks at the color, and then places it back in the urn without showing it to the others. The person then publicly guesses whether the urn is majority-red or majority-blue. The first person follows a general decision rule: if he sees a red marble, he guesses majority-red, and if he sees a blue marble, he guesses majority-blue. If the second person’s marble matches the first guess, then he should guess this color as well. Conversely, if the marble is different, the person’s indifference would most likely lead them to choose their own color. Thus, the first two people convey perfect information about their marble. The third person simply guesses the color he sees if the previous two guesses were opposites. The fascinating case occurs if the third person draws a marble that opposes the color of both the previous guesses. By probability, the third person would guess the color matching the previous guesses regardless of the color he observed. In this way, an information cascade is formed. The fourth person and everyone that follows is placed in the same situation as the third person. They will choose the color coinciding with the majority because of the assumed probability. Clearly, the rational behavior of these people nevertheless leads them to make potentially incorrect decisions due to the information cascade.

Comments

Leave a Reply

Blogging Calendar

November 2017
M T W T F S S
 12345
6789101112
13141516171819
20212223242526
27282930  

Archives