Skip to main content



Information Cascades and Trading

The following research paper from the London School of Economics analyzes the impact of information cascades in trading. The efficient market hypothesis, a standard assumption in the world of finance, argues that the price of any given security at a certain time is at equilibrium with respect to all of the information about the security that exists in the market. The EMH is often cited as proof that there are no arbitrage opportunities within the market.

However, this research paper by Dasgupta and Prat argues that as long as traders have a reputation to maintain, and possess career concerns, there cannot be a market that is informationally efficient i.e. a market where prices converge to a single value if an infinite number of trades were allowed to be made. They basically argue that traders that trade on behalf of institutions are guided by their own personal objectives and incentives. Their guiding principles are clearly not public knowledge, and therefore standard financial models cannot fully capture the incentive structure that governs institutional trading.

Often, career goals lead to conformist action, where traders will do what is known as “window dressing” or pick a portfolio of investments that conform to the beliefs of others in their organization in order to make themselves seem more agreeable because this would result in a higher chance of them moving up within the organization.

Dasgupta and Prat develop a model where the payoff of a fund manager is the sum of the monetary payoff that their portfolio makes, as well as a payoff function of the reputation that they earn due to a certain sequence of trades.  They show that in a market full of career-oriented traders, prices will never converge to their true liquidation value even though prices evolve to incorporate the private information of fund managers. This is because fund managers that trade according to their private signal will have a positive profit that tends to zero whereas managers that take a “contrarian” view have a high probability of the trade becoming incorrect and looking dumb in the eyes of others. Therefore, fund managers will tend to conform and their trade will stop reflecting their private information.

Furthermore, they show that a “reputational benefit/cost” or the difference between the expected value of trade for a regular trader vs. a trader with career concerns, can often exist and the reputational benefit/cost lead to a systematic mispricing of securities. Past research showed that information cascades weren’t possible in a market because after every investment decision the price of a security would adjust to reflect the expected value of the asset based upon the information revealed by the past trades. However, Dasgupta and Prat show that this is no longer the case. When you take a look at traders who trade based upon market information and also their career concerns, markets can become informationally inefficient and generate mispricing. In section 4 of the paper, they show mathematically how such market conditions can result in information cascades which don’t necessarily result in a market breakdown but can still result in mispricing of securities.

These results are interesting because they contradict a lot of basic assumptions made about efficient markets and help explain market anomalies, cascading prices, and also the existence of arbitrage opportunities in markets.

http://else.econ.ucl.ac.uk/newweb/esrc_seminars/june2005/Dasgupta.Amil.pdf

Comments

Leave a Reply

Blogging Calendar

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

Archives