Background: Efficient Market Hypothesis
The Dumb Agent theory is predicated upon the Efficient Market Hypothesis (EMH). This is used to explain a frequently observed phenomenon, and was coined and best explained by Eugene Fama in 1970, who defined EMH as:
“… a market that is efficient in processing information. The prices of securities observed at any time are based on ‘correct’ evaluations of all information available at that time. In an efficient market, prices ‘fully reflect’ available information.”
In a nutshell, EMH states that markets are efficient, or as efficient as possible, and the fewer artificial constraints imposed on them (such as government intervention and insider trading), the more efficient they will be. A full description of EMH will be coming to the site shortly.
Theory: Dumb Agent Theory
The Dumb Agent Theory takes the EMH one step further.
It states that a market, at any given point in time, reflects the true value of its underlying stock. Therefore, although many experts may spend vast amounts of time looking over charts and information about companies and stocks in order to discover information that others do not know, this will be futile in the long term, since all available information is already incorporated into the stock price.
It may seem that the Dumb Agent theory does not differ from the EMH, but there is one fundamental difference, and that is that no one person needs to know the true value of a stock for the market to find its true value. When many people invest, some will offer too much while others will offer too little, but the equilibrium point will equal the true value of the stock. As soon as new information is available it is incorporated into the stock price, and the non-relevant information is weeded out. This may seem more like mysticism than economics, but it is merely the Invisible hand of Adam Smith at work.
One of the first examples is from a 1968 study by Ball and Brown, the first one on market efficiency, which stated that 80% of a stock price can be forecasted before an earnings announcement. This must therefore mean that the outcome of the announcement is already incorporated into the stock. The fact is, no one investor need know for sure what the announcement will be, but since many investors are playing in the market with all the information related to their stocks available to them as a whole, the equilibrium price will reflect the earnings announcement.
This can be shown in a study performed by Bernard and Thomas on Post Earnings Announcement drift, wherein it is interesting to look at the Pre earnings announcement drift. Bernard and Thomas decided to look at companies that had “surprised” with their most recent earnings announcement. This surprise was measured relative to:
1. A simple time series model for expected earnings which gives a measure called “Standardized Unexpected Earnings” (SUE).
2. Earnings expectations by analysts.
3. Stock price reaction once the announcement is made
Therefore these were stocks that announced a level of earnings that was different (at varying levels) from what was expected beforehand. As can be seen in the graph below, the bigger the surprise (i.e. the larger the difference between expected and announced earnings), the bigger the gain or loss of the stock before the announcement. This gain or loss becomes quite evident already 20 to 30 days before the announcement. Barring illegal insider trading (which is highly unlikely for such a big group of stocks), this is a perfect example of how many traders find true value more efficiently than any one individual expert, analyst or formula.
(Bernard – Thomas, 1989)
Another example, in a slightly different context, is that of Michael Maboussin, who is an investment strategist at CSFB, and also teaches a course in the Columbia Business School. Every year, he asks his class to estimate IBM’s total assets for the year 1989. It is quite safe to say that no one in the class will know this answer by heart. However, every year, the mean of the students’ choices comes within 5% of the correct number. This follows the theory that markets are efficient (more precisely, the dumb agent theory), and that a group of people investing will turn a more accurate value for their investment than any one individual could.
The 2003 Iraq war can be seen as yet another example. Well before it started, petroleum prices had factored in a “War premium”, with the West Texas Intermediate Spot oil prices already being 52% higher at the end of 2002 than 12 months prior (and steadily climbing to 73% higher by February 2003). This, therefore, showed how the start of the war was foreseeable by observing markets.[i]
It has also been observed with the Challenger shuttle disaster (explained more in length in James Surowiecki’s Wisdon of Crowds. After it occurred, NASA’s expert committee needed several months in order to find the cause of this event, at first blaming the explosion on the space shuttle external tank manufactured by the Martin Marietta corporation. The markets, however, gave their verdict immediately. Morton Thiokol, manufacturer of the faulty O-Rings, quickly received a 12% loss in share price, while the other shuttle contractors remained relatively free of market punishment[ii]. The final confirmation came six months later, when the NASA committee charged with investigating the event found that only Morton Thiokol’s O-rings were to blame, while the space shuttle external tank and indeed all the inputs by other companies, were not to blame.
Interesting resources for further study of Dumb Agent Theory:
- · James Surowiecki’s Wisdon of Crowds, available for purchase here.
- · The Influenza Prediction Market, an explanation of which can be found here.
- · To see the Dumb Agent Theory in action, you can visit these U.S. Election Prediction Markets, tracked in real time.
- · Dumb Agent Theory entry on Wikipedia.
[i] Andres Leigh et al. “What do Financial Markets Think of War with Iraq”, Stanford University, March 17, 2003
[ii] Daniel Gross, “The Disaster Market: Can Wall Street Figure Out the Cause of a Space Shuttle Crash Faster than NASA’s Experts?”, Slate-online, http://www.slate.com/id/2086811/, August 8, 2003