There is a longstanding hypothesis regarding securities markets called the “efficient markets hypothesis.” The thrust of the "weak" form of that hypothesis is that the prices of securities “impound” or reflect all publicly available information about the firm whose shares are being traded. For instance, suppose that a published report gives good reason to believe that the market for Firm A’s products is going to be weak in the coming year, so that the profits of the firm are going to fall. Traders who are aware of this report may seek to liquidate their holdings of Firm A’s shares so as to move their funds into what they think are more profitable investments. The result may be a fall in the price of Firm A's shares. Others who passively observe the fall in price and infer from that fact that some people know something adverse about Firm A’s prospects and sell, too. Note that these others have inferred information from stock market price movements. They presumably do not know the details of Firm A’s situation.
The contention that markets can efficiently aggregate and disseminate information is, of course, an old one. But it is one that has led to the creation of new and fascinating markets in recent years. These markets are frequently referred to collectively as “prediction markets.” The public became aware of these new uses in Summer, 2003, when a political firestorm erupted in Washington, DC, over the proposed creation of a market in terrorism futures. The story is beautifully told in the Surowiecki book mentioned below. Here are the barest facts. Admiral John Poindexter and the Defense Advanced Research Project Agency (DARPA) wanted to institute a futures market in certain political events. For example, the Policy Analysis Markets (PAM, as they were tentatively called) proposed to define a futures contract that might have said something like this: “The holder of this future will receive $10 if foreigners have perpetrated a terrorist attack on U.S. soil that has killed or could have killed more than 30 non-terrorists by January 1, 2005; if no such attack takes place or does take place but kills fewer than 30 non-terrorists, the holder receives nothing.” In ways that we leave you to work out, the value of this instrument will fluctuate between the date of its issuance and its due date. (What would it mean if this future was trading at $5.45?) Why did DARPA want to institute such a market? Their hope was that the values of these futures would (like the prices in an efficient stock market) relect a concensus of opinoin about the likelihood of serious terrorist attach by creating an incentive for the discovery and revelation of information about the subject of the futures.
To elaborate very slightly, the original DARPA proposal actually called for two different markets. One would have been what is called an “internal market”—that is, one operated by and open to only those within a certain closed class, such as employees or members of an organization. The contemplated internal market would have been open only to those within the governmental intelligence community and would have allowed those professionals to buy and sell futures on terror attacks (or other relevant political events). The PAM was to be an external market, open to the public. Think, for a moment, about the extent to which the internal market among disparate intelligence professionals might have allowed widely dispersed intelligence information to come together to signal knowledge of or belief in an imminent terrorist attack. The 9/11 Commission, which issued its report in Summer, 2004, famously recommended that there be consolidation of the various governmental intelligence operations in the U.S. under a single National Intelligence Director. Might not an internal market open to disparate intelligence operations gain the advantages of competition among the agencies and yet provide for information coalescence and revelation? (Judge Richard Posner wrote a thoughtful critique of the 9/11 Commission report for The New York Times. You can see it at http://www.nytimes.com/2004/08/29/books/review/29POSNERL.html.)
There was a storm of criticism from politicians and others about the proposed terrorism futures market. In response, DARPA dropped the idea (and some on-line markets began a futures market on whether Admiral Poindexter would resign by the end of the Summer). But it may not have been such a bad idea. Prediction markets have proved extremely useful in wide-ranging contexts. Some large firms, for instance, use internal markets to try to elicit information from their employees about the likely success of new products. Suppose that Firm B wants to introduce a new laser printer to the market. There have been lots of forecasts of sales in the first year, most of them very optimistic. But suppose that the managers at Firm B are somewhat skeptical of those forecasts. Instead they establish an internal prediction market with shares that pay, say, $1 if sales in the first six months exceed a stated amount. Now, the managers hope, that in an anonymous-trading market employees will reveal their true beliefs about the printer’s success in the prices they are willing to pay to acquire the future.
There have been many other recent prediction markets. For example, since 1988 the Tippie College of Business at the University of Iowa has operated the Iowa Electronic Markets, where you can purchase futures on well-defined political outcomes. (See http://www.biz.uiowa.edu/iem/.) The market in the presidential race has been a better predictor of the outcome of those elections than any poll. A host of interesting markets are available at the Net Exchange (www.nex.com) and TradeSports (www.tradesports.com) websites. In the Summer of 2004, Ladbrokes, the British book-makers, opened markets in some pressing questions of science, such as whether there is life on Titan, whether the Higgs boson will be found, and whether experiments will find gravitational waves. For a wonderfully interesting argument on the socially beneficial aspects of more gambling on basic scientific research, see the work of Robin Hanson, an economist at George Mason University. For example, see the article, “Could Gambling Save Science?” at http://hanson.gmu.edu/gamble.html in which Professor Hanson outlines a proposal for “ideas futures.”
We strongly recommend two recent works on the economics of prediction mar-kets: Justin Wolfers and Eric Zitzewitz, “Prediction Markets,” from the Spring, 2004, Journal of Economic Perspectives and James Surowiecki, The Wisdom of Crowds (2004).
TSU
