Orin Kerr asks about the accuracy of electronic markets. Saul Levmore has an interesting review of the evidence:
Efficient Markets and the Role of Regulation: Lessons from the Iowa Electronic Markets and the Hollywood Stock Exchange Symposium: Revisiting the Mechanisms of Market Efficiency, 28 J. Corporation L. 589 (2003).
Levmore reviews the evidence that even small fantasy markets with little money at stake and thin trading do an extraordinary job predicting outcomes. Those who log on through their law school or law firm computers might be able to download the article at heinonline.org.
I’m reasonably certain that Tradesports.com predicted the outcome of every state in the Nov. 2004 election, even New Hampshire, Iowa, Wisconsin, and New Mexico. That said, markets are frequently wrong; a recent Tradesports example is that traders overwhelmingly thought that Rehnquist would step down before O’Connor. The question is whether experts can usually do a better job than markets. It would seem that the answer is generally No.
UPDATE: By the way, Edith Clement’s odds (actually, probabilities) have dropped to 32-48%, while Edith Jones’s odds (probabilities) have risen to 27-31% by 5:10pm ET.
2d UPDATE: Cass Sunstein is dicussing information aggregation and markets at Lessig’s blog.
And Jones is now (5:30 ET) at 50-54%, while Clement has dropped to 20-28%.
Comments are closed.