March 14 discussion: Trading war futures
|
GEOFF COLVIN: Well, it's a big question: When will war with Iraq start, and how long will Saddam last? You can scarcely punch your remote without hearing some expert's opinion, but there may be a way of predicting war or any other future event that's better than any expert. Online betting sites are consistently beating the experts at predicting election results, Oscar winners, even the weather.
It's much more than just fun and games. The Pentagon has given a group of economists $1 million to apply the idea that for accurate predictions, bettors are better. One of those economists is Robin Hanson of George Mason University.
Robin, I gather when working for the Pentagon, you can't say a lot about what you're up to, but they have said publicly - this is DARPA, the Defense Advanced Research Projects Agency, the people who invented the Internet -- they have said, "they're interested in market-based techniques for avoiding surprise and predicting future events." Now, it sounds to me a lot like betting on college hoops. Is it the same basic idea?
ROBIN HANSON: Well, it's the same basic idea as that, and as of the stock market, currency markets or future markets. Basically, all the familiar financial markets you know of, including these bettor markets, basically work with the same mechanism.
COLVIN: Which is placing bets on how you think events are going to turn out, whether it's whether a stock will go up or down, whether someone will win a basketball game, or whether a war will start by the end of March.
 HANSON: That's right. Some of these things are worth more to find out than others, but it's the same mechanism of finding out the answer: The market asks the question and basically offers money for anybody to fix the current estimate, and people do, in fact, respond and fix the estimate.
COLVIN: This is what's so intriguing to people, that there's money to be made, or least money to be made and lost. And we're going to talk with one of the people who's running one of those sites, John Delaney, CEO and founder of TradeSports.com in Dublin, Ireland.
John, your site is taking bets right now on March Madness and all kinds of sporting events. It's also taking bets on some of the most important questions facing the world today. I wonder what you can tell us your market is telling you right now, specifically about the chances of a UN resolution authorizing force in Iraq.
DELANEY: For sure. On TradeSports.com, which is a person-to-person trading exchange -- not a sports book or a book maker -- we list 500, 600 different contracts, and two of those contracts would be on the UN resolution, whether one will pass authorizing the use of force, before the end of March or before the end of June.
Currently, the March contract has a probability of around 12 to 13 percent. And then on the June contract, around 12 or 14 percent.
COLVIN: Pretty low odds in both cases. What about the odds Saddam will remain in power, let's say first, by the end of March.
DELANEY: Indeed. Around 740,000 contracts have traded on these so far, and the probability currently that traders in TradeSports believe Hussein has to go by the end of March is around 30 percent. But if you look out as far as the June contract, the probability there on TradeSports.com is around 86 percent.
COLVIN: 86 percent by the end of June, so overwhelming sentiment that he'll be toast before the summer.
DELANEY: That's correct.
COLVIN: Thanks, John.
We have to point out that the Justice Department says Americans betting with offshore sites are breaking the law, though that law doesn't seem to be enforced very thoroughly.
Robin, let's talk about another example of where this has worked particularly well in practice, and that's with predicting elections. What's happened?
HANSON: A group of people at the University of Iowa have the Iowa electronic markets, and since 1988 they have been predicting U.S. presidential elections and the primaries and a bunch of other elections. One remarkable statistic that comes out of that is that over this 12-year-or-longer period, they've made basically 600 times when they've been able to compare the market price on a day with poll results, national poll results on that same day trying to predict the election, and out of 600 times, over three-quarters of the time, the market was a more accurate estimate of the outcome than the polls were.
COLVIN: That's interesting, because we know that markets can and do go insane sometimes, as we have learned all too painfully in recent years. Can't it happen just as easily if people are trading election futures or weather futures as if they're trading stocks?
HANSON: Well, certainly they can go crazy. It's the "as easily" question, so we have to compare institutions.
In all the head-to-head competitions we've seen with institutions in real-life situations, markets seem to have done better. That doesn't mean they'll always do better -- I assume you're referring to the dot-com bubble. Markets look like they made a mistake there, okay. But what other institution in American society did better? Was the media all telling us it was a bubble? Were academia all telling us it was a bubble?
COLVIN: It's a valid point. Now it sounds, though, like part of your message is that investors probably can't hope to beat the market -- stock market or any other market.
HANSON: Actually, there's a lot of economic theory that says that people should stay away (from the markets) because, on average, they lose money. But fortunately for the rest of us, they seem to go in there and trade, and they provide the rest of us this benefit of their information on their estimates. So, people who go trade in these markets, maybe on average they lose, but the rest of us benefit by the information they create and provide for us.
COLVIN: It's fascinating stuff. Robin, thank you so much.
HANSON: You're most welcome.
|
