Bastiat’s Bastions

Late last night (Oct. 1, 2006), I couldn’t sleep and turned on cable news, “Fair and Balanced.” The twin talking heads, the “Beltway Boys,” were making their predictions on how many seats the Democrats would pick up in the House and in the Senate, and whether it would be enough to upset the current Republican majorities in those two chambers. They both thought the Democrats would fall short of winning the Senate, but were split over the House, with one saying they would fall short and the other saying the Dems would get the majority in the House.

One thing to remember about the talk of the talking heads is that Talk is Cheap. Talking heads don’t have that much to lose if they miss things a bit. So, I went to the tried and true method of predicting the outcome of those races, the electronic market at the University of Iowa. At www.biz.uiowa.edu/iem, you can see the results of a real live betting market on the House and the Senate. These markets work better than the polls on predicting the outcome of these races, because they use all information available, including the polls. At the Iowa Electronic Markets, the traders who think that the Dems will take over the House have an incentive to buy up contracts that have the Republicans losing the House. The market, then, weighs the opinions of those who think the Dems will win in proportion of the money bet on the Dems winning, and weigh the opinions of those who think the Reps will maintain their majority in the House in proportion to the money these people bet. The market, then, is a panel of experts who are willing to place bets on their projections, weighted by the money they are willing to bet.

These markets really work very similarly to the Double-Auction markets I demonstrate in my principles of economics classes. They also work very similarly to sports books and pari-mutuel betting (horse racing). Besides the Iowa Electronic Markets, Tradesports.com (http://www.tradesports.com/), a popular sports betting site, also often runs bets on such political events. In fact, if people start betting more on one horse, then the odds for that horse winning goes up. If the markets did not work this way, the odds makers would lose and be forced to pay out more than they take in.

Can anyone really know what will happen in these elections? Not really, at least not until the elections are over. Can we make better predictions than the polls? Here the answer is “yes.”

For a full explanation as to how the market works, go to the www.biz.uiowa.edu/iem site, or go to an article I wrote for the Bayou Business Review some years back at http://www.nicholls.edu/mcoats/newopeds/iowapoliticalmarket.htm.

A couple of years ago, Robin Hanson suggested the same type of betting market could be set up to predict terrorism events. You can read Hanson’s paper, “Designing Real Terrorism Futures,” at http://cipp.gmu.edu/news/PE-of-Terror.htm (I presented my own paper with Robert Tollison of Clemson and Gohkan Karahan of NSU at that conference). Hanson’s idea is that markets are good ways to combine information from many sources, such as the CIA and the FBI, because as we know, these agencies tend not to share information.

Hanson’s “terrorism futures market” never got off the ground. Even though we know that one of the problems that left us vulnerable to the attacks on 9/11 was that the CIA and FBI don’t communicate, and worse still, information sometimes gets squashed in the bureaucracy. Hanson’s market would have given people in the know, the FBI and CIA operatives, an incentive to commincate their information in a simple, easy to understand number, a price. Politicians who didn’t bother to read the proposal got up in arms at the notion of a market for terrorism and squashed the idea before it could even get a fair hearing.

By the way, if you go to the Iowa Electronic Market, you will see that the Beltway Boys weren’t that far off.

MC

This entry was posted on Monday, October 2nd, 2006 at 9:18 am and is filed under General, Political Economy, Prediction Markets. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.