Tuesday, July 29, 2003

Poindexter, Wolfowitz, Daschle, Dorgan, Wyden & Warner:

Policy Analysis Market (PAM): http://www.policyanalysismarket.org is now a blank page.

Sigh. Here's a great example of what politics and law are all about, and how they often clash with rational thought. As one of my law professors said during the very first class: "Don't expect law to be about logic. It's not. It's about emotions -- and emotions are not logical."

A fellow Caltech alumnus of some renown, Ret. Adm. John Poindexter, now at DARPA, implemented a plan to use futures markets as predictors of terrorism. Regardless of the fact that the idea makes for really great political fodder, and it sounds a little kooky, the idea actually works. It is based on the fact that people tend to make more conservative and therefore probably truthful guesses about outcomes of uncertain situations when they have a stake in them, and therefore accurately mirror the wider public. These market set-ups have been shown to be more accurate than pundits with things like elections and box office returns, as in the University of Iowa's Tippie College of Business Iowa Electronic Markets Program (IEM). The research involved in the ideas behind the IEM project is not trivial, and has the financial backing of the Department of Education, the National Science Foundation, Microsoft, and NASDAQ. Heck, I remember listening to Minnesota Pubic Radio's "Markeplace" and "Future Tense," who had spots on the idea on November 6th 2000, and February 28th 2003 (see archive), and sang its praises.

The terror futures market program at DARPA was being set up with the help of people who are not fools either: NetExchange and The Economist. The study was not really intended to be focused on individual terrorist acts, but on much wider phenomena, at which these types of things excel. The focus was more along the lines of: "How would public opinion in Jordan shift if X or Y happened."

I was amazed to see how few of the news stories that ran about the Pentagon's defense and then withdrawal of the program actually referred to the research showing that this technique works - the focus of the stories was of course more on the political scandal.

If the Senators are shocked, shocked by 'betting parlors' they should be consistent and shut down all futures markets. Especially those dealing in pork bellies, but I imagine there might be an objection from New York's junior Senator. The amazing thing is that no one seems to have picked up on the sheer hypocrisy of the situation at the Pentagon briefing dealing with the cancellation of the DARPA Program. Immediately after dealing with a question about how shutting down the program was the right thing to do, DOD spokesman DiRita took a question about how the NYSE had ticked up 85 points in one hour in response to rumors about the capture of Saddam Hussein. "Hellllooooo!" The NYSE is a futures market in constant operation, only this one is a little more difficult for offended Senators to shut down for political profit.

Luckily, only $600K of the planned $8 million DARPA investment in PAM had been sunk - but the quotes in the news from the legislators made it sound like the $8 million had already been used.

I do have to agree that there were some skewed incentives involved in PAM, and that perhaps more analysis was needed on political implications of this particular application of the technique. Manipulation through deliberate acts of terrorism make this an easier market to control for the perpetrators than controlling the outcome of an election or movie-goers viewing choices. But the fundamental idea behind the market remains sound - if you have money on the line, you tend to make a better guess. Any arguments about the technique itself, well, are just a load of pork bellies.