The Chronicle of Higher Education

                            August 15, 2003, Friday

SECTION RESEARCH; Pg. 16

LENGTH 458 words

HEADLINE Defending the 'Terrorism Futures' Market

BYLINE DAVID GLENN

BODY

   "The idea of a federal betting parlor on atrocities and terrorism is
ridiculous and it's grotesque," declared Sen. Ron Wyden (D-Oregon) last month.
The project that so exercised him was the Policy Analysis Market, a trial
program financed by the Pentagon's Defense Advanced Research Projects Agency.
The market was designed to allow participants to place conditional bets, such
as If there is a coup in Jordan in the first quarter of 2004, then economic
activity (or terrorist casualties) in the United States will increase (or
decrease) in the second quarter. Players would have staked their own money, and
could have won as much as $100 per correct prediction. The day after Senator
Wyden made his comments, the agency shut the program down.

   Robin D. Hanson, an assistant professor of economics at George Mason
University, designed the market's combinatorial betting system and helped select
the policy questions at stake. He believes critics like Senator Wyden were
grievously mistaken.

   Q.Why might an "information market" provide better guidance than reports from
the CIA?

   A.I have a sort of general skepticism about experts. There are literally
millions of people who know relevant things about what's likely to happen in the
Middle East. There are only a few hundred, or perhaps a thousand, credentialed
experts, who have a small tap onto some of those things. The hope is to get
input from other people, and to create a forum that encourages people to
contribute when they think they know things, but to shut up when they know they
don't. Losing money is one way to do that.

   Q.What about the concern that terrorist organizations might use such a market
to profit from their attacks, or to mislead their enemies about what they're up
to?

   A.The market wouldn't have involved predictions about specific attacks -- the
question involved aggregate casualties on a quarterly basis. It's the difference
between predicting the crime rate and predicting which bank will be robbed next.
... This was a research project, and we would have been glad to change elements
of it in response to public concerns. No one ever came to us.

   Q.How do you reply to the objection that betting on calamity is simply in bad
taste?

   A.Intelligence is fundamentally in bad taste. Intelligence is about paying
people to tell you about bad things that might happen. If you're repelled by the
idea that somebody might profit from having told somebody about a bad thing that
might happen, you're repelled by intelligence.

   In the name of intelligence, people lie, cheat, steal, and kill. Compared to
those sorts of things, our proposal was very mild. We were simply going to take
money from some people and give it to others based upon who was right.