Policy Analysis Market
Once again, DARPA’s Information Awareness Office has come up with a wonderful idea to fight terrorism. Unfortunately (and also, once again) the same office has demonstrated such an incredible lack of public relations skills that the same brilliant idea has been canned.
By now, I’m sure that you’ve heard of the proposed Policy Analysis Market, which was designed to create a sort of futures market for terrorist attacks and other such events. To quote from the website’s homepage (which is now only available through the Google cache).
Analysts often use prices from various markets as indicators of potential events. The use of petroleum futures contract prices by analysts of the Middle East is a classic example. The Policy Analysis Market (PAM) refines this approach by trading futures contracts that deal with underlying fundamentals of relevance to the Middle East. Initially, PAM will focus on the economic, civil, and military futures of Egypt, Jordan, Iran, Iraq, Israel, Saudi Arabia, Syria, and Turkey and the impact of U.S. involvement with each.
The contracts traded on PAM will be based on objective data and observable events. These contracts will be valuable because traders who are registered with PAM will use their money to acquire contracts. A PAM trader who believes that the price of a specific futures contract under-predicts the future status of the issue on which it is based can attempt to profit from his belief by buying the contract. The converse holds for a trader who believes the price is an over-prediction Ð she can be a seller of the contract. This price discovery process, with the prospect of profit and at pain of loss [sic], is at the core of a marketÕs predictive power.
At its core, the idea being proposed here is that of using a genetic algorithm to attempt to predict regional instability. (Genetic algorithms have actually been used to simulate stock markets before.) The basic idea behind a genetic algorithm is simple: You take a shotgun approach to predicting the answer to a question. Those rules (or genes) that do a better job of predicting the answer are weighted more heavily than those which don’t, and the process is repeated with the weighting in place. In a market, the feedback is equivalent to how much money each trader has to invest in futures. If you’re right, the amount of influence you have over the market goes up. If you’re wrong, the amount of influence you have goes down. Thus, most of the control over the market is given to those individuals who have the best track record in predicting events.
Unfortunately, DARPA decided to put this plan out on the internet, and to use real money. Both of these ideas have merit in an abstract, academic sense — using the internet opens the market to a wider range of perspectives, and using real money ensures that traders take the market seriously. Unfortunately, those ideas combine to form a public relations nightmare, which we saw explode today. Nothing kills a project like the phrase “Betting on Terror” splashed across every newspaper in the country.
Sadly, this is an almost exact repeat of what happened to the Terrorism (formerly Total) Information Awareness program. What started out as a very logical approach to identifying suspicious behavior using statistical patterns got demonized into visions of a 1984-esque police state.
Much of the problem is very, very bad PR management on the part of DARPA and the Information Awareness Office. The IAO in particular seems not to have realized that any proposal placed on the Internet is instantly subject to public scrutiny, and that using language that may be appropriate in an academic environment may cause a firestorm in a non-academic environment. The result of this political shortcoming is that potentially useful terrorism-fighting systems are getting the axe, and it is doubtful that any of the axed systems will be viable in the near future. Hopefully, DARPA can get its act together before another good idea hits the fan.
Update
The New York Times has exhibited a wonderful degree of schizophrenia on this issue. Yesterday they presented an editorial Poindexter’s Follies, an editorial calling for the immediate firing of John Poindexter and the closure of the “wacky intelligence operation” he runs at DARPA. Today, a separate piece in the business section argues that the futures market was actually a good idea killed by bad press. Determining which section of the paper might know more about how a futures market works is left as an exercise for the reader.
Markets do not always operate perfectly in the larger world of stocks and bonds. The idea that they can reliably forecast the behavior of isolated terrorists is ridiculous. — Editorial staff, New York Times, July 30
Similar markets have been organized to predict shifts in Federal Reserve monetary policy, the outcome of political conventions and sales of consumer products. The results are that markets typically perform at least as well, and generally better, than feasible alternatives, and they are much cheaper to organize. — Hal R. Varian, New York Times, July 31