http://www.newyorker.com THE FINANCIAL PAGE DECISIONS, DECISIONS by James Surowiecki Issue of 2003-03-24 Posted 2003-03-17
For months, traders on the world's stock and commodity exchanges have been preoccupied with one issue: war with Iraq. If you think that war will be over quickly, you buy stocks. If you believe that it will drag on, you sell them. And if you think that the Middle East will be plunged into chaos you buy gold. For speculators on the New York Merc or the International Petroleum Exchange, every trade is like a wager on what will happen, and when, to Saddam Hussein. For speculators on web sites like TradeSports and NewsFutures.com, though, every trade actually is a wager on Saddam. At each of these sites, markets have been set up where, by buying and selling futures contracts, you can bet on such questions as when war will begin and when Saddam will fall. (The market for "Saddam futures" on TradeSports says that there's a seventy-nine-per-cent chance he'll be gone by the end of April.) There is, perhaps, something ghoulish about gambling on war. (Think of Adam Trask, in John Steinbeck's novel "East of Eden," rejecting as tainted the money his son Cal made speculating on bean futures during the First World War.) But in a sense the NewsFutures traders are only trying to do what op-ed writers, TV pundits, and presidential advisers attempt to do every day: predict the future. The big difference is that the betting markets are far more likely to be right. The market in Saddam futures is the latest version of what are sometimes called "decision markets." They've been around for more than a decade, and in that time their track record in forecasting events has been exceptional. The best-known example is the Iowa Electronic Markets. The I.E.M., which was founded in 1988 and is open to all comers, allows people to buy and sell shares based on the percentage of the vote that they think candidates will get in upcoming elections. At any moment, the price of a candidate's shares reflects the market's collective judgment of how well he'll do. If George W. Bush's shares are trading at 56.4, the market is anticipating that he'd get 56.4 per cent of the vote. The I.E.M. routinely outperforms major national polls. In the last four presidential elections, for instance, almost six hundred different polls were conducted, and the I.E.M.'s market price on the day each of them was released turned out to be closer to the election results seventy-five per cent of the time. And the I.E.M.'s election-eve predictions in those four contests were off by an average of just 1.37 per cent. The Hollywood Stock Exchange, which allows people to speculate on box-office returns, opening-weekend performance, and the Oscars, has also been prescient. Traders' predictions of opening-weekend returns are more accurate than the movie industry's forecasts, and the Exchange has done a good job of foreseeing nominations as well. Last year, its traders correctly predicted thirty-five of the forty Oscar nominees in the top eight categories. And then there's the eerily accurate Foresight Exchange, where speculators can gamble on a wide variety of topics-say, whether Catholic priests will be allowed to marry by 2005, or when physicists will discover a Higgs boson. Why do decision markets work so well? They are extremely efficient at aggregating information and tapping into the collective wisdom of a group of traders, and groups are almost always smarter than the smartest people in them. As in financial markets, the incentive to get the better of others (whether the reward is profit or mere satisfaction) causes traders to seek out good information. The absence of hierarchy-markets don't have vice-presidents-insures that no single person has too much influence and that diverse viewpoints don't get shut out. Decision markets also skirt the political and personal issues that so often clog the flow of information within organizations. Because people are rewarded only for being right, they have no incentive to hide information, pursue agendas, or go along with the crowd. Of course, an organization can't rely on such prophecies unless it is run by leaders who don't claim to have a monopoly on wisdom. Leaders like that are hard to come by, which may be why decision markets have been ignored, for the most part, by corporations and governments. That may be changing, though. Hewlett-Packard has used artificial markets for sales forecasts. Essentially, H.-P. employees bought and sold shares depending on what they thought sales in a particular month would be. The number of people participating was small-never more than twenty-six-and each market ran for only a week, but in the course of three years the markets outperformed the company's official forecasts seventy-five per cent of the time. And then there's the Defense Department, which recently set up something called the FutureMAP program, to investigate and experiment with decision markets. Two companies are already working with the Pentagon to build and test markets that would allow defense and intelligence analysts to speculate on such strategic and geopolitical issues as "the probabilities of specific kinds of failure within our national infrastructure"-that is, what are the most likely targets of another terrorist attack?-and the possible consequences of economic and political turmoil abroad. This may be surprising, given the Bush administration's command-and-control management style, but it's a smart response to the limitations of traditional intelligence analysis-limitations that in the past few years have been painfully exposed. It may be hard to see TradeSports' would-be war profiteers as avatars of the future, but there it is. The future belongs to the speculators, not the spooks.
