The `Election Futures' Market: More Accurate Than Polls?

As The U. Of Iowa Goes, So Goes The Nation?

Iowa seems an unlikely spot for the financial world to focus its attention. But the University of Iowa is host to what is arguably today's hottest futures market--an electronic exchange that allows traders to bet on the outcome of the national election.

What began as an Ivory Tower experiment in remote Iowa City is beginning to have an impact on stocks and bonds. When trading at the Iowa Electronic Market on Oct. 23 suggested that Congress might fall to the Democrats, the Dow Jones industrial average took it on the chin. The capital markets are paying attention because the IEM has forecast election results with surprising accuracy since its formation in 1988. It predicted the vote totals of the past two Presidential elections within two-tenths of a percentage point, outperforming national polls. It also has closely tracked elections overseas, never wavering from Boris Yeltsin, for instance, as he won reelection in Russia.

The market operates like a stripped-down model of the Chicago futures pits, except that all trading is conducted on the World Wide Web at Participants can open accounts with up to $500 and trade any of 16 contracts. Some 7,000 players have more than $200,000 riding on them. The most popular play is a winner-take-all bet on Clinton vs. Dole. Each contract has a Clinton unit and a Dole unit, which together costs $1 to buy. Traders make their bet by selling the unit they think will lose. When the contracts expire on election day, the winning candidate's units pay $1 and the loser's are worthless. Lately, traders have been snapping up Clinton and dumping Dole. In the lively aftermarket for units, Clinton was selling for 95 cents as of Oct. 29 and Dole for 5 cents. So if Dole pulls an upset, his supporters in the market can win a dollar for every nickel they wager.

That's not the only Presidential contract. Traders in the "vote-share" market are betting Clinton will win on Nov. 5 by a wide margin--though not as wide as some polls indicate. From early March to late October, Clinton units have climbed from 51% to 57%. Rather than winner-take-all, traders who paid 51 cents would realize 57 cents per unit at expiration if Clinton gets that much of the vote. On the other hand, Dole holders have watched their contracts sag from 49% to 43% over the same period. In contracts involving House and Senate control, Iowa traders riled the stock and bond markets by bidding up the value of Democrats-take-the-House units to 44 cents--suggesting a 44% chance of Gingrich & Co. losing control of the next session. As of Oct. 29, though, the IEM's view on Democrats seizing power had retreated to 37%.

TRACK RECORD. Interesting, but is it a legitimate indicator? Some major-league traders think so. The idea that markets work better than polls appeals to folks toiling on trading floors. Bond analysts at Merrill Lynch & Co. and Lehman Brothers Inc. follow the Iowa exchange, and some market professionals have joined in the betting. At heart, traders see value in subjecting political winds to market forces, and even Iowa's small stakes are an improvement on mere poll responses: "They're putting their money on the line, and that means something," says Chicago Board of Trade Chairman Patrick Arbor. Pollsters, meantime, typically have problems predicting turnout and getting representative samples, explains Robert Forsythe, associate dean at Iowa's business school and a driving force behind the market. Voters who answer polls sometimes lie about their preferences or whether they vote at all.

Even as advocates hale free markets in action, critics attack the ballot-box bourse for its short history, thin capitalization, and alleged tendency to veer out of sync early on in the election cycle. Yet those gripes hardly matter if the market's overall performance is good--and so far, the record favors Iowa. "We've yet to be disappointed," Forsythe boasts. While the Iowa market is no doubt unrepresentative, it rewards those who are good at predicting elections. "The basic idea is right. People will do what's economically rational," notes Martin S. Fridson, chief high-yield bond strategist at Merrill Lynch.

So if the Iowa market is gaining credibility, isn't it also inviting manipulation? Why not sell short stock-index futures, then bet your Iowa account on Democrats retaking the House in the hope that your order will change the political odds and frighten Wall Street again? To be sure, the potential risks have caused the market's founders some jitters. In 1992, they invited the Commodity Futures Trading Commission to regulate it. They also take comfort in the $500 trading limit, which curtails the ability of single players to move prices: An avid Perot supporter tried it in 1992, but the market swung back within hours.


In the beginning, the IEM was aimed at providing hands-on experience to students hindered by their school's "geographic disadvantage"--a polite way of saying that, market-wise, it's in the middle of nowhere. Today, the political market is set to grow. While its founders scoff at the Brazilian hedge-fund manager who tried to open an account with $12 million, they see nothing wrong with raising the stakes to, say, $5,000 per account. And since demand is strong for the Iowa approach in Russia, Turkey, and other nations where polling remains primitive, "You're naturally inclined to think global," says Gary C. Fethke, Iowa's B-school dean. Today, Iowa. Tomorrow, the world.

Before it's here, it's on the Bloomberg Terminal.