IF YOU HAVEN'T HEARD

of the Iowa Electronic Markets (IEM) yet, you probably will soon, for one reason alone: this year's hotly contested presidential election. Forget the polls, which shift from day to day based on who frowned or sighed the most during the latest debate. Since its inception in 1988, the IEM has predicted the outcome of the presidential elections more accurately than 76% of the polls it follows, according to its own research.

There are several good reasons for that, says Thomas Rietz, associate professor of finance at the University of Iowa and a board member of the IEM. One of the biggest, he says, is enlightened self-interest (also known as greed). While there's nothing stopping a poll respondent from bending the truth, he says, "Our traders are putting their money on the line with their forecasts."

Traders can invest up to $500 in any individual futures market on the IEM, a rule designed to limit the influence of any single individual.

The details of how the IEM works are pretty complicated you can learn more on its Web site. What's important is what it's telling us now about who will win the U.S. presidency on Nov. 2. As of Thursday morning, Bush led Kerry by about four percentage points in the IEM's Presidential Vote Share market, which represents a bet on the percentage of the popular vote each candidate will receive. In the Winner-Takes-All market, a bet on who will win the popular vote, Bush's lead is greater, at 56 cents to 44 cents, as of Oct. 6 (the winner gets $1 for each contract, hence winner-takes-all). That represents a 24% fall-off for Bush, however, since he peaked at 74 cents on Sept. 27, just before the first presidential debate.

Rietz says interest in the IEM is up sharply from the Bush-Gore contest four years ago. "In 2000, in the Winner-Takes-All market, we ran that for 27 weeks and had 263,000 contracts traded. That's roughly 10,000 contracts a week," he says. "From Sept. 1. to Oct. 5, we've traded 317,000 contracts. That's about 65,000 contracts a week."

SmartMoney.com recently asked Rietz about the IEM, the presidential elections and whether or not the president's popularity affects the stock markets.

SmartMoney.com: Why are electronic futures markets more accurate than the polls?

Thomas Rietz: There are several reasons. The first is that we ask a different question than the polls do. The typical poll asks, "If the election were held today, who would you vote for?" In order to profit in our markets, you need to be able to predict how the entire population will vote in November. So the pollster doesn't allow you to take into account the effects of the campaigns, the tendency of people to change their minds and the different things that occur between now and the election.

The second thing is that polls depend on a random sample of voters, and most of them poll up to around 1,000 people, and it's the first folks they get on the phone when they call on a particular night. It's very difficult, in that context, to get a random sample. We don't rely on random samples, we rely on our traders trying to predict what the entire population will do. We have about 3,000 [traders] right now.

The third thing that makes a difference is that when a pollster calls you, you have no incentive to actually tell the truth. Our traders are putting their money on the line with their forecasts.

SM: The IEM predicted Bush would win the popular vote in 2000, which he didn't. What went wrong?

TR: IEM predicted about a two-thirds chance that he would win the popular vote, and Gore barely won. In our Winner-Takes-All market, we predict the probabilities of an event. In this case, it was either Bush or Gore taking the majority of the popular vote. When you predict the probability of an event, what you're going to see is an outcome either it happened or it didn't. We predicted that there was a two-thirds chance that Bush would take the popular vote, but that means we also said there was a one-third chance that Gore would, and that's what happened.

SM: How has the IEM moved since the first presidential debate last week?

TR: Take a look at our Winner-Takes-All market, which is asking what the probability that Bush will win the popular vote is. The Winner-Takes-All graph on our Web site combines contracts and asks what the probability of Bush winning is. We have four contracts in the market now. Two of them correspond to Bush winning. One of them is Bush winning with 50% to 52% of the vote, another is more than 52%. The first graph is the sum of those two, which gives the overall probability of Bush winning vs. Kerry, and that's the vote that most people look at. You can see that Bush peaked on Sept. 27 at 74 cents, which corresponds to a 74% probability of Bush winning the popular vote. That was three days before the debate. On the 28th, he closed at 70, on the 29th, he closed at 65, and on the 30th he closed up at bit at 68. Since the debate he'd fallen off all the way down to 58 on Oct. 3.

SM: Some people have noted a link between how well the stock market is doing and how high George Bush's poll numbers are. That's led some people to suggest that the market wants Bush to win. What do you think?

TR: The Iowa markets are correlated with the stock market as well. The change in Bush's probability to win is correlated with the daily return on the return on the S&P 500 from one day to the next on our markets. As the probability of Bush winning drops, so does the S&P 500. Here are the possible causes for that. The market goes down as the economy goes bad, which means people are more likely to elect Kerry and throw Bush out of office. That's one possible explanation of the correlation. The other explanation is that as people see that Bush is more likely to win, they become more confident and bid up the stock market. Which one it is, I don't know. The causation is anybody's guess.

SM: TradeSports.com, another electronic market, has different numbers from IEM, with 61% for Bush and 39% for Kerry as of Oct. 6. Why the disparity?

TR: You should check to see if TradeSports is predicting the exact same thing we are. They have contracts similar to ours, but they're slightly different. I think they're asking "Who will be president?," and that's not what we're asking. We're asking who will win the popular vote.

SM: This year's presidential race is the most heated in years, for obvious reasons. Have you seen a big increase in the attention the IEM has gotten this year, as opposed to previous presidential elections?

TR: I think so. Our volumes are very high. In 2000, in the Winner Take All market, we ran that for 27 weeks and had 263,000 contracts traded. That's roughly 10,000 contracts a week. From Sept. 1. to Oct. 5, we've traded 317,000 contracts. That's about 65,000 contracts a week.