Presidential Futures: Pools of Ignorance

These markets create plenty of buzz, but don't put your money in them, says Barry Ritholtz, Maxim Group's chief market strategist

Voters have plenty of resources to find out who's leading the latest polls, President Bush or Senator John Kerry, or which states are leaning blue or red. One is the Presidential futures markets, which provide a way for investors to bet on who they think will win the Oval Office. Essentially, investors can put real money in and trade contracts for each candidate -- or for states and electoral votes. Some of the real-money exchanges, such as the Iowa Electronic Markets, claim they can predict the election outcome.

That may be true for some races, but these exchanges shouldn't be taken too seriously, says Barry Ritholtz, chief market strategist at New York investment firm Maxim Group. "They're fun, and that's all they're supposed to be," says Ritholtz, who compares the Presidential futures exchanges to the popular fantasy-football and fantasy-baseball leagues.

"I get very uncomfortable when I see people talking about making investment decisions based on stuff that, to me, are a small and fun little thing. It's goofy, but nothing you should bet real money on."

BusinessWeek Online's Karyn McCormack recently spoke with Ritholtz about these exchanges. Edited excerpts of their conversation follow:

Q: How big are the Presidential futures exchanges?


Two are generating the most buzz these days. They have a tiny amount of money invested. As of late July, the Iowa Electronic Markets had about $16,000 invested in the "Presidential Winner-Takes-All Market" contract. Someone described it as "a bunch of college students playing with their lunch money." To some degree, there's an element of truth in that.

In the Intrade market, $2 million is invested in the biggest contract, the "President George W. Bush" contract.

They are so utterly dwarfed by the massive size of the U.S. markets. The domestic-equity market is $13 trillion. The U.S. Treasury market is about $5 trillion, and the corporate-bond market is another $9 trillion. If markets have any ability to allow us to try and figure out most likely outcomes, the basis of it is that they're so massive, and there's so much money at stake.

Q: Do you think the Presidential futures markets can predict the outcome of the election?


The underlying thesis -- markets as a predicting mechanism -- is flawed, and here's why. First of all, I don't think we can collectively pool our ignorance and create wisdom out of that. If you put 10 million people in a room and none of them know what's going to happen in the future, you don't generate any sort of sage prognostication.

However, when you look at the stock markets, and when you learn to read the tea leaves, you can glean a little insight as to what the best and brightest investors actually believe is happening. The problem is, it's so subjective, and unanticipated events are always leaping up. Nobody really anticipated the Asian virus that had a very deleterious affect on the economy and markets in late 2002 and early 2003. There's simply no way to anticipate an outbreak like that.

The whole concept of the markets forecasting the future is dubious. The old joke is the market predicted 10 of the last 4 recessions. And it's very much true.

Q: Can the futures exchanges be considered a leading indicator?


When you look at the actual way these things work, they're a lagging indicator. They're not a leading indicator telling you what's going to happen. These futures very much track the polls. And the polls are three, four, or five days old. So people are betting based on a poll that came out yesterday that's telling you what people were thinking a week ago. So you end up with this very rearward-looking analysis.

The perfect example of a leading indicator is the yield spread in the bond market. When the short-term interest rate gets inverted -- when you actually get paid more money to lend money short term than long term -- that invariably is a sign that a recession is coming. It's historically accurate.

Q: Did the futures exchanges predict who would win the debates?


The day before the first debate, the Iowa "Presidential Winner-Takes-All Market" futures were enormously spread and gave Kerry essentially no chance. The incumbent looked like a lock. Obviously, that just wasn't the case.

Right before the Democratic and Republican conventions, none of the futures exchanges accurately foretold the sort of bounce either candidate would get. It was very obvious they were looking at what happened two weeks before the convention as opposed to looking forward, and saying, "we're looking for a bounce for both candidates."

The problems with the future exchanges are: The underlying thesis that markets can predict the future is dubious. Second, even if it's true, there's such a tiny amount of money at stake. Third, they seem to be a lagging indicator because they're tracking polls. And I think it has become clear that the polling data, especially by some researchers, has been very misleading.

Q: Are any polls accurate?


When I look at a sector or stock, I know which analysts have been very good in the past. You want to go with a guy with a good track record. When I look at polling, it's the same thing: Who was the most accurate pollster in 2000, and who was the least accurate?

Look at the Bush-Gore election, Gallup was predicting a double-digit lead for Bush, when in fact, Gore ended up winning the popular vote by 540,000 votes. That tells me there's a flaw in Gallup's methodology that has been consistent from election to election. It turns out that [John] Zogby was the guy who nailed it dead on the head. (Editor's Note: Ritholtz's assertion isn't quite correct. Gallup's final poll before the election gave Bush 48% of likely voters and Gore 46%. Zogby's final poll had the race at a 46%-46% dead heat. In statistical terms, both pollsters had the race a tie.)

Q: It looks like a dead heat right now.


We have an electorate that hasn't made a clear choice. Everyone remembers how awful it was in post-November election in 2000, when for the first time in my memory, you didn't know the outcome of the election. That was certainly a little disconcerting. I think there's a reluctance to live through that again.

If you look at the history of reelection campaigns, they're rarely close. The history of Presidential politics is either the incumbent is either approved by a significant percentage of the population or disapproved.

I believe we're not going to have another close election. It's hard to believe that there are so many undecided voters out there. If you haven't decided at this point, do your homework and make your choice. I get a sneaking suspicion that the undecideds are far less undecided than they've let on -- they are just taking advantage of the opportunity to be wined and dined by the media and the pollsters.

As we progress to Nov. 2, I would expect to see the election break one way or the other. I heard someone say the other day, "I'm going to hold my nose and pull the lever for this candidate or that candidate." And I think that's what's going to happen. I'll be very surprised to see this election as close as the last one.