Political Markets

I’ve referred to the Iowa Electronic Markets on political futures in past posts. The exchange creates contracts that pay either $0 or $1, depending on what happens. For example, a George Bush Presidential election contract will pay you $1 for each contract that you own if Bush wins, and will pay you zero if he loses. Effectively it works out to a form of betting on the outcome of the election. Note that there are other markets that do similar things, such as Tradesports.com.

The puzzling thing to me is that when people are putting their money on the line by betting on one candidate or another, they tend to bet on Bush despite his lag in the polls. For a nice way to see Kerry’s slight but beginning-to-be-noticeable lead in the polls, take a look at this assessment by Chris Bowers at MyDD. My conclusion from these poll results is one that others have also come to recently: if the election were held today, John Kerry would win.

Despite this, however, more people are betting on Bush than Kerry. This chart shows the prices of the Bush and Kerry contracts at the IEM since the beginning of June (Bush is yellow, Kerry is blue):

And this chart shows the price of the Bush contract at Tradesports.com since the beginning of 2003:

While the Tradesports.com contract has taken a pretty sharp fall over the past week, it’s still selling above 50, indicating a slightly better-than-even chance for Bush to win, according to the people betting on this exchange. While on the IEM, participants there had Bush and Kerry neck-and-neck for a few days last week, but recently Bush’s lead has opened up again.

My question is a simple one: why aren’t the political futures markets following the poll results? Do the participants in those markets believe that the poll results are wrong? Do they think that people are going to change their mind and opt for Bush between now and November 2, perhaps because they expect an “October surprise”? Or how about this less likely but more interesting (at least to an economist) theory: maybe the participants in the markets tend to be Kerry supporters and are using the political markets as a way to hedge or provide themselves with some insurance (I’d rather have Kerry win, the thinking goes, but if he loses at least I’ll make some money in the political markets).

I’m sure there are some other good explanations out there that I can’t think of — if you have an idea, let me know. Coming up with an explanation for this discrepancy probably won’t shed light on any significant political dynamics this year (though it might), but regardless, this is an interesting little puzzle to think about.

Kash