What is the election market telling us?

Professor Bainbridge posts on this, in response to mine and others. Specifically, what does it mean that Bush futures are at 70?

Here's some thoughts (keeping in mind that I was an English major):

1. Those who are buying or holding think that the expected value of a Bush victory is more than .70, where a win is worth $1.00 and a loss is worth 0. Note that they might all think that the expected value is 1, or they might all think it's .71, and they may hold these views strongly or weakly.

2. Those who are selling think that the expected value of a Bush victory on this basis is less than .70. Note that they might all think that the expected value of a Bush victory is 0 or they might all think it's .69 and they might hold these views strongly or weakly.

3. Everybody else either thinks that the expected value of a Bush victory is less than .70, which again might be 0, and might be a strong or weak opinion, or may have no opinion at all, or not participate in this type of market.

So where does this leave us? I think it's a fair guess that the traders' expectations are evenly distributed, so that the average expectation of all participants in this market is of a .70 chance of a Bush victory, and that 2/3 are within one standard deviation of this number.

Note that this leaves out those who don't trade, and this group might be biased toward speculation, economics, or other things that make them non-representative.

Do these caveats detract from the value of these markets? The answer is, compared to what? What some guy (non-blogger of course) in his pajamas says when a pollster calls him and asks him a bunch of questions he hasn't been thinking about?

And now, let the nasty comments about my innumeracy begin.

Comments

[Steven Kurson]

James Surowiecki's book "The Wisdom of Crowds" has a very interesting discussion of prediction markets, including the IEM, Tradesports, and some of the internal markets that corporations have used to aggregate information and forecasts. He argues, much as you do here, that the price of a state-contingent contract (like the ones offered in the IEM) reflects the weighted-average judgment of all the participants in the market.

Interestingly, "participants" would also include those people who are thinking about trading but decide not to because, crudely speaking, they believe the price to be accurate. Those would be many of the people in your group #3.

Surowiecki shows, pretty convincingly as far as I'm concerned, that the collective judgment of a group of interested problem-solvers (like, in this case, the market participants, but this is true not just in markets but a whole host of other scenarios) is, over time, going to be superior to the judgment of just about anyone in the group. So there's real reason to think that the IEM and Tradesports are offering valuable forecasts.