Jim Lindgren and Orin Kerr are having an inter-conspiracy spat over at Volokh regarding the performance of the Supreme Court nomination trading market. Kerr started the fracas by criticizing the market. Cf. Lindgren's response here; Kerr's reply here and (more pointedly) here; Lindgren calling Kerr "incorrect and misleading" here; Kerr's comparing tradesports to a gossip page here; and Lindgren's final, very informative, post here.  Whew!  And I thought our discussion of Wisconsin's policy on damage caps was extensive!

The gravamen of the dispute, for those who don't feel like boosting the Conspiracy's already heady circulation figures, seems to be how to judge the success of the trading market.

• Kerr shows that the market followed the conventional wisdom lock-step throughout the day, reflecting Clement as a 88% chance early on, and Roberts had (at times in the day) a 1% chance of being the nominee.  He argues that in the context of this particular choice, which was solely in Bush's control, market forces are unlikely to be particularly informative.  That is, unlike, say, political markets which function to bring together and weigh large amounts of data, nomination markets function to bring together and weigh lots of fluff - folks talking with no basis for their opinion.  Kerr's implicit point is that it simply isn't true that a few days ago, A.G. Gonzales had a 50% chance of being the nominee, nor that Judge Clement had an 88% chance - in fact, both of those individuals had no chance at all, and the market for them was bubble-like.
  
• Lindgren points out that (1) the market pushed market in advance of the news networks (by about an hour); and (2) there is value to a mechanism that produces the best group estimate of the likelihood of a future event.  He also discounts (common) criticisms of these trading markets as insufficiently liquid and serious. (Go here to download Levmore's interesting article on the topic.)

Lindgren summarizes his position as follows:

I am claiming that markets (however "good" or bad they are in absolute terms) should be better than experts on balance, or at least better than experts who lack actual first-hand knowledge of the forthcoming decision. So Orin and I may be talking past each other. I am asserting that I would expect a comparative advantage for markets over experts; Orin is questioning whether markets would "do a good job" predicting individual decisions compared to group ones.

This seems to me to be a hard problem.  I think that markets are "better than experts on balance" in evaluating this sort of problem, if  by "better" you mean "more quickly responsive to new information" and "experts" you mean "a randomly selected talking head"  and "on balance" you mean "most of the time."  Which is to say, if I want a neutral view of the current and commonly prevailing wisdom on who the next nominee/president/dance fad/movie hit will be, I look first to the trading markets.  All of which suggests that I think Lindgren has the better of this particular argument.

Of course, if I want guarantees on any issue, I ask one of my very smart colleagues what he thinks.  His constant, and constantly reassuring, response:  "It depends."

Incidentally, the instant conventional wisdom on Robert's confirmation is that it is merely 80% likely.  I would have guessed something much higher.  What do they know that I don't?

[UPDATE 7:35 PM: Original post corrected to clear up some typos.]

| Bookmark

TrackBack URL for this entry:
https://www.typepad.com/services/trackback/6a00d8345157d569e200d8344e3e5253ef

Links to weblogs that reference Trading Markets at Volokh: