Is there research on this? I think an important part of market design should be who to include, and who to exclude? Too often I see people take a reflexive libertarian position on this. But what are other approaches? For example, with a political market prediction market, how would it function if it excluded people who didn't know what the electoral college system was? I can see less dumb money as a discentive for smart people to participate. But markets chasing dumb money also seem to create dumb outcomes, such as the relative capitalization of tobacco companies and car companies compared to many biotech companies.
This isn't yet as formulated as coherently as I'd like, so feel free to rip in with your criticisms, it may motivate me to make it more coherent, although as I type I realize I'd rather be focusing on microsociological concerns to maximize my individual persistence odds.
I think there's a distinction between IQ and having different preferences from you, though they may correlate.
Posted by: TGGP | August 17, 2008 at 06:08 PM
How do you define competency? From your political futures example, we can remove the people who don't know how the college electoral system works, but do we keep the people who know how it works but underestimate the number of electoral votes California has? That's kind of incompetent. What about those traders who know every fact about the voting process but don't follow the news as often as the next trader? That's also kind of incompetent.
Who sets the level of dumbness? You? The exchange? A regulatory body? I think trading that doesn't result in a benefit to one of the involved parties indicates some form of incompetency; however, this dumb money can really only be discovered ex post.
Posted by: mpkomara | August 18, 2008 at 11:57 AM
"I think trading that doesn't result in a benefit to one of the involved parties indicates some form of incompetency; however, this dumb money can really only be discovered ex post."
Your post is filled with the type of good questions that I think should be explored experimentally (if they already have been, I'm interested in links to the results).
I'm surprised you end with such a baldly assertive conclusion. You don't think that's an empirical question?
Posted by: Hopefully Anonymous | August 18, 2008 at 12:26 PM
Incompetent traders generate volume and liquidity. What incentive does an exchange have to exclude them? Perhaps at a minimum they need to exclude the trader so incompetent he is capable of taking down the exchange with a catastrophic mistake. Other than that, I fail to see the benefit to the exchange to exclude them. You suggest "dumb outcomes". What does Intrade care if their markets are poor predictors? Seems like a call out to arbitrageurs to visit their site.
Posted by: mpkomara | August 18, 2008 at 11:59 PM
mpk, I think I anticipate your comment in my first post. Incompetent traders generate volume and liquidity, but they also generate noise that may make prediction markets less accurate.
What I'm arguing for here is experimentation with prediction markets that look at how they fair when restricting various kinds of less competent, literate, or intelligent traders. Or alternatively, if such experimentation has already been done, I'm blegging for information about the results.
Posted by: Hopefully Anonymous | August 19, 2008 at 01:42 AM