3 ms·

Here are some problems with prediction markets that Tyler Cowen enumerated in 2006: https://marginalrevolution.com/marginalrevolution/2006/03/why_dont_busine.ht

Here are some problems with prediction markets that Tyler Cowen enumerated in 2006: https://marginalrevolution.com/marginalrevolution/2006/03/why_dont_busine.html https://marginalrevolution.com/marginalrevolution/2006/03/wh... That was before machine learning became a thing. I wonder if ML can detect patterns in voting behavior to downrank/uprank certain predictions to get closer to the likely truth. Obviously the more liquid the market is the better. Edit: Someone downvoted me but I'm not sure they're aware that Robin Hanson (the author) is a big advocate of prediction markets, which are also a central feature of the original post. I think highlighting the issues with prediction markets and speculating about a solution is germane to this discussion, especially since prediction markets can be very valuable for society.

pjscott 6y ago
Those look very specific to using employee-only prediction markets to guide decision-making within a single company. Which objections, if any, do you think apply to Hanson's proposal for tradable archaeology prize obligations?
- keenmaster 6y ago
  Abstracting away from the corporation-specific concerns, there is an incentive problem. Vested interests can seek to lower the usefulness of the market's "price as a signal" by placing large diversionary bets. Granted, this is relatively less likely to be a problem in the archaeology space. What definitely may be an issue is lower liquidity. Every time I see a post like this I think "where are all the prediction markets?" Regulations are one reason, but that's far from the only reason. If information/prediction markets are enhanced using new technology, regulators will begin to respect them more and allow for widespread use in non-academic contexts.