Why doesn’t business use more prediction markets? They would seem to make sense, right? Bet on ideas. Aggregate information. We’ve all read Hayek.
Hal Varian’s answer is that the most valuable predictions are very sensitive, and you don’t want those to be public. My answer is different: prediction markets are too easy to game. They work decently as long as no one really cares about the answer, but as soon as someone does care—i.e., someone can make a lot of money from influencing the results—then the prediction market will be rigged. The only way to stop that from happening is to make the market expensive, but then you lose almost all your players.
It’s hard to rig the actual market because the cost of rigging it is usually higher than the amount of money you can make from rigging it. But as soon as that cost drops so much as a nano-penny lower, someone, somewhere will decide to screw with it.