I think I basically get the idea behind prediction markets. People take their money seriously, so the opinions of people who are confident enough to bet real money on those opinions deserve to be taken seriously as well. That kid on the schoolyard who was always saying “wanna bet?” might have been annoying but he also had a point: your willingness or unwillingness to bet does say something about how seriously your opinions ought to be taken. Furthermore, there are serious problems with the main alternative prediction method, which consists of asking experts what they think is going to happen. Almost nobody ever keeps track of whose predictions turned out to be right and then listens to those people more. Some predictions involve events that are so rare or so far in the future that there’s no way for an expert to accumulate a track record at all. Some issues give experts incentives to be impressively wrong rather than boringly right. And so on. These are all good points, and they make enough sense to me to convince me that prediction markets deserve to be taken seriously and tested empirically. If they reliably produce better predictions than the alternatives, then they deserve to win the day.*
But there is aparticularclaim that is made about prediction markets that I am skeptical of. It starts with the well-known idea, usually associated with Friedrich on Hayek, that a major virtue of free markets is that there is all kinds of useful information spread out in local chunks throughout the economy, which individuals can usefully exploit but a central planner never could, which is reflected in market prices, and which in turn cause resources to be allocated efficiently. It then goes on to argue that prediction markets have a similar virtue. As an example, suppose there’s a prediction market for a national election, and you happen to know that Candidate X is more popular in your little town than most people think. There’s no way that some faraway expert could have known this or incorporated it into his or her prediction in any way, but it gives you an incentive to bet on Candidate X, which causes your local information to be reflected in the prediction market price. Lots and lots of people doing the same thing will cause lots and lots of such little local pieces of information, which couldn’t have been obtained any other way, to also be reflected in the market price. But it seems to me that this “Hayekian” mechanism should work a lot less well in the prediction market context than in the standard context. In the standard version, you benefit directly from a piece of local information that only you happen to have. If you know that a particular machine in your factory only works right if you kick it three times on the left side and then smack it twice on the top, then you can do that and directly reap the benefits, and the fact that you were able to do it (i.e., the fact that output in your particular industry is very slightly less scarce than someone who didn’t know that trick would have thought) will be reflected in the market price. In contrast if you’re the only one who knows that Candidate X is surprisingly popular in your little town (say because you’re the mail carrier and you count yard signs along your route), could you really benefit from trading on that information? There are a number of barriers to your doing so. First, there are transactions costs associated with trading. Second, there is garden-variety risk aversion: if you’re risk-averse then you won’t want to invest a large share of your total wealth in this highly risky and urnhedged asset, which means that you won’t bet much and so the price won’t move much to reflect your information. Third, in order to believe that your little piece of local information constitutes a reason to bet on Candidate X, you’d have to believe that the current price accurately reflects all the *other* pieces of information besides yours. In some sense you should believe this: if you thought the price was off and you thought you knew which direction it was off, that would be a good reason to bet against the mispricing. But even if you had no actionable beliefs regarding a mispricing, you might just not have a lot of confidence that all the other information has been aggregated correctly. This would translate into another form of risk, and so risk-aversion would kick in once again. Fourth, there may be uncertainly about whether you really are the only one who knows knows your piece of local information (maybe the paper boy also noticed the yard signs, but then again maybe not). If you’re not sure, then you’re not sure to what extent that piece of information is already reflected in the current price. Again, you might have beliefs about this, and those beliefs might be right on average (though they might not) but it is yet another layer of uncertainty that should have an effect similar to ordinary risk aversion.
I asked Robin Hanson about this once at lunch a few years ago, and we had an interesting chat about it, along with some other George Mason folks. I won’t try to summarize everyone’s positions here (I’d feel obligated to ask their permission before I’d even try), but suffice it to say that I don’t think he foreswore the Hayekian idea entirely as an argument in favor of prediction markets. And there is a quote by him here that seems to embrace it. In any case, I’d be interested to know what he thinks about it. And of course it matters whether or not this Hayekian claim is being made for prediction markets, and it matters whether the claim is correct, because whether or not prediction markets have this additional theoretical advantage should go into one’s priors about their merits before evaluating whatever empirical evidence becomes available.
*The question is not purely an empirical one though. There are issues related to how susceptible prediction markets are to manipulation, how well they’ll work when the people doing the betting about what will happen also have some influence over what does happen, whether they’ll work for rare or distant events, or for big picture questions where in some states of the world there’s no one around to pay out the winnings, and so on. So even a strong empirical finding that prediction markets work in more straightforward settings is not the last word on the subject, which means that there will be a continuing role for theoretical arguments even as more evidence comes in.
Hayekian Prediction Markets?
I think I basically get the idea behind prediction markets. People take their money seriously, so the opinions of people who are confident enough to bet real money on those opinions deserve to be taken seriously as well. That kid on the schoolyard who was always saying “wanna bet?” might have been annoying but he also had a point: your willingness or unwillingness to bet does say something about how seriously your opinions ought to be taken. Furthermore, there are serious problems with the main alternative prediction method, which consists of asking experts what they think is going to happen. Almost nobody ever keeps track of whose predictions turned out to be right and then listens to those people more. Some predictions involve events that are so rare or so far in the future that there’s no way for an expert to accumulate a track record at all. Some issues give experts incentives to be impressively wrong rather than boringly right. And so on. These are all good points, and they make enough sense to me to convince me that prediction markets deserve to be taken seriously and tested empirically. If they reliably produce better predictions than the alternatives, then they deserve to win the day.*
But there is a particular claim that is made about prediction markets that I am skeptical of. It starts with the well-known idea, usually associated with Friedrich on Hayek, that a major virtue of free markets is that there is all kinds of useful information spread out in local chunks throughout the economy, which individuals can usefully exploit but a central planner never could, which is reflected in market prices, and which in turn cause resources to be allocated efficiently. It then goes on to argue that prediction markets have a similar virtue. As an example, suppose there’s a prediction market for a national election, and you happen to know that Candidate X is more popular in your little town than most people think. There’s no way that some faraway expert could have known this or incorporated it into his or her prediction in any way, but it gives you an incentive to bet on Candidate X, which causes your local information to be reflected in the prediction market price. Lots and lots of people doing the same thing will cause lots and lots of such little local pieces of information, which couldn’t have been obtained any other way, to also be reflected in the market price.
But it seems to me that this “Hayekian” mechanism should work a lot less well in the prediction market context than in the standard context. In the standard version, you benefit directly from a piece of local information that only you happen to have. If you know that a particular machine in your factory only works right if you kick it three times on the left side and then smack it twice on the top, then you can do that and directly reap the benefits, and the fact that you were able to do it (i.e., the fact that output in your particular industry is very slightly less scarce than someone who didn’t know that trick would have thought) will be reflected in the market price. In contrast if you’re the only one who knows that Candidate X is surprisingly popular in your little town (say because you’re the mail carrier and you count yard signs along your route), could you really benefit from trading on that information? There are a number of barriers to your doing so. First, there are transactions costs associated with trading. Second, there is garden-variety risk aversion: if you’re risk-averse then you won’t want to invest a large share of your total wealth in this highly risky and urnhedged asset, which means that you won’t bet much and so the price won’t move much to reflect your information. Third, in order to believe that your little piece of local information constitutes a reason to bet on Candidate X, you’d have to believe that the current price accurately reflects all the *other* pieces of information besides yours. In some sense you should believe this: if you thought the price was off and you thought you knew which direction it was off, that would be a good reason to bet against the mispricing. But even if you had no actionable beliefs regarding a mispricing, you might just not have a lot of confidence that all the other information has been aggregated correctly. This would translate into another form of risk, and so risk-aversion would kick in once again. Fourth, there may be uncertainly about whether you really are the only one who knows knows your piece of local information (maybe the paper boy also noticed the yard signs, but then again maybe not). If you’re not sure, then you’re not sure to what extent that piece of information is already reflected in the current price. Again, you might have beliefs about this, and those beliefs might be right on average (though they might not) but it is yet another layer of uncertainty that should have an effect similar to ordinary risk aversion.
I asked Robin Hanson about this once at lunch a few years ago, and we had an interesting chat about it, along with some other George Mason folks. I won’t try to summarize everyone’s positions here (I’d feel obligated to ask their permission before I’d even try), but suffice it to say that I don’t think he foreswore the Hayekian idea entirely as an argument in favor of prediction markets. And there is a quote by him here that seems to embrace it. In any case, I’d be interested to know what he thinks about it. And of course it matters whether or not this Hayekian claim is being made for prediction markets, and it matters whether the claim is correct, because whether or not prediction markets have this additional theoretical advantage should go into one’s priors about their merits before evaluating whatever empirical evidence becomes available.
*The question is not purely an empirical one though. There are issues related to how susceptible prediction markets are to manipulation, how well they’ll work when the people doing the betting about what will happen also have some influence over what does happen, whether they’ll work for rare or distant events, or for big picture questions where in some states of the world there’s no one around to pay out the winnings, and so on. So even a strong empirical finding that prediction markets work in more straightforward settings is not the last word on the subject, which means that there will be a continuing role for theoretical arguments even as more evidence comes in.