Statistical groups elicit and incorporate the views of group members. Deliberating groups are supposed to do the same. A very different way to aggregate information is to rely on prices and on the signals that prices give. To see why that method might work, consider the familiar informal challenge when people disagree on some question: “Want to bet?” The point of the challenge is to suggest that the speaker is really confident of her judgment, enough so as to ask the person with whom she disagrees to back her conviction with money. Much of the time, the challenge is successful in the sense that it operates to establish to all concerned that one or another belief is weakly held. When people ask for a bet, they are bypassing deliberation and offering a good clue about the confidence with which they hold their views. But if we want to get access to many minds, we might see if economic incentives can be used far more formally and systematically.
Prices are a product of the independent judgments of many people, each with his own dispersed beliefs, judgments, and tastes. Of course, markets are very far from perfect, and I devote some discussion to their errors, which can be egregious; but much of the time, they do an excellent job in aggregating privately held information. Suppose that we are inclined to celebrate the information aggregating properties of markets. lf so, it would seem plain that if we are attempting to improve on the answers pro-
duced by statistical averages and deliberating groups, we might consider an increasingly popular possibility: Create new markets, Prediction markets, in some ways a recent innovation, have proved remarkably successful at forecasting future events. They do far better, in some domains, than deliberating groups. Such markets are worth sustained attention, in part because they offer important lessons about how to make deliberation go better or worse, and in part because they provide a useful model for any private and public organization that seeks access to many minds. Prediction market theorist Robin Hanson has, in fact, coined the term “futarchy” for a system in which such markets play a large role in governmental decisions. A central advantage of prediction markets is that they give people the right incentive to disclose the information they hold. People stand to win money if they get it right and to lose money if they get it wrong. Recall that in a deliberating group, members may have little incentive to say what
they know. By speaking out, they provide benefits to others, while possibly facing high private costs. Prediction markets
realign incentives in a way that is precisely designed to overcome these problems. People can capture, rather than give to others, the benefits of disclosure. Because investments in such markets are generally not revealed to the public, investors need not fear that their reputation will be at risk if, for example, they have predicted that a company’s sales will be low or that a certain candidate will be elected president.
With prediction markets, the use of private information will be reflected in prices. Suppose that a prediction market suggests that it is likely that intelligent extraterrestrial life will be discovered within the next year. Suppose that for a $100 bet, you can receive $200 if you predict that no such discovery will be made. This market’s prediction is preposterous (I speculatel), and skeptics are likely to make money if they are willing to bet otherwise. (Want to bet?) Of course, many prediction markets raise harder questions. The key point is that those with specialized information can gamble as they see Ht. People who
do not want to participate, because they have no idea what to predict, have every incentive to stand aside. To be sure, some
people might refuse to participate, not because they lack information, but because they lack the money to gamble. But at least it can be said that such markets give people a powerful incentive to reveal what they know. In these crucial ways, prediction markets have large advantages over surveys. Suppose that the question is how many children I have. A survey of numerous people is far less likely to produce the right answer than a market, in which informed people can make money. What is true for mundane or personal questions is true for larger questions as well, including the likelihood of natural disasters, changes in the economy, the outcome of an election, and the success of a movie. Recall that under the Condorcet Iury Theorem, the average vote of a large group will be wrong if most group members are likely to err. In a prediction market, the existence of incentives greatly increases the likelihood that each investor will prove to be right. Those without information will not participate; those with a lot of information will participate a great deal. Crucially, the problems that infect deliberating groups are reduced in prediction markets.
Excerpts taken from: Sunstein, Cass R. „Infotopia“