Prediction markets

Prediction markets are speculative markets created for the purpose of making predictions. Participants trade contracts whose payoff depends on the outcome of future events—elections, policy decisions, product launches, scientific discoveries—and the market price aggregates the participants' beliefs into a probability estimate. The theoretical justification comes from the efficient market hypothesis: if participants have financial incentives to be correct, and if information is dispersed among many traders, the equilibrium price should reflect the best available estimate of the true probability.

Prediction markets have been deployed in corporate settings (Google's internal markets forecast product deadlines), political settings (Iowa Electronic Markets predict elections), and scientific settings (markets on replication outcomes). Their track record is mixed: they often outperform polls and expert panels, but they are vulnerable to manipulation, thin trading, and the same cognitive biases that afflict individual forecasters. Most critically, prediction markets predict outcomes, not the affective consequences of outcomes—a distinction that matters when the goal is not merely to know what will happen but to know how people will feel about it.

The deeper question is whether prediction markets are tools for discovering truth or tools for manufacturing consensus. When a market price becomes public, it can influence the very events it is trying to predict—a reflexive loop that complicates the interpretation of market accuracy. The market does not just aggregate information; it produces information, and the information it produces is not always veridical.