Prediction Markets and the Economics of Crowd-Sourced Forecasting

In 1945, Friedrich Hayek wrote about how central planning fails due to the knowledge problem; namely, that no individual, nor any central decision maker, can possibly obtain sufficient information to make effective economic choices for the entirety of society. He suggested that the solution to this problem lay in prices, which condense the dispersed and fragmented information that exists in society into a single number that allows individuals to take action.

Over the last 80 years, prediction markets have applied the same argument as Hayek when making forecasts. Instead of relying on the input of a panel of "experts" or using econometric modelling to price future outcomes, prediction markets allow thousands of individuals to trade contracts tied to future events, with the prices of these contracts reflecting the aggregate opinions of participants. Prices are updated in real time as new information about outcomes becomes available.

The volume of trades in prediction markets is staggering when compared to historical data from the last decade. For example, during the week beginning February 26, 2026, prediction markets traded over $4.3 billion worth of contracts across more than 266,000 active contracts, covering decisions by the Federal Reserve, cryptocurrency prices, geopolitical events, and other developments. For economists, prediction markets provide a real-time laboratory to evaluate information aggregation, behavioral bias, and price discovery.

How Prediction Markets Aggregate Information

Prediction Markets function based on a simple principle: People can buy/sell contracts in anticipation of specific future events. If a contract sells for 70 pence, you can interpret this to mean that the prediction market will pay you £1 if the outcome happens (70% chance) or nothing if the outcome doesn’t happen.

Participants have a direct financial motivation to trade based on their beliefs about whether the market has over-priced/under-priced outcomes. If someone believes the market is underpricing an outcome, they can purchase an inexpensive contract and if they are correct, will make a profit. If a participant believes the market is overpricing an outcome, they can sell their contract. The constant effort of buyers and sellers will be reflected in a market price that, theoretically, accurately reflects the best estimate of probabilities of the event.

Current large prediction markets are Kalshi (CFTC-regulated US Exchange) and Polymarket (decentralized platform). Each predict over 36 million transactions per week with approximately $780 million total open interest. The DeFi Rate tracker aggregates live volume, open interest, and category-level data from both, providing a side-by-side comparison of where trading activity is concentrated.

The Efficient Market Hypothesis, Applied to Events

Conventional markets have long been seen as an example of the Efficient Market Hypothesis (EMH). According to the semi-strong form of EMH, all available public information is included in asset pricing and will be reflected in prices as long as a fair market price exists for each asset at any given time. Prediction markets leverage EMH by functioning similarly to stock and bond pricing theory; however, they allow participants to trade on events with binary outcomes. For example: Will the Fed cut interest rates in March? Will Bitcoin end the month above a certain dollar value? Is there a possibility of a ceasefire? Etc.

Studies have indicated that prediction markets are effective at aggregating and conveying information. Specifically, a 2002 Penn State study concluded that prediction market prices were more accurate than expert forecasts and polling in United States elections 74% of the time, consistent with the Iowa Electronic Markets, an academic prediction market that has outperformed major polling averages since 1988.

The key distinction between traditional financial markets and prediction markets relates to the way each market is structured. Polling merely asks individuals to express a belief or preference regarding an outcome based on their opinion; however, there is little consequence if their belief or preference turns out to be incorrect. In contrast, a prediction market requires participants to wager their own capital based on their beliefs. The use of financial capital creates a disincentive to express exaggerated beliefs and improves the ability of prediction markets to minimize noise, correct for overconfidence, and enhance information evaluation. As such, the market mechanism is moderately effective at reducing anchoring and availability heuristics, as well as herd behaviour; however, none of these behaviours can be fully eliminated.

Where Behavioural Biases Still Appear

For many years, cognitive biases have been studied by behavioral economists and continue to be prevalent in prediction markets. One well-known cognitive bias is the "favorite-longshot" bias, which occurs when traders assign too much weight to low-probability outcomes and not enough to high-probability outcomes. For example, an event may have fundamentals suggesting that a contract should be valued at 2%, but instead it trades at 5% because traders overvalue the small chance of an upset. This pattern can be seen across many types of events (sports, politics, and finance).

Other behaviors exhibited in prediction markets include cognitive biases related to recency. A particular contract may trade rapidly following a news event before settling back into a rational trading range, for example, after an interest rate announcement by the Fed. There is typically high volatility for two days after the market receives new information, as traders respond to each data release and react based on speculation.

Finally, herding behavior is also present in prediction markets, whereby prices increase as subsequent buyers purchase contracts at higher prices than fundamentals would suggest. This pattern can also be observed in equity markets through momentum trading; disciplined traders can take advantage of mispricing created by herding behavior.

Applications Beyond Finance and Politics

In 2026, there will be many contracts offered beyond elections and interest rate contracts. The geopolitical events category is one of the fastest growing, with millions of dollars traded weekly on ceasefire timelines, military engagements, and leadership changes. These contract prices provide an up-to-date measure of global political risk and can supplement traditional measures such as the VIX and credit default swap spreads.

There is also an increasing number of technology markets, and this category is expanding at a rapid rate. Traders can now place bets on which company will have the leading AI model at the end of the month or whether a particular cryptocurrency will reach a designated price point. Sports contracts represent the largest volume traded per day on regulated exchanges.

Anyone interested in researching these trends across all contract categories can find a breakdown by sector through DeFi Rate’s prediction markets platform, which visually represents volume flow from politics, sports, crypto, and finance in real time.

What This Means for Economic Forecasting

According to Philip Tetlock, the leading expert on superforecasting, the most successful individual forecasters share characteristics similar to those of traders in active prediction markets. They tend to update their beliefs regularly, use base rates appropriately, and avoid committing too strongly to one narrative for extended periods of time. Prediction markets create these habits by design, as they provide trading opportunities each time new information is released, and the aggregate of these transactions produces continuously evolving probability estimates of future events based on current knowledge.

Prediction markets provide an uncommon but extremely valuable tool for today’s economists, researchers, and practitioners. They offer real-time, financially incentivized forecasts that reflect what the entire participant pool believes about likely future events. As volume continues to increase—recently experiencing thirteen-fold growth since last year—prediction markets may become a standard reference alongside traditional economic indicators such as the Bank of England's inflation expectations survey.

The debate has shifted from whether prediction markets should be utilized to how quickly and widely they will be adopted, and whether they will become a fundamental forecasting method used by economists or remain a curiosity at the fringes.