Study finds Kalshi prediction markets as reliable as Wall Street on Fed bets

Kalshi, the prediction site known for turning headlines into tradable bets, is now being taken seriously by economists. A new study shows that Kalshi’s forecasts on Fed interest rates and economic numbers are just as accurate as the ones coming from Wall Street.

The paper tracked data from 2022 through June 2024 and was written by three economists, one of them from the Federal Reserve.

The research found that Kalshi’s traders matched professionals like the ones surveyed by the New York Fed. In one key case, Kalshi actually beat them. That was when the Fed made a surprise 0.5-point rate cut in September 2024. While Wall Street was caught off guard, Kalshi traders already had it priced in.

Study shows Kalshi matched experts on rates and inflation

This paper isn’t peer-reviewed yet, but it landed right before another Fed meeting. On that event, Kalshi and Polymarket both showed a 99% chance the Fed would hold steady.

The fed funds futures market gave it a 97.2% chance. Every single one of the 92 economists tracked by Bloomberg also expected no change.

The study also compared how well Kalshi did on inflation and unemployment numbers. The traders on the platform were just as close to the Bloomberg survey group.

And on one key metric, headline Consumer Price Index (CPI), Kalshi beat the professionals in a way the authors said was statistically significant.

On Kalshi, users trade simple ‘yes’ or ‘no’ contracts that settle at $1. If a contract trades for 32 cents, that means the market is saying there’s a 32% chance of that thing happening. Many economic contracts are based on whether something like CPI will come in higher than a certain number.

The study found that the most likely outcome on Kalshi, what’s called the modal forecast, was accurate by the night before most Fed decisions in that time period. Even when the Fed didn’t do what people expected, Kalshi’s prediction still held up.

Kalshi gives constant updates economists can’t match

Jared Katz, one of the authors and a Ph.D. student at Northwestern, said the reason Kalshi works well is that it gives more than just one number.

“The real advantages are you’ve got a distribution instead of a point estimate, and you can look at how it responds very quickly after events,” he said. “It’s even better that the forecasts are pretty accurate.”

Katz wrote the paper with Jonathan Wright from Johns Hopkins and Anthony Diercks from the Fed. Wright said these markets work because people are putting real money into their opinions. “The overarching theory is the wisdom of the crowds getting information from lots of people aggregating their beliefs,” he said.

The idea isn’t new. Earlier studies showed that even small prediction markets can do better than experts. That’s part of what helped Kalshi beat regulators in 2024. They won the legal right to offer contracts in elections. Not long after, they added bets on sports and entertainment too.

Kalshi still faces bias risks, but looks stronger in finance

The study didn’t pretend Kalshi is perfect. Some researchers found that prediction markets tend to overpay for low-chance events. One wild example? At one point, traders gave a 4% chance that Jesus would return to earth that year.

But a different review done by a Coinbase engineer looked at Kalshi trades and saw that the long-shot bias was weaker in the platform’s finance-related contracts. It was mostly the sports and entertainment side that showed those flaws.

Still, for Wright and the others, prediction markets like Kalshi are tools they wish they had a long time ago. “I and others are super fascinated by this because we’ve always wanted to get markets for the things that economists care about, and those don’t really exist,” Wright said.

By now, Kalshi has shown it can compete. It has matched economists, outperformed them on CPI, and picked up surprise Fed cuts. It gives live updates, shows all possible outcomes, and lets anyone trade on real-world data.

Claim your free seat in an exclusive crypto trading community – limited to 1,000 members.