Recession odds 2026–2027: what Kalshi and Polymarket are pricing in
Prediction markets give a US recession in 2026 only a 4–6.5% chance, but about 21% for 2027. What the markets measure, how they resolve, and the data that would move them.
Prediction markets see little chance of a US recession starting in 2026 and a meaningful but minority chance in 2027. On 16 October 2026, Polymarket's "US recession by end of 2026?" traded at 6.5%, Kalshi's "Recession this year?" at 4%, and Kalshi's "Recession in 2027?" at 21%.
Snapshot from 16 October 2026. Live: Polymarket 2026, Kalshi 2026 and Kalshi 2027.
Recession odds at a glance
| Market | Platform | Price |
|---|---|---|
| US recession by end of 2026 | Polymarket | 6.5% |
| Recession this year (2026) | Kalshi | 4% |
| Recession in 2027 | Kalshi | 21% |
| UK recession in 2026 | Polymarket | 6.5% |
| Japan recession in 2026 | Polymarket | 4.5% |
Why 2026 is priced so low
With a quarter of the year left, a recession "in 2026" would have to show up in data that is mostly already known, or start in the final months and be confirmed later. Markets need evidence, not just worry, and the economic data through the summer has not shown the broad decline a recession call requires.
For comparison, the US has had about a dozen recessions since 1945, roughly one every six or seven years. That base rate works out to around 15% a year, which is why a full-year 2027 market at 21% reads as "somewhat elevated risk", not "recession expected".
What counts as a recession in these markets
This is where people get caught out. There are two common definitions, and markets pick one in their rules:
- The NBER's official dating. In the US, the National Bureau of Economic Research decides when recessions start and end. It looks at jobs, income, spending and output, and often announces a recession months after it began. Kalshi's recession markets are tied to this official dating.
- Two consecutive quarters of falling GDP, the "technical recession" used in headlines and in many other countries. Some markets use GDP data, or GDP data or an NBER call, whichever comes first.
A market can therefore resolve "No" even in a year that felt like a recession, or take a long time to settle. Read the rules on the market page for the definition and the deadline.
What would move recession odds
- Jobs reports. A jump in unemployment is the classic early warning; the "Sahm rule" flags a recession when the three-month average jobless rate rises half a point above its low of the past year.
- GDP releases, especially a negative quarter.
- The Fed. Markets on 16 October gave an 86.5% chance that the Fed holds rates on 28 October and 13.5% to a hike, a sign traders are more focused on inflation than on a downturn. Live: Fed decision in October.
- Shocks: a long government shutdown, a financial-market crash or a spike in energy prices.
All of these are on our economy odds page and the resolution calendar, which lists data releases and the markets they settle.
Prediction markets vs economists
Surveys of economists and models based on the yield curve publish recession probabilities too. Markets differ in two ways: the number updates every minute, and traders lose money when they are wrong. That does not make markets always right; they missed turning points in the past like everyone else. Their advantage is speed and a clear, single number. More on what prices can and cannot tell you: how do prediction markets work.
Recession odds: frequently asked questions
What are the odds of a recession in 2026?
About 4% on Kalshi and 6.5% on Polymarket on 16 October 2026.
What are the odds of a recession in 2027?
About 21% on Kalshi on 16 October 2026.
How does Kalshi decide if there was a recession?
Its recession markets follow the NBER's official dating. Check the market rules for the exact deadline, because NBER announcements can come long after the fact.
Can I bet on a recession?
Yes. Buying "Yes" on a recession market pays $1 per contract if a recession happens under the market's definition; buying "No" pays if it does not.
