What Are the Odds of a U.S. Recession in 2026?
Updated July 21, 2026
By most credible measures, the U.S. has roughly a 1-in-4 chance of entering a recession before the end of 2026. Here's what's driving that number — and what could move it.
Recession talk has been circling Wall Street and Washington for the better part of two years now, but in 2026 it has picked up a new and specific urgency. The question is no longer theoretical. Between a Middle East conflict that briefly closed the Strait of Hormuz, oil prices pushing past $100 a barrel, and a labor market showing its first real cracks, investors and economists are trying to put a number on something that resists easy quantification: how likely is a U.S. recession by the end of this year?
Prediction markets are one of the more honest attempts to answer that question – not because they're always right, but because they have real money behind them.
Updated May 13, 2026
Polymarket odds
~22%
Prediction market
Goldman Sachs
25%
Bank forecast
JPMorgan
35%
Bank forecast
NY Fed model
5.7%
Yield curve signal
Forecaster Comparison — Recession Probability by End of 2026
Polymarket
~22%
NY Fed model
5.7%
Goldman Sachs
25%
JPMorgan
35%
Moody's Analytics
~49%
Kalshi peak (March 2026) 34%
Hormuz spike
Kalshi current (May 2026) ~17.5%
record low
April payrolls +115k jobs
Unemployment 4.3%
Polymarket currently prices the probability of a U.S. recession by end-2026 at around 26%, with the platform noting that March nonfarm payrolls added 178,000 jobs and unemployment held steady at 4.3%. That's a relatively sanguine reading, though the number has moved considerably over the past several months.
Kalshi's recession market jumped above 34% in early March – its highest level since November – following the dramatic rally in oil prices above $100 per barrel after Middle Eastern producers cut output amid the closure of the Strait of Hormuz. The gap between Kalshi and Polymarket at any given moment reflects the fact that these platforms attract different pools of traders and use slightly different contract structures. They're worth watching together rather than in isolation.
To put the current range in perspective: when the contract went live in late September 2025, "Yes" shares were trading in the mid-40s before sliding into the high-20s by late 2025. The improvement in sentiment was real – and then the Hormuz disruption hit.
The mechanics are simpler than they sound. Each contract is structured as a yes/no question – "Will the U.S. enter a recession by end of 2026?" – where participants buy shares in either outcome at prices between $0 and $1. Those prices directly reflect implied probability: a share trading at $0.28 means the market assigns a 28% chance of recession.
What makes this meaningful rather than just speculative is that Polymarket's economy markets have seen over $171 million in trading volume, giving the prices genuine information content. When large numbers of people with money on the line collectively move a price, that movement tends to reflect something.
The definition of "recession" baked into these contracts also matters more than most readers realize. The Polymarket contract resolves "Yes" if either the Bureau of Economic Analysis reports two consecutive quarters of negative real GDP growth, or the National Bureau of Economic Research publicly announces a recession. The NBER definition is notably broader – it incorporates employment, real income, and industrial production, not just GDP. An economy can feel recessionary to ordinary Americans while technically not qualifying under one definition but qualifying under the other.
Prediction markets don't exist in a vacuum. The major banks have their own forecasts, and right now they're telling a story of meaningful but not overwhelming risk.
Goldman Sachs raised its recession probability to 30% following the oil price surge, nudging its full-year GDP growth estimate down to 2.1% and raising its headline PCE inflation forecast to 3.1% by December – while still maintaining that a recession is not its base case. The bank sees U.S. growth cooling to just 1.25–1.75% in the second half of 2026, a level it describes as close to "stall speed."
JPMorgan had already entered 2026 with a 35% recession probability baked into its outlook, citing consumption downshifting in developed markets. JPMorgan's Bob Michele has pushed back against more optimistic readings, arguing that the Iran conflict is not merely an inflation "speed bump" and that price pressures could remain sticky well beyond current forecasts.
Moody's Analytics Chief Economist Mark Zandi had argued that recession odds were near even before the war broke out – a notably more pessimistic baseline than either Goldman or JPMorgan.
Meanwhile, the New York Federal Reserve's yield curve model currently assigns approximately a 25% probability that the U.S. will be in a recession by November 2026. That model has a strong long-run track record – an inverted yield curve has preceded each of the last eight U.S. recessions – though it threw a false positive during 2023–2024, when the curve inverted sharply and no recession followed.
Understanding why recession odds sit where they do requires looking at the specific pressures converging on the economy right now, rather than recycling the standard list of macro indicators.
Energy costs. The Hormuz disruption was the single biggest near-term shock. Goldman expects Brent crude to average around $105–$115 per barrel through early spring before retreating to $80 by year-end – but that assumption hinges on roughly six weeks of supply disruptions. A prolonged conflict would entrench energy costs, crimp consumer spending, and force the Fed into an increasingly uncomfortable position.
The labor market. This remains the most important buffer against a downturn, and it's holding better than many expected. March payrolls added 178,000 jobs with unemployment steady at 4.3%, near historic lows. But February's payrolls actually fell by 92,000 – Goldman economist David Mericle described it as "a reminder that job growth is still too low," with the bank's estimate of underlying job creation sitting barely above zero. One month is not a trend, but February's number was a warning.
Inflation and Fed policy. Goldman estimates that Trump's tariffs have already added more than 70 basis points to core inflation, though underlying inflation net of tariff effects looks more contained – core CPI near 1.75% and core PCE near 2.25%. The Fed is effectively being squeezed: inflation remains above target, which limits its ability to cut rates, even as growth slows. Rate cuts, by most accounts, aren't coming soon.
Consumer spending. This is the variable that ultimately determines whether slower growth tips into contraction. Consumer spending accounts for roughly 70% of U.S. GDP, and the household balance sheet has deteriorated since the pandemic savings buffer was drawn down. Rising credit card delinquencies and higher borrowing costs represent a slow drain that doesn't show up dramatically in any single data release – until it does.
The honest answer is: useful, but imperfect. Prediction markets aggregate information from a diverse pool of participants with real financial stakes, which tends to make them faster and less ideologically rigid than individual forecasters. They update within minutes of a new data release, whereas a bank's official forecast might not be revised for weeks.
But they have well-documented limitations. Liquidity remains concentrated among a relatively small number of active traders. Those traders are susceptible to the same momentum-chasing and news-overreaction that affects financial markets generally. And the contract definitions can create perverse outcomes – an economy that enters a technical downturn in Q4 might not resolve the NBER question for another 12–18 months, since the NBER's Business Cycle Dating Committee is notoriously slow to make official calls.
The 2022–2023 period is instructive here. U.S. GDP contracted in both Q1 and Q2 of 2022, which by the technical GDP definition would have triggered a recession contract – yet the NBER never declared a recession, because employment and income held up throughout.
The most common mistake people make when reading these numbers is treating them as binary: either there's a recession or there isn't, and a 26% probability means "probably not." That framing misses the point.
A roughly one-in-four chance of a significant economic contraction within the next eight months is not a reassuring number. It means the base case is continued – if sluggish – growth, but that the range of plausible outcomes is wide and the downside scenario is genuinely on the table. The specific path from here depends heavily on how long the Strait of Hormuz remains disrupted, whether the labor market weakness seen in February was an anomaly or the start of a trend, and whether the Fed has room to respond if conditions deteriorate.
Prediction markets are tracking all of that in real time. They won't tell you what's going to happen. But they do offer a reasonably honest read on what informed participants think the odds are – and right now, those odds suggest cautious concern rather than alarm.
What is the probabability of a US recession in 2026?
The probability of a US recession in 2026 is currently estimated at 25%–35%, based on prediction markets like Polymarket and forecasts from Goldman Sachs and JPMorgan.
Related topics