The Brief

The US economy lost 92,000 jobs in February — the worst monthly decline since October and the third net loss in five months — while unemployment rose to 4.4 percent, well above forecasts. Markets sold off sharply as oil prices crossed $100 a barrel for the first time since 2022, with Goldman Sachs and Moody’s raising recession probability estimates to between 35 and 42 percent.

The Report

The Bureau of Labor Statistics reported Friday that nonfarm payrolls fell by 92,000 in February, a sharp reversal from economists’ median forecast of roughly 60,000 jobs gained. Unemployment ticked up to 4.4 percent from 4.3 percent in January, and the labour force participation rate dropped to 62 percent — its lowest level since December 2021 outside pandemic months.

The losses cut across sectors. Healthcare shed 28,000 jobs, driven partly by a four-week Kaiser Permanente nurses’ strike that sidelined more than 31,000 workers across California and Hawaii. Leisure and hospitality lost 27,000. Manufacturing dropped 12,000 against a forecast of a 3,000 gain. Construction fell 11,000, attributed in part to severe winter weather, and transportation and warehousing declined by the same figure — extending a sector slide of 157,000 jobs since February 2025. Federal government payrolls contracted by 10,000, continuing a pattern that has eliminated 330,000 federal positions since October 2024.

Prior months were revised substantially downward. December’s initially reported gain of 48,000 became a net loss of 17,000 — a swing of 65,000 — while January was trimmed by 4,000 to 126,000. The combined revision erased 69,000 previously counted jobs.

The report arrived against a volatile backdrop. US and Israeli operations in Iran, now in their second week, have disrupted shipping through the Strait of Hormuz. Oil surged from roughly $70 a barrel before the conflict to above $100 — the highest since 2022 — pushing the national average for petrol up 19 percent in a month to $3.45 a gallon. Goldman Sachs warned that sustained prices at current levels could raise US inflation from 2.4 to 3 percent by year-end.

Markets responded immediately. The Dow fell 903 points on Friday, the S&P 500 and Nasdaq both dropped more than 1.5 percent, and the VIX climbed 24 percent to its highest reading since April. The Atlanta Fed’s GDPNow tracker for Q1 2026 fell from 3.2 to 2.1 percent in the four days following the report.

Morgan Stanley’s Ellen Zentner described the Federal Reserve as caught “between a rock and a hard place” — weakening employment supporting a rate cut, rising energy costs arguing against one. Markets priced a 95.5 percent probability of rates holding at the March 17–18 FOMC meeting. Governor Stephen Miran, who has dissented at every meeting since September, continued to advocate for cuts approximately one full percentage point below the current 3.5–3.75 percent range.

The White House characterised the data differently. Deputy press secretary Kush Desai called it a “blockbuster, expectation-shattering jobs report,” while economic advisor Kevin Hassett stated “the economy is really strong.” The six-month average of job creation is effectively zero.

The Angle

The debate over February’s number — whether it reflects genuine deterioration or a temporary collision of strikes, weather, and geopolitical noise — is not uninteresting. But it is happening at the wrong resolution. A single month’s payroll figure is a seismograph reading. The question is what the underlying geology looks like, and on that the data is less ambiguous than the month-to-month argument allows.

The six-month rolling average of job creation is zero. Net employment since the tariff regime took effect in April 2025 is negative 19,000. Long-term unemployment has climbed from 1.5 to 1.9 million in twelve months. Hiring plans in February fell to the lowest level since Challenger began tracking them in 2009. None of these figures are distorted by a nurses’ strike in California.

What makes this particular moment structurally distinct is the convergence. An energy shock arriving simultaneously with a labour market already running on fumes is a different problem from either condition in isolation. The Fed’s dilemma is not a policy puzzle — it is the predictable output of an economy that spent the last eighteen months accumulating vulnerabilities while the headline numbers were still close enough to positive to allow everyone to look away. The January report that briefly encouraged optimism has already been revised into irrelevance.

The White House response is worth noting not for its content — administrations defend their economies; this is unremarkable — but for what it reveals about the distance between the narrative and the instrument panel. “The economy is really strong” is a statement made while the Atlanta Fed’s own GDP tracker drops a full percentage point in four days. That gap between declared reality and measured output has a tendency to close. The question is which direction.

The tools available to correct this — rate cuts, fiscal stimulus, trade policy reversal — each carry costs that interact badly with an oil price above $100 and inflation already above target. The Fed can support employment or contain prices. It cannot, at this moment, do both. That is not a difficult situation. It is a constrained one. The difference matters, because difficult situations have solutions and constrained situations have trade-offs, and the political system’s preference is overwhelmingly to pretend the first when facing the second.