September payrolls rose 29,000 while the household survey showed 406,000 more people working. The house view: the October hike is dead, and the December one is still on.
The Labor Department's two surveys of the same month told opposite stories on Friday. The establishment survey counted 29,000 new payroll jobs in September. The household survey, which asks people whether they are working, counted 406,000 more employed. Traders and headline writers focused on the first number.
The household figure deserves more weight. The payroll miss moves the next Fed hike from October to December, but leaves the case for it intact. The report shows none of the conditions that would undermine that case: rising unemployment, falling participation or slowing output.
The case for patience rests on two points. The Bureau of Labor Statistics release shows July revised from a gain of 21,000 to a loss of 10,000 and August from 162,000 to 133,000, taking 60,000 jobs out of the prior two months. The 12-month average gain is 45,000, and the three-month average is 51,000. A labour market adding jobs at that pace is not the one the Fed assumed when it raised rates in September.
The consensus had been higher, too. CNBC reported that the Dow Jones estimate was 84,000 and that Reuters polled economists at 90,000, so the print missed by roughly 55,000 to 60,000. Government shed 17,000 jobs and information 10,000; health care added 17,000 and construction 11,000.
The household survey is noisier month to month, which is the usual reason to discount it. Even so, this month it showed a clear gain. The labour force grew by 485,000 and the participation rate rose 0.2 point to 61.8%. The unemployment rate was 4.2%, inside the 4.1% to 4.3% range it has held since March, and the broad underemployment measure, U-6, edged down to 7.6%, its lowest since January 2025.
Together, those figures describe an economy where firms hire slowly and fire rarely, while more people look for and find work. That describes constrained supply as much as weak demand. A year of gains averaging 45,000 has left the unemployment rate inside its range, which says the pace of hiring needed to hold it steady is now small. Against that yardstick, 29,000 is a soft month, not a break. A recession shows up as firings, and CNBC reports that weekly jobless claims are low and one measure of layoffs is at a four-year low.
The rest of the data agrees with the household survey. CNBC said Commerce revised first-quarter GDP growth to 2.5% and the second quarter to 2.2%, and the Atlanta Fed's tracker has the third quarter at 3.7%. The Fed's preferred inflation gauge, on a core basis, runs at 3% annually. The 10-year Treasury yield fell nearly 6 basis points to 5.18% on Friday, after Reuters reported that 10-year and 30-year yields had touched their highest levels since 2002 on Thursday. A 10-year yield near 5.2% is tightening financial conditions on the Fed's behalf, which is part of why stocks liked the number. CrossCheck's Todd Schoenberger told Reuters the print was "good news for stocks" because the bond market was "organically" doing the central bank's job.
Pay growth gives the Fed a reason to wait. Average hourly earnings rose 5 cents, or 0.1%, to $37.81, a 3.0% annual gain and the slowest since May 2021. Wall Street had expected 0.3% for the month and 3.1% for the year. Against core inflation of 3%, real wages are about flat; against August's headline PCE inflation of 3.4%, they are falling. Navy Federal's Heather Long told CNBC that wage growth is "wiped out entirely by inflation." A Fed worried about a wage-price spiral can relax a little.
The committee held at 3.50% to 3.75% in July on a 9-3 vote, with three dissenters wanting a hike, and then raised a quarter point in September to a range of 3.75% to 4.00%, its first increase since 2023 under Chair Warsh. Its own projections put the policy rate at 4.00% to 4.25% by the end of 2026, which is one more hike, according to Reuters. A central bank that moved against 3% core inflation with unemployment at 4.2% does not turn on one soft payroll print. Slower pay growth gives it a reason to wait a meeting.
Jefferies' Thomas Simons called the report "the nail in the coffin for an October hike," and he is right about October. The broader argument fits the data less well: that August was a rebound and the revisions reveal weak hiring as the underlying trend. Low hiring and low firing, falling slack in the household survey and unemployment near 4.2% suggest a shortage of workers at least as much as weak demand. Inflation alone can justify a hike. Inflation is still at 3%.
Pricing followed the logic. CME FedWatch put the odds of a hold at the October 27-28 meeting at 82.8%, and traders gave an October hike roughly one chance in six, down from about one in four before the data. The next hike is now expected in December. Equities treated the number as a gift: Reuters reported the Nasdaq Composite at a record in morning trading, up 1.7%, with Nvidia also at a record and Brent below $100.
Anyone adding bonds at 5.2% on the argument that the Fed is finished is betting that the household survey is wrong and the payroll survey is right, and that the GDP tracker is wrong too.
The Fed should hold on October 28 and say so without drama. The number that could change the December call arrives with the next jobs report on November 6: unemployment at 4.4% would break the range and make a hike harder to defend. At 4.3% or below, a quarter-point increase in December stays the base case.