Construction was one of the few sectors still adding jobs in July, with 22,000 hires.
Construction was one of the few sectors still adding jobs in July, with 22,000 hires.
American employers shed jobs last month for the first time since the spring, and the unemployment rate went down anyway. July payrolls fell by 23,000, according to the Bureau of Labor Statistics, against a consensus that had pencilled in a gain of about 80,000. The jobless rate, forecast to rise to 4.3%, instead slipped to 4.1%.
Those two numbers only look contradictory until you see who stopped counting. More than 260,000 people left the labor force in July, and a worker who is no longer looking for a job is no longer counted as unemployed. The rate improved because the pool shrank, not because the hiring did.
Why the Jobless Rate Fell While Jobs Disappeared
The two headline figures in the monthly report come from two different surveys, which is why they can move in opposite directions. Payrolls come from a survey of employers; the unemployment rate comes from a survey of households, and it counts only people who are actively looking for work.
Heather Long, chief economist at Navy Federal Credit Union, put the distinction bluntly on the day of the release: “Unemployment rate: 4.1% ->This went down for the WRONG reasons. Over 260,000 left the labor force.”
The participation data backs her up. The overall labor force participation rate sat at 61.4% in July and has now fallen by 0.7 percentage points since January. Roughly 1.8 million people have been out of work for 27 weeks or longer, a group that accounts for 25.5% of all unemployed workers. Another 4.8 million are working part time for economic reasons, meaning they want full-time hours and cannot get them.
There was one genuinely encouraging detail underneath. Prime-age participation, which strips out students and retirees, edged up to 83.4% from 83.3% in June, recovering a fraction of a sharp drop from 83.9% the month before.
The Revisions Did More Damage Than the Headline
The 23,000 decline was not the most consequential number in the release. The revisions were.
May’s gain was cut from 129,000 to 63,000, a downward revision of 66,000. June’s was cut from 57,000 to 20,000, a revision of 37,000. Together, 103,000 jobs that had been reported as real in earlier months were removed from the record. The spring recovery that policymakers and markets had been working from was, in retrospect, substantially weaker than it appeared at the time.
The split between private and public payrolls tells the same story twice. Private employers added 30,000 jobs in July against an estimate of 78,000, and June’s private figure was itself revised down from 49,000 to 30,000. June’s government number flipped outright, from a reported gain of 8,000 to a loss of 10,000.
Lindsay Rosner of Goldman Sachs noted the seasonal echo: “History doesn’t repeat, but sometimes it rhymes. For the third time in as many years, July jobs data saw a mid-summer loss of momentum.”
Where the Jobs Went, Sector by Sector
The losses inside July payrolls were concentrated rather than broad. Government payrolls fell by 53,000, the single largest decline, driven overwhelmingly by local government and local education, where employment dropped by roughly 57,000 even as other levels of government added a little.
Retail trade lost 19,400 positions. Supercenters and general merchandise stores accounted for 21,300 of that on their own, and gas stations shed 4,600, offset partly by a 9,500 gain at sporting goods, hobby, music and book retailers. Financial activities fell by 14,000, with credit intermediaries down 8,800 and insurance carriers down 6,700; that sector now sits 121,000 jobs below its May 2025 peak.
The growth side of the ledger was steadier but slower. Private education and health services added 25,000 and construction added 22,000, the two biggest gains of the month. Professional and business services rose 18,000, information 11,000, transportation and warehousing 9,700, other services 9,000, manufacturing 5,000, wholesale trade 4,700 and utilities 700. Mining and logging fell by 2,000.
Within that education and health grouping, health care on its own accounted for 22,000 of the gain, with ambulatory health care services contributing 18,100. That is well short of the 36,000 average monthly gain health care had posted over the previous twelve months. Long the most reliable engine in the US labor market, it is now visibly downshifting.
The World Cup Hangover in Leisure and Hospitality
Leisure and hospitality shed 40,000 jobs, the second-largest drop of the month, and the explanation is unusually specific. The sector had staffed up through a home World Cup, and July is when that work ended. Stadium concessions, hotels, bars and restaurants across the host cities that had hired for the tournament wound those roles down as the crowds went home.
That makes the figure easier to read but no less real for the people affected. It also means the underlying trend in the sector will not be legible until August and September data strip the tournament effect out. Anyone drawing a conclusion about consumer spending from this single month should wait.
What It Means for the Fed in September
The unusual feature of this cycle is that a weak jobs report has landed while the Federal Reserve is debating whether to raise rates, not cut them. Inflation has stayed stubborn, energy costs are high, and the target range currently sits at 3.5% to 3.75%.
Traders repriced immediately. On the CME FedWatch tool, the probability of the Fed holding steady in September rose to 55.9% from about 45% the previous day, while the odds of a 25 basis point hike fell to 44.1% from 55%. For the end of the year, one hike remains the single most likely outcome at 44.9%, followed by two hikes at 26.8% and no change at all at 23.6%.
Ellen Zentner of Morgan Stanley framed the limits of that shift: the weak payrolls print “may ease the pressure on the Fed to raise rates at its September meeting, but next week’s inflation data will still likely be the deciding factor.”
Equity markets treated it as relief. The S&P 500 rose 0.4% in morning trading, the Nasdaq Composite gained 0.96% and the Dow Jones Industrial Average added 0.13%.
Analysts at Indeed’s Hiring Lab argued the deeper problem is the lack of margin for error, warning the report “exposes just how little cushion the job market has left should the Fed be forced to tighten into a slowdown.”
The Pay Packet Is Falling Behind
Average hourly earnings rose 3.2% year over year in July, below the 3.5% economists expected, and June’s figure was revised down from 3.5% to 3.4%. With inflation still elevated, wage growth at that pace is unlikely to be keeping pace with prices, which means the squeeze households report feeling is arithmetic rather than sentiment.
This is the part of the data that tends to outlast a single month’s noise. Slower nominal wage growth alongside a shrinking labor force points to an economy where fewer people are competing for work and the ones who stay are not winning bigger raises for it. Debates about how work itself is structured, from hybrid arrangements to the four-day week and what the evidence actually shows, take on a different character when pay is no longer outrunning the cost of living.
What to Watch Next
Three things will decide whether July was an aberration or an inflection.
The first is next week’s inflation reading, which Zentner and most of the market now treat as the real determinant of the September decision. The second is the August payroll print and, just as importantly, what it does to July’s number. After three consecutive months of significant downward revisions, the first estimate has earned some scepticism. The third is participation: if the 260,000 who left in July stay out, the unemployment rate will keep flattering an economy that is not actually absorbing workers.
Not everyone reads the trajectory as alarming. Jeffrey Roach of LPL Financial argued the “labor market is experiencing an orderly slowdown, and labor stress indicators remain historically low.” Guy Berger of Homebase called it “a fairly mediocre report overall” rather than a break.
Ger Doyle, North America regional president at ManpowerGroup, offered the most useful frame for workers reading the numbers: “What’s becoming clear is that demand isn’t disappearing; it’s being redirected. We’re entering a labor market where opportunity is increasingly concentrated around specific skills, industries, and investments.” July payrolls fell, but construction sites, hospitals and professional services offices were still hiring, just not in the places, or at the pace, the last three months had promised.
This article was AI-assisted and edited for accuracy.
