The economy shed 23,000 jobs in July and the unemployment rate went down anyway, which is the single most misleading pair of numbers the government will publish this year.
It fell because only 61.4% of American adults are still in the labor force, the lowest share since February 2021, and you cannot be counted as unemployed if you have given up.
That is the part of Friday’s report nobody says out loud on cable. The headline number improved. The thing the headline number is supposed to measure, whether an ordinary person can find work, got worse.
What Actually Happened in July
Payrolls fell by 23,000 in the Bureau of Labor Statistics release, a sudden reversal after four months of gains. Economists surveyed by Dow Jones had penciled in 83,000 additions. Missing a forecast by 106,000 jobs is not a rounding error, and it was not the worst part of the release.
The worst part was the revisions. BLS cut May’s gain by 66,000 to 63,000 and June’s by 37,000 to 20,000, wiping 103,000 jobs off the two prior months. June was reported five weeks ago as a modest but real 57,000, and we wrote at the time that the economy was running on fumes. It turns out June was a third of what we were told. The spring recovery that pundits spent the summer describing did not exist. It was a data artifact, corrected quietly on a Friday morning in August.
Average hourly earnings rose two cents to $37.62, up 3.2% over the year. That is a raise that roughly keeps pace with prices and buys nothing extra.
Who Lost the Jobs
Local government education shed 50,000 positions. That is not an abstraction. That is teachers’ aides, bus drivers, cafeteria staff, special-education paraprofessionals, and school secretaries, cut in the weeks before a school year starts, in districts that were already asking the people who stayed to work second jobs to cover rent. Retail lost another 19,000 heading into back-to-school season, which is the retail calendar’s second-biggest hiring window.
Health care kept adding. It has been carrying the entire labor market for the better part of two years, which is less reassuring than it sounds. An economy where the only reliable job creation is nursing homes and hospital systems is an economy that has stopped generating opportunity everywhere else.
The Number That Explains the Number
Here is the mechanism the coverage keeps skipping. The unemployment rate is a ratio, and its denominator is people who are actively looking. When someone quits searching, the numerator drops, the denominator drops, and the rate can fall in a month when the country destroyed jobs. That is exactly what happened.
Underneath it sits a structural change almost nobody outside the Fed research departments is talking about. Economists call it the breakeven rate: how many jobs the economy has to create each month just to absorb new workers and hold unemployment flat. For most of the last decade it was somewhere north of 100,000. Federal Reserve Board staff estimated in April that labor force growth is now near zero, driven by an aging population and collapsed net immigration, which means breakeven employment growth is near zero too. The Federal Reserve Bank of Kansas City reached the same conclusion, and Dallas Fed researchers tied the shift directly to net outflows of unauthorized immigrants.
Read that consequence carefully, because it is the whole ballgame. If breakeven is near zero, negative months are roughly as likely as positive ones, and the unemployment rate can sit politely at 4.1% while hiring is effectively dead. The rate stops being a thermometer. It becomes a number that reflects how few people are left in the pool, not how many of them are working.
Deporting workers does not show up as a spike in unemployment. It shows up as a labor market that looks calm on paper while ordinary people describe an economy where nothing is hiring.
The Pain the Rate Cannot See
Ask anyone who has been on the market since spring. The complaint is never that they got laid off. It is that they applied to 200 jobs and heard back from four.
The data agrees. The hires rate, meaning hires as a share of total employment, has fallen for three straight years and now sits at levels last seen in 2013 and 2014, deep in the slog out of the Great Recession. The Economic Policy Institute has argued that this depressed hiring rate, not layoffs, is the main thing crushing young college graduates. The Federal Reserve Bank of New York’s tracker for recent graduates puts their unemployment rate near 5.6% and their underemployment rate at 42%. Nearly half the class of 2026 with a degree is working a job that never required one.
Long-term unemployment tells the same story from the other end. Some 1.8 million people have now been out of work 27 weeks or more, a quarter of everyone counted as unemployed. That share does not budge when hiring freezes, because a freeze is not dramatic. Nobody gets fired. You simply cannot get back in.
And the burden is not distributed evenly, as it never is. Black workers faced a 6.6% unemployment rate in June against 3.6% for white workers, and unemployment among young Black workers jumped from 14.1% to 16.6% in a single month. When the door closes, it closes on some people first.
What Happens Next
Markets did the only arithmetic they know. Stocks slipped, and traders moved from pricing no chance of a half-point cut at the September 16-17 FOMC meeting to pricing a 17% chance of one, with at least two cuts now expected by year end.
Cheaper money is not a bad thing. It is also not a job. Rate cuts work by making it attractive for firms to expand, and they take three to four quarters to reach anyone’s paycheck. For a school district that just cut 50,000 aides, a September cut changes nothing about this school year.
The real risk is what officials do with a 4.1% headline. It is a genuinely low number, and it will be quoted, framed, and posted as evidence that the labor market is fine. Every month that story holds is a month nobody feels obligated to act, while participation drains, hiring stays frozen, and the people who quit looking vanish from the statistic that decides whether we have a problem.
The rate is not measuring the economy anymore. It is measuring who is left.