On 7 August the Bureau of Labor Statistics reported that American employers shed 23,000 jobs in July, against a consensus forecast of roughly 80,000 added. The same release showed the unemployment rate falling, to 4.1 per cent. Both are true at once, because the rate counts only people who are looking for work. Participation stood at 61.4 per cent, down 0.7 points since January, and revisions took a further 103,000 jobs out of May and June.
Three weeks later, on 28 August, the chair of the Federal Reserve told the Jackson Hole symposium: “I believe the labor markets are consistent with full employment.” On the arithmetic he has a case. Unemployment claims, on a four-week average, are near their lowest level in decades, and the jobless rate has barely moved in two years. The question is what a stable rate is measuring when the denominator underneath it is shrinking.
A rate that falls because people stopped looking is not the same statistic as a rate that falls because people found work.
That is not a debating point, and it has numbers attached. In July, 1.8 million Americans had been out of work for twenty-seven weeks or more — a quarter of everyone counted as unemployed. Another 4.8 million were working part time because they could not find full-time hours. All of them sit inside the 4.1 per cent, and none of them are evidence of a market that is clearing.
Some of the flat labour supply is a policy choice rather than weather. Work published in January by Wendy Edelberg, Stan Veuger and Tara Watson put net migration to the United States in 2025 at between minus 295,000 and minus 10,000, the first negative year in at least half a century. On their estimates, the monthly payroll growth needed merely to hold the employment rate steady falls to somewhere between minus 20,000 and plus 50,000. A print near zero can mean stability or stagnation, and the headline cannot tell you which.
The honest concession is that none of this makes 4.1 per cent a lie. Layoffs are not running high. People already in work are largely staying in it, and the claims data the chair cited is as strong as he says. The failure is at the entrance rather than the exit — a hiring problem, not a firing problem — and that is a different disease requiring different medicine. Nothing in the record supports the claim that the series is fabricated.
Britain has begun administering some. On 26 August the Department for Work and Pensions announced the first placements under its Jobs Guarantee: six months of paid work, up to twenty-five hours a week, for eighteen- to twenty-four-year-olds who have claimed Universal Credit for eighteen months, backed by £2.5 billion and incentives worth up to £8,000 a business. Pat McFadden, the work and pensions secretary, made the case in a sentence: “Young people want to get on in life but for too long too many have been denied that opportunity.”
Whether it works is an empirical question and the answer will take years to arrive. But it is aimed at the part that is actually broken, and it is an instrument no central bank possesses. The Federal Reserve can change the price of money. It cannot place a twenty-two-year-old in a first job. Mistaking the one for the other is how a headline rate stays flat while a cohort falls quietly out of the count.
