When a Quiet Labor Market Stops Working: Signs of a Deeper Freeze

The employment report released on August 7 brought the risk I described in this space into much sharper focus. Payrolls fell by 23,000 in July. In absolute terms, that is a modest decline, and the Bureau of Labor Statistics rightly characterized it as little changed. But the revisions that came with the report matter more than the headline. May was marked down from 129,000 to 63,000, and June from 57,000 to 20,000. Together, those revisions removed 103,000 jobs from what we thought we knew about late spring and early summer. Over the past 12 months, payroll growth has averaged just 34,000 per month.

The unemployment rate stood at 4.1 percent, essentially unchanged. On its own, a rate that low would ordinarily be reassuring. The difficulty is that it is being sustained by a labor market where very little is happening. In fact, the average number of people leaving the labor force over the past two years has reached 3.3 million per month, with about half leaving due to job-search burnout or retirement.

Why Labor Market Movement Matters

When I wrote about the June figures, my concern was not that the labor market was about to collapse. It was that it might stop moving. The unemployment rate had remained within a narrow range for roughly eighteen months, and that surface calm concealed something underneath. A healthy labor market depends on churn. Workers change jobs, move between employers, and find positions that make better use of what they can do. That movement is not simply a byproduct of a functioning market. It is part of what makes the market function. It is how pay rises, how firms improve their workforces, and how people end up where they are most productive.

The worry was that slowing hiring could harden into something closer to a freeze, and that a freeze, if left in place long enough, would eventually result in outright job losses rather than merely slower gains. The July report is the first month in which that progression appears to be taking shape.

What the Flow of Unemployment Reveals

The most telling numbers in this report are not the payroll figure or the unemployment rate. They are the figures showing how people move in and out of unemployment.

The number of people unemployed for less than five weeks fell to 2.0 million in July and is down 344,000 over the year. Fewer people are becoming newly unemployed, which sounds encouraging and, in isolation, is. But the long-term unemployed, those out of work for 27 weeks or longer, still numbered 1.8 million and accounted for 25.5 percent of all unemployed people. That share has barely moved over the past year and contrasts with the much lower average share of long-term unemployed of 20.8 percent during the post-pandemic period up until last year.

Read those two facts together, and the picture is of a market where the door has largely closed in both directions. Relatively few people are being pushed into unemployment, and relatively few of those already there are finding their way out. That combination is precisely what a sclerotic labor market looks like in the data. The aggregate rate stays low not because opportunity is abundant, but because so little is changing hands.

Temporary layoffs point in the same direction. Their number rose by 153,000 in July, to 921,000. Firms appear to be adjusting headcount at the margin while holding back on hiring that would allow displaced workers to move quickly into something new.

Slower Wage Growth Fits the Broader Pattern

Average hourly earnings for private nonfarm employees stood at $37.62 in July, up two cents on the month and 3.2 percent over the year. That is not the pay pattern of a market in which workers retain meaningful leverage. It is the pattern in which leverage has largely disappeared, which connects directly to the argument I made in the second issue of my Substack, Kugler Labor Notes & Beyond, about real earnings.

Weakness is also becoming concentrated in identifiable places rather than spreading evenly. Employment in financial activities has now fallen by 121,000 since its May 2025 peak. Local government education shed 50,000 positions in July, and retail trade lost 19,000. Health care continued to add jobs, up 22,000, though at a slower pace than its twelve-month average of 36,000. When hiring narrows to a small number of sectors while job losses broaden across more sectors, the reallocation that a dynamic market depends on becomes harder for everyone outside those sectors.

What the Recent Gains Mean, and What Comes Next

None of this erases what workers gained during the tight labor market of 2021 through 2023. Those real wage increases were real. The fact that the subsequent cooling did not immediately reverse them or trigger a wage-price spiral was a genuinely favorable outcome, one that deserves more acknowledgment than it usually receives.

But favorable outcomes are not permanent. A market that stops moving stops serving the people inside it, and a market that has begun shedding jobs has moved beyond merely stiffening. My concern remains what it was in June, only now there is more evidence to back it up.

Two dates are worth marking. On August 28, the Bureau of Labor Statistics publishes its preliminary estimate of the annual benchmark revision to the establishment survey, which will tell us whether the payroll picture we have been working from was accurate. The August employment report follows on September 4. If it confirms the pattern set in July rather than reversing it, then the freeze I described earlier was not where this ends. It was the warning.