The Job Market Isn't Bad. It's Frozen. Here's the Difference.

There is a number everybody quotes about the job market and a number almost nobody does, and this month they are telling opposite stories.

The Bureau of Labor Statistics published its July JOLTS report on September 1. It got very little coverage, which is a shame, because it contains the clearest picture of what job seekers are actually running into right now.

Here is the whole thing in two lines.

Job openings in July: 7.3 million. The monthly average in 2019, before any of this, was 7.2 million. Openings are normal.

The quits rate in July: 1.9%. The monthly average in 2019 was 2.3%. Quitting is not normal.

That gap is the story. The demand side of the market looks roughly like it did in a good year. The worker side has seized up. Those two facts produce a job search that feels nothing like 2019 even though the headline number says it should.

What the quits rate actually measures

The quits rate is the share of employed people who voluntarily left a job in a given month. It is the closest thing the federal data has to a confidence reading. People quit when they believe something better is available and they will not be punished for reaching for it.

It peaked at 3.0% in late 2021, which is what the Great Resignation looked like in a statistic. It has now sat between 1.9% and 2.0% every single month of 2026.

I want to be careful here, because this number gets oversold. You will see "the lowest quits rate in a decade" in a lot of coverage. That is not right. The rate touched 1.9% twice during 2024 as well, and it ran far lower than this in the depths of the 2009 recession. The defensible statement is narrower and still striking: quitting is running below its pre-pandemic norm and about a third off its 2021 peak, and it has not moved all year.

Why a frozen market is different from a bad one

A bad market has too few openings. You can see it, everyone agrees about it, and the advice writes itself.

A frozen market has a normal number of openings that are not turning over. That is harder to perceive and it changes the mechanics of a job search in ways nobody tells you.

Start with where openings come from. A posting exists because a company created a new role, or because somebody left one. When quitting runs at 2.3%, a large share of what you see is backfill. When it runs at 1.9%, that share shrinks. The total may look the same while its composition quietly changes underneath you.

That matters, because a backfill and a new role behave differently. A backfill has a defined shape, a predecessor, a team that already knows what the job is, and usually genuine urgency, because work is piling up on somebody. A newly created role often has none of that. It has a budget line, an aspiration, and a hiring manager still arguing internally about what it should be. Those are the roles most likely to stall, get re-scoped mid-process, or quietly disappear.

So the frozen market does not simply hand you fewer chances. It shifts the mix toward the kind of opening that wastes the most of your time.

The July detail that got no coverage at all

One line in the release deserves more attention than it received.

Hires in professional and business services fell by 188,000 in July. Not openings. Hires. That is consulting, accounting, legal services, marketing, staffing, corporate support functions, a large share of what people mean when they say white-collar work.

Meanwhile openings rose by 76,000 in durable goods manufacturing.

Read those two together and you get a sharper picture than any headline provides. The part of the economy that hires people who read blog posts like this one is hiring less, while the part that does not is hiring more. The aggregate can stay flat while your specific market gets worse, and that is exactly the sort of thing an average is designed to hide.

Why nobody is quitting

The obvious answer is fear, and there is survey evidence for it. But there is a cleaner explanation sitting in payroll data.

ADP publishes pay growth split by whether a person stayed in their job or changed jobs. In August, gross pay for people who stayed rose 4.4% year over year. For people who changed jobs it rose 7.3%.

Switching still pays. It pays about 2.9 percentage points more than staying.

Note what that does and does not say. You may have seen a claim going around that the switching premium collapsed from around 20% to around 7%. That is wrong, and it is wrong in an instructive way: it compares the job-changer growth rate to itself and calls the result a premium. The premium is the gap. The gap is 2.9 points, positive, and roughly in line with recent years.

Which makes the freeze harder to explain as pure economics. Moving still pays more than staying. People are staying anyway. That is a confidence problem, not a compensation problem, and confidence problems take longer to thaw than pay problems do.

What to actually do with this

Four things follow, and none of them are "apply to more jobs."

Weight your screening toward evidence that a role is real. In a market with proportionally fewer backfills, the base rate of soft, speculative and stalled postings goes up. That is not a reason to despair. It is a reason to spend your first fifteen minutes on the listing rather than on your resume. A posting that names a team, a predecessor, a manager, or a specific project is describing a job that exists. One built entirely from adjectives may not be.

Treat listing age as a stronger signal than usual. In a market that turns over slowly, an old posting is less likely to be a busy recruiter and more likely to be a role nobody is urgently trying to fill. Listing age was already one of the best available predictors of whether you hear back. In a frozen market it gets better, not worse.

Read your own sector, not the headline. The professional and business services number is the whole argument for this. National aggregates are averages across industries that are not moving together. Find the line item that covers your work and read that instead.

If you have a job and you are hesitating, price the hesitation. Staying is a choice with a number attached, and this year the number is about 2.9 points of pay growth. That may well be worth it to you. Plenty of good reasons exist to sit still. Just make it a decision rather than a default, because "everyone is staying put" is a description of a market, not advice about your life.

The honest summary

The job market in the back half of 2026 is not collapsing. Openings are at a normal level and the August payroll print came in well above expectations.

It is also not fine, and telling people it is fine because openings look normal misses the mechanism. A market with normal demand and frozen mobility produces long searches, stale postings, and a lot of applications into roles that were never quite real. That experience is not a personal failing and it is not in the headline number.

It is in the one nobody quotes.

---

Try JobIntel free at jobintel.com. See every job. Know if you fit.