The September jobs report came out on Friday morning at 8:30, and the number was 29,000.

That went through the markets like a cold front. Economists had forecast 84,000. The previous two months had already been revised down by a combined 60,000, so even the summer, which had looked reasonable, was worse than reported.

Here is the other number from that same week: the U.S. economy grew at 2.2% in the second quarter. The revision came out Wednesday, higher than the preliminary estimate of 1.5%. The economy is growing. Hiring has nearly stopped. Both of those things are true, and I spent most of this week trying to understand what to do with the combination.

The people who follow this for a living reached for the phrase “low hire, low fire.” It’s a good phrase. It means companies aren’t adding workers and they aren’t cutting them either. The economy is, in some technical sense, stable. Unemployment is 4.2%. That’s elevated from where it was, but it’s not a crisis number. There are no mass layoffs. There are just very few people being hired, and the people who need a job are discovering that a market that supposedly has 4.2% unemployment doesn’t have a spot for them.

I grew up watching my father work a line at an auto parts plant in Anderson, Indiana. He started at Delco-Remy in 1961 and was on that floor for twenty-seven years. Every morning, steel-toed boots, the same route to the plant, the same men at the gate. The economy “grew” in most of those years, in the sense that the GDP number was positive. My father cared nothing about GDP. He cared whether there was work and whether the work paid enough and whether it would still be there next year.

That’s not a sentimental argument. It’s a measurement argument. GDP measures what an economy produces. It says nothing, on its own, about where the production lands. When GDP grows at 2.2% and the labor market adds 29,000 jobs, something in that gap is worth reading carefully.

Here is what the September report said about health care: 17,000 jobs added. The twelve-month average is 33,000 a month. Health care is the floor of the American labor market. It doesn’t have bad months. The fact that it had one in September is not a catastrophe, but it’s the kind of signal you write down and don’t forget.

Financial activities lost 7,000 jobs and has shed 129,000 positions since May of last year.

Wages are up 3% year over year. Inflation is running at 3.4%. The people working in this economy are, by one reasonable measure, getting poorer at a rate of roughly four-tenths of a percent a year. Not dramatically. Not a crisis. Just quietly, steadily, not keeping up.

The Federal Reserve meets October 28 and will have to decide what to do with all of this. One month’s jobs report doesn’t make a trend, and the Fed is not in the habit of overreacting to a single data point. But the three-month average for job gains is sitting around 51,000 a month. That’s stagnation, not growth. The question they’ll be deciding on October 28 is whether cutting rates helps a labor market that’s frozen rather than declining, and the honest answer is that nobody is entirely sure what that does.

The phrase “frozen labor market” is doing a lot of work right now, and I’m not sure it’s doing it honestly. A frozen market means the people inside it are fine. It says nothing about the people trying to get in.

There’s a version of the October 2 headline that writes itself: “Jobs Miss Badly, But Economy Still Growing.” That’s accurate. The economy is still growing. The 29,000 number is one month’s reading in a volatile series. The revision to summer payrolls might just be the statistical correction that always comes.

But I’ve been reading these reports since the Reagan years, and the specific combination in Friday’s number, GDP up, wages below inflation, hiring nearly stopped, health care below trend, is the kind of reading that tells you the economy is working for some people and not others, and that the people it isn’t working for are not well-represented in the headline figure. They’re in the denominator. They’re not in the growth.

The Fed will meet October 28. The economists will revise their models. The headlines will keep saying what they say. The labor market will remain what it is: not bad, not good, and for a specific group of people, something close to invisible.