My initial reaction is that the drop in February looks a lot like noise, and I want to see another month or two of data. But if there really is a sustained drop in multiple job holding, that suggests some weakness in labor demand.
No, it's not your imagination: Your morning cup of coffee has gotten a lot more expensive. Blame tariffs (and also climate change).
More from Sydney Ember last fall: www.nytimes.com/2025/09/11/b...
Core services excluding housing, which the Fed has at times highlighted as a measure of underlying inflation, has stalled out recently (and picked up sharply on a three-month basis).
This report, more than most, feels like a real look back in time, so I'm not going to do a long thread. But a couple quick charts, starting with the headline: Overall inflation was steady in February, although it has picked back up a bit on a three-month basis.
Remember that the year-over-year inflation rate is still being depressed by the lingering effects of the government shutdown last fall, which disrupted data collection and led the BLS to essentially assume rents didn't rise at all in October.
Data: www.bls.gov/news.release...
Live coverage: www.nytimes.com/live/2026/03...
U.S. consumer prices rose 0.3% in February and were up 2.4% from a year earlier. "Core" prices, excluding food and energy, were up 0.2% month/month and 2.5% year/year. But remember that all these numbers are from before the war in Iran sent gas prices soaring. #NumbersDay
I don’t think I said (and certainly didn’t intend to say) there’s been an increase in layoffs. I was trying to embody the audience member who was going to”what do you mean there’s no impact, I keep hearing about all these companies cutting jobs because of AI!”
This is tomorrow! Join us for what is shaping up to be a great discussion of AI and the labor market. #EconSky
Interestingly, the number of people working part time because they couldn't find full-time work fell in February, so the broader (U6) measure of un- and underemployment fell, even as the official (U3) unemployment rate rose.
The unemployment rate ticked back up to 4.4 percent in February. On its own, that doesn't mean much, but unemployment has been drifting very gradually higher, though it's still quite low by historical standards.
Government job cuts are part of the story, but nowhere close to all of it. Even looking just at the private sector, job growth has slowed to a crawl.
Between the decline in employment in February and the downward revision to December, job growth has now effectively slowed to zero. #numbersday
U.S. payroll employment FELL by 92,000 jobs in February and the unemployment rate ticked up to 4.4 percent.
Data: www.bls.gov/news.release...
Live coverage: www.nytimes.com/live/2026/03... #NumberDay
More in my story today. Oh, and happy #jobsday!
www.nytimes.com/2026/03/06/b...
But we shouldn't take this system for granted. There are very real strains on it, and a very real need to modernize, including by relying less heavily on surveys. That will require resources. And if that doesn't happen, the quality of the data will erode.
If you were following the monthly numbers last year, you understood that job growth was slowing (and approaching zero late in the year), but that unemployment remained low, a "low-hire, low-fire" stasis. That basic narrative has not changed.
But ultimately I do still think these numbers are useful, as long as you're using them properly, which means viewing them in context. That means watching the trend not just the individual datapoint. It meant looking at the details of the reports, and also at other sources.
The second question is harder. We've had some big revisions recently, including another big downward benchmark revision to payrolls. I think it's reasonable for people to question the value of a report that can't reliably distinguish a healthy gain from an outright loss.
Or take the word of the remaining employees there, who I'm in touch with and who I am confident would blow the whistle if something were amiss. Or just dig into the numbers, which have not exactly posted a consistently rosy picture for the administration.
First, the easy question: No, there is no evidence of political interference in the BLS's numbers. You don't have to take my word for it. Take the word of @erikamcentarfer.bsky.social, the person Trump fired after he didn't like the BLS's numbers.
I get a lot of questions these days about data from BLS and other agencies. Some of this is about fears of political interference. Some of it is about the reliability of the data, especially in light of recent big revisions. In a story today, I tried to tackle both.
www.nytimes.com/2026/03/06/b...
What do we know about AI’s impact on the labor market? On March 10, Peter Orszag, @jedkolko.bsky.social, @marthagimbel.bsky.social, Bharat Chandar, and @bencasselman.bsky.social will dive into the data and what we know:
Featuring @jedkolko.bsky.social @marthagimbel.bsky.social @bharatchandar.bsky.social and Nathan Goldschlag. At the end of the discussion, the audience will get to pick a winner, and everyone else will immediately be replaced by Claude. (I'm kidding. In reality, all of us will be replaced by Claude.)
Really looking forward to this event, in which we'll try to tackle one of the most important questions in the economy right now: What does A.I. mean for the labor market, both now and in the near future?
Join us next week!
www.hamiltonproject.org/event/unders... #EconSky
Signs BlueSky is coming into its own: I have my first impersonator on here. No, @casselmanbe44.bsky.social is not me. This, here, is me. (Thank you to an eagle-eyed observer, whom I won't drag into this, for flagging!)
The folks at @gothamist.com capture it perfectly: "It's Monday in New York City, where we're doing this again."
Stakes in private companies. Handshake deals with CEOs. President Trump's economic policies are a long way from the principles that once defined the Republican Party.
I grappled with how we should understand his policies -- and what will come after him.
🎁🔗
www.nytimes.com/2026/02/22/b... #EconSky
I looked at the sub-sector breakdown for another reader, and most of the recent growth has been in hospitals and doctors’s offices. I haven’t looked into the family members piece, but the payroll numbers only include W-2 employees.
Grocery prices were up just 2.1 percent in January from a year earlier, though they're still running hotter than in 2024. But if people don't seem to be celebrating, that may be because grocery prices are up 30 percent since 2020.
