When are 335,000 new jobs not 335,000 new jobs? When Washington makes them up and the media dutifully report them with all due breathlessness.
But first, let’s look at how the media reported January’s job numbers.
January Jobs Report Was a Blowout. Disregard the Seasonal Noise. —Barrons
The U.S. created an extraordinary number of jobs in January. Here’s a deeper look. —NPR
U.S. employment soars by 353,000, stunning Wall Street. —MarketWatch
If you go to the more usual suspects, you’ll see “Another shockingly good jobs report shows America’s economy is booming” from CNN, “Blockbuster Jobs Report Backs Up Fed’s Patience as It Waits to Cut Rates,” from the NYT, and this gag-worthy bit from NBC News: “The great American jobs machine keeps revving in an election year.”
[…] Yeah, about those jobs…
For the dismal numbers behind the screaming headlines, let’s go to economics Ph.D. (and portfolio manager) Robert Barone at Forbes.
[…] Barone wrote, “The workweek, itself, contracted to 34.1 hours from 34.3 in December and 34.4 in November. This is the lowest number since March 2020 (the pandemic) and, before that, November 2008 (Great Recession).”
“Rosenberg Research calculated that, despite the job gains, total hours worked actually contracted.” [Emphasis added.] In fact, the Household Survey — “the one the media ignores,” as Barone put it — “showed up with a -31K headline jobs number, and a fall of -63K in full-time jobs.” Worse, the labor participation rate remains down. Americans are working more part-time jobs because there are fewer full-time jobs to go around — a sizable fraction of last month’s jobs went to people taking on a second or third job.
So how did the BLS come up with 335,000 new jobs when the Household Survey showed a net loss of 31,000? For that, let’s go to Seeking Alpha:
While the BLS reported 353k jobs, the Household Survey actually reported a loss of -31k. December was an even bigger miss with 333k vs -683k.
The BLS also publishes the data behind their Birth/Death assumptions. January is always a big drop, but this was smaller than last year (-144k in Jan 2023 vs. -121k in Jan 2024). Birth/death specifically refers to assumptions made about new business being formed.
“Birth/Death assumptions” means just that. The BLS assumes that so many jobs were created or destroyed and (kind of) fixes the imaginary numbers when the real data come in. Now you know why jobs always seem to get revised downward. Seeking Alpha adds that “many believe the Household survey is a more accurate measure of employment when compared to the headline number.”
No kidding.
