Job growth revised down by 911,000 through March, signaling economy on shakier footing than realized
Source: CNBC
Published Tue, Sep 9 2025 10:00 AM EDT Updated 4 Min Ago
The labor market created far fewer jobs than previously thought, according to a Labor Department report Tuesday that added to concerns both about the health of the economy and the state of data collection.
Annual revisions to nonfarm payrolls data for the year prior to March 2025 showed a drop of 911,000 from the initial estimates, according to a preliminary report from the Bureau of Labor Statistics. The total revision was on the high end of Wall Street expectations, which ranged from a low around 600,000 to as many as a million. The revisions were more than 50% higher than last years adjustment.
The numbers, which are adjusted from data in the quarterly census and reflect updated information on business openings and closings, add to evidence that the employment picture in the U.S. is weakening. Most of the time span for the report came before President Donald Trump took office, indicating the jobs picture was deteriorating before he began levying tariffs against U.S. trading partners.
Tuesdays revisions are not by themselves a reflection of current conditions as they go back as much as a year and a half. However, recent months data also has been pointing to a soft labor market. The summer months of June, July and August saw average payroll growth of just 29,000 per month, below the breakeven level for keeping the unemployment rate steady.
Read more: https://www.cnbc.com/2025/09/09/jobs-report-revisions-september-2025-.html
March 2024 - March 2025 revision.
From the source -
Link to tweet
@BLS_gov
·
Follow
Preliminary benchmark revision for March payroll employment is -911,000 (-0.6%) https://bls.gov/news.release/prebmk.nr0.htm
#BLSdata
10:02 AM · Sep 9, 2025
Article updated.
Original article -
The Bureau of Labor Statistics on Tuesday released annual payrolls revisions for the year prior to March 2025.
This is breaking news. Please refresh for updates.

IronLionZion
(49,903 posts)He'll fire anyone except himself apparently. Since the buck stops somewhere down in the statistics level and not at his disastrous policies.
Wiz Imp
(7,142 posts)These revision do not really paint Trump's administration in a bad light - at least on the surface and not yet. This revision level (reminder, it's just an initial estimate and the final revision was about 400,000 smaller last year than the prelim estimate), is for March 2025. So the overestimates occurred almost entirely during the Biden administration. That's not saying that the BLS Commissioner did anything wrong, she didn't. But Trump will say this proves she was making the numbers higher for Biden than they really were and doing it on purpose. I don't think this will prompt any additional "firings" unless he tries to dissolve BLS altogether, or end the CES program which I wouldn't put past him.
EYESORE 9001
(29,044 posts)
twodogsbarking
(15,775 posts)durablend
(8,509 posts)Wiz Imp
(7,142 posts)He can blame it on Biden and the BLS Commissioner he already fired. See post #2.
twodogsbarking
(15,775 posts)IronLionZion
(49,903 posts)He'll fire everyone until he gets numbers he likes.
progressoid
(51,927 posts)for correcting the numbers.
AZ8theist
(6,896 posts)Wiz Imp
(7,142 posts)However, looking at the monthly QCEW data, it appears the divergence of the 2 programs really spiked in December 2024. And it continued to grow each month through March 2025. That makes some sense. Since Trump was elected in November, it should not be a surprise that job growth could sharply decline immediately following that, particularly with his promise of job killing tariffs.
JCMach1
(28,982 posts)republianmushroom
(21,375 posts)I really love this part,...'signaling economy on shakier footing than realized'. Realized by whom ? Most with any sense knew it was very shaky.
progree
(12,313 posts)Last edited Tue Sep 9, 2025, 06:10 PM - Edit history (4)
occurred in the 9 2/3 months of the Biden administration, and how much occurred in the 2 1/3 months of the tRump administration (remember this report covers the 12 months through March 2025), will be disappointed. I've looked thru all of it --
News release: https://www.bls.gov/news.release/prebmk.nr0.htm
Technical note: https://www.bls.gov/news.release/prebmk.tn.htm
Calculation aggregation procedures: www.bls.gov/opub/hom/ces/calculation.htm#aggregation-procedures
Monthly estimates for the year preceding the March benchmark are readjusted using a "wedge-back" procedure. The difference between the final benchmark level and the previously published March sample-based estimate is calculated and spread back across the previous 11 months. The wedge is linear; eleven-twelfths of the March difference is added to the February estimate, ten-twelfths to the January estimate, and so on, back to the previous April estimate, which receives one-twelfth of the March difference. This procedure assumes that the total estimation error since the prior benchmark accumulated at a steady rate. ((I don't see any table or such of these in the current one, they may be talking about the final benchmark report that will be released in February 2026. In any case, it's just spreading the 911,000 point-in-time March 2025 cumulative difference (or whatever the number is in the final estimate) evenly back through the 12 months, so it's not an attempt to find the actual monthly differences -- Progree))
. . . Benchmark revisions from 1979 forward are available in the CES total nonfarm benchmark revisions table ( https://www.bls.gov/web/empsit/cestn.htm#tb5 ).
See the CES Benchmark Article ( https://www.bls.gov/web/empsit/cesbmart.htm ) for more details about the most recent benchmark.
=======================================
The 911k 12-month number amounts to an average of 75.9k/month average
Here are the nonfarm payroll jobs numbers that were reported 9/5/25 for perspective
https://www.democraticunderground.com/10143524396#post11
The headline payroll job numbers (+22,000 in AUGUST) come from the Establishment Survey
https://data.bls.gov/timeseries/CES0000000001
Monthly changes (in thousands): https://data.bls.gov/timeseries/CES0000000001?output_view=net_1mth
YEAR: JAN FEB MAR etc.
2022: 225 869 471 305 241 461 696 237 227 400 297 126
2023: 444 306 85 216 227 257 148 157 158 186 141 269
2024: 119 222 246 118 193 87 88 71 240 44 261 323
2025: 111 102 120 158 19 -13 79 22
The numbers at this time series will not be revised by the benchmark adjustment until the final benchmark report to be released in February 2026.
(The last 2 months -- July and August -- are still subject to revisions in the regular usual monthly jobs report procedures)
ETA 9/9 300p ET: In the Biden months of the benchmark revision period -- April 2024 thru January 2025 - job growth averaged 153.6k/month.
The average per-month benchmark revision is 911k/12 = 75.9k
If one reduces the 10 Biden months by 75.9k/month, that leaves only 77.6k/month average job growth in the 10 Biden months
(yes, in the above calculation, I'm including January 2025 as a Biden month, even though he was president for 2/3 of the month. That's because the survey week for the Establishment Survey that produces the numbers in the table above is the week that includes the 12th of the month -- that week in January is entirely before the inauguration).
For the tRump 2 months (Feb & March), job growth averaged 111k/month.
If one reduces the tRump months by 75.9k/month, that leaves only 35.1k/month average job growth in the two tRump months
If one accepts the April 2025 and following numbers in the table above as-is, but uses 35k/month for February and March, then one has for the tRump months:
FebMarApr MayJunJul Aug
35 35 158 19 -13 79 22 = 335k = 47.9k/month average job growth in the tRump months
Any benchmark revisions to April 2025 and beyond are likely to be downward, reducing the tRump averageEnd ETA
=======================================
Note that this is a preliminary report:
DallasNE
(7,872 posts)I bring this up because yesterday, a DU poster included a link to a Bloomberg report that is at odds with this CNBC report.
https://finance.yahoo.com/news/another-us-jobs-markdown-sets-100000691.html
This link shows that one year ago, the report showed a downward revision of 818,000 for the period ending March 2024. The CNBC report describes these revisions as "more than 50% higher than last years adjustment ". The CNBC report also states, "for the year prior to March 2025." My understanding is that it includes March 2025, so it should read "for the year ending March 2025". The CNBC report goes on to say, "Most of the time span for the report came before President Donald Trump took office, indicating the jobs picture was deteriorating before he began levying tariffs against U.S. trading partners." Nobody claims that the job picture was not deteriorating as Trump came into office. Of course, the period covered precedes nearly all of the tariffs put in place by Trump.
And this is just from a quick glance.
BumRushDaShow
(160,177 posts)has been around (in it's current form) since the early '90s.
Here was the thread from last year - https://www.democraticunderground.com/10143294175 (same source - CNBC)
The way they characterized it last year was this -
Published Wed, Aug 21 2024 10:37 AM EDT Updated Wed, Aug 21 20243:37 PM EDT
Jeff Cox
@jeff.cox.7528
@JeffCoxCNBCcom
(snip)
As part of its preliminary annual benchmark revisions to the nonfarm payroll numbers, the Bureau of Labor Statistics said the actual job growth was nearly 30% less than the initially reported 2.9 million from April 2023 through March of this year.
(snip)
Meaning that the "actuals" for that period were "30% less" than originally reported (headlines) for that period. In that last year's report, they actually give you what the total of the "initial report" numbers were for the period, which makes it easier to get the calculation.
For some reason, most of the other sites reporting this data release this year (including CNBC, MarketWatch, Yahoo! Finance) were missing the TOTAL of what was was originally reported during that period, in order to more easily compare and calculate the "x%" difference.
Here is a Barron's excerpt from today that actually gives the total of the "initial report" figures for that period to compare -
By Megan Leonhardt
Updated Sept 09, 2025, 11:24 am EDT / Original Sept 09, 2025, 12:01 am EDT
(snip)
The Bureau of Labor Statistics reported on Tuesday that it marked down previously reported net payroll gains by 911,000 in the 12 months through March. That means instead of adding about 1.8 million positions as originally reported, the U.S. economy created only 847,000 jobs. Economists estimates of how big the preliminary benchmark revisions would be varied widely, but most expected to see a reduction in previously reported monthly payroll growth.
(snip)
BaronChocula
(3,333 posts)where the Dukes are watching their fortune being drained on the commodities floor.
I am aware that pisspants' economic policies will hurt people but at least many of them will be those who support(ed) him. I do believe in reaping what you sow.
33taw
(3,237 posts)lastlib
(26,704 posts)I don't think we can put it past this Regime and its toadies.