Most of this period was under Biden. This is the largest revision in history (excluding the pandemic).

So it looks like about 850k jobs were created, not 1.8 million. That's a mean of 70k per month, mostly immigrants I'm sure.

Quote:

---

Annual revisions to nonfarm payrolls data for the year prior to March 2025 showed a drop of 911,000 from the initial estimate.

---

https://www.cnbc.com/2025/09/09/jobs-report-revisions-september-2025-.html

Quote:

---

The preliminary estimate of the Current Employment Statistics (CES) national benchmark revision
to total nonfarm employment for March 2025 is -911,000 (-0.6 percent), the U.S. Bureau of Labor
Statistics reported today.

---

https://www.bls.gov/news.release/prebmk.nr0.htm

Preliminary estimate = made up number to suit whatever message the DC bureaucracy wants to send.

Ag87H2O said:

---

Preliminary estimate = made up number to suit whatever message the DC bureaucracy wants to send.

---

Definitely prelim but also influenced heavily by the QCEW which is an actual count of employment submitted by employers as mandated by law. The initial measures are imputed (made up) numbers based on falling participation in a voluntary survey.

I would put a LOT more stock into the revisions than I would the original numbers that are now being revised.

Made up BS to prop up Biden and help Kamala and in the face of that with crippling inflation they STILL cut interest rates to further help the Dems.

Now we really will get a rate cut and at .5 but probably a full point.

Obi Wan would be all over this.....

Prosperdick said:

---

Made up BS to prop up Biden and help Kamala and in the face of that with crippling inflation they STILL cut interest rates to further help the Dems.

Now we really will get a rate cut and at .5 but probably a full point.

---

CPI and PPI are both due this week. It'll be interesting to see what Powell does. Most still expect 0.25 this month.

Were the original numbers inflated to help Biden or are the new numbers slashed to hurt Trump? Who knows. We are a banana republic doing banana republic things.

Unexpectedly.

I would love to be this bad at my job and still get paid .

HumpitPuryear said:

---

Were the original numbers inflated to help Biden or are the new numbers slashed to hurt Trump? Who knows. We are a banana republic doing banana republic things.

---

Or ...



......and follow along with me here......













.............govt numbers are trash. Always.

BigRobSA said:

---

HumpitPuryear said:

---

Were the original numbers inflated to help Biden or are the new numbers slashed to hurt Trump? Who knows. We are a banana republic doing banana republic things.

---

Or ...



......and follow along with me here......





.............govt numbers are trash. Always.

---

I agree in general, but they were waaaaaaay off this time and there was no assignable cause, e.g. pandemic, housing crash, etc...

Trump will turn this around

It happened last year too. August is when BLS does their annual revision. https://texags.com/forums/16/topics/3482058

Somehow.... then mainstream media at CNN and NYTimes will slime Trump as causing this.....

Sims said:

---

Ag87H2O said:

---

Preliminary estimate = made up number to suit whatever message the DC bureaucracy wants to send.

---

Definitely prelim but also influenced heavily by the QCEW which is an actual count of employment submitted by employers as mandated by law. The initial measures are imputed (made up) numbers based on falling participation in a voluntary survey.

I would put a LOT more stock into the revisions than I would the original numbers that are now being revised.

---

I don't think enough people realize this. I frankly did not realize how much of the initial numbers are totally guessed / inferred until recently. We should frankly stop the practice because it's not remotely accurate considering the standard deviation from initial estimates until the real numbers are calculated months/years later.

Initial estimates are totally fake news and should be discounted immediately.

I'm glad all of this is coming out. We really need to get better and more accurate information.

I don't think I'm buying it. I looked back at the last 8 years.

In the 2 years prior to the pandemic, the revisions were +0.04% and -0.03%. No biggie.

The last two years, the revisions were -0.4% and -0.6%, two of the three largest revisions in history and both overestimates. They were even larger than the revisions during and coming out of COVID.


eta: looking back the previous decade before Trump and the revisions are small, in line with Trump's first two years.

YouBet said:

---

Sims said:

---

Ag87H2O said:

---

Preliminary estimate = made up number to suit whatever message the DC bureaucracy wants to send.

---

Definitely prelim but also influenced heavily by the QCEW which is an actual count of employment submitted by employers as mandated by law. The initial measures are imputed (made up) numbers based on falling participation in a voluntary survey.

I would put a LOT more stock into the revisions than I would the original numbers that are now being revised.

---

I don't think enough people realize this. I frankly did not realize how much of the initial numbers are totally guessed / inferred until recently. We should frankly stop the practice because it's not remotely accurate considering the standard deviation from initial estimates until the real numbers are calculated months/years later.

Initial estimates are totally fake news and should be discounted immediately.

---

Meh It's based on available data. Much like the AP poll in week one. A guess based on whatever information they have. Even when wrong, I still like to have the best guesses

Logos Stick said:

---

Prosperdick said:

---

Made up BS to prop up Biden and help Kamala and in the face of that with crippling inflation they STILL cut interest rates to further help the Dems.

Now we really will get a rate cut and at .5 but probably a full point.

---

CPI and PPI are both due this week. It'll be interesting to see what Powell does. Most still expect 0.25 this month.

---

I'll be shocked if it's not at least 0.50.

this literally happened LAST YEAR AS WELL!

I remember posting on it.

and last year was one million revisions downward.

at least now we will finally get Fed Reserve rate cuts!

That's a very hopeful, myopic way of looking at it.

If available means what they choose to use, then yeah, fine, use "available."

Right now there is about a 45% response rate to the establishment survey. They extrapolate those responses across all industries and non-responsdents. Over that data, they apply a "birth/death" model that takes the data from the extrapolated data and assumes that x amount of companies were created and y amount of companies were shut down (birth and death). Once they make up how many companies were created or shutdown, they apply the extrapolated jobs figures from the 40% of voluntary respondents across the make believe companies.

They do all that and then publish a number that drives several trillion dollars in capital allocations world wide.

Now if by available you mean the readily available digital data companies use and could potentially publish on a daily basis about hirings/firings that they choose to ignore...then no...they don't use available data.

That is a really, really crappy job of estimating. Whoever is in charge of calculating the estimate should be fired, and her work be audited by competent professionals.

That data from private companies is required to be reported quarterly. It's not available in real time.

Great, then use it quarterly. Why are we still unsure of what the March 2024 period looks like and just getting a prelim revision?

My point is...as things stand now...they actively choose not to modernize or revisit what data is available that they could make use of.

BusterAg said:

---

That is a really, really crappy job of estimating. Whoever is in charge of calculating the estimate should be fired, and her work be audited by competent professionals.

---

The data was much more reliable when private companies were filling out the surveys. We didn't see adjustments like this until the survey participation fell through the floor. It's more of a case of garbage in, garbage out.

Didn't Trump fire the head of the Bureau of Labor Statistics last month over allegations of not reporting correct numbers? Maybe we're just getting the truth now and this is that adjustment.

Sims said:

---

Great, then use it quarterly. Why are we still unsure of what the March 2024 period looks like and just getting a prelim revision?

My point is...as things stand now...they actively choose not to modernize or revisit what data is available that they could make use of.

---

They do that already when they do the revisions to the prior two months, and when they do their quarterly adjustments. But even then, the data isn't complete because some data comes in annually.

Modernization would be great. That's going to take Congress acting and appropriating money to do it. Do you see any desire for them to get off their collective butts and do something about it? I don't.

HTownAg98 said:

---

BusterAg said:

---

That is a really, really crappy job of estimating. Whoever is in charge of calculating the estimate should be fired, and her work be audited by competent professionals.

---

The data was much more reliable when private companies were filling out the surveys. We didn't see adjustments like this until the survey participation fell through the floor. It's more of a case of garbage in, garbage out.

---

So, are you saying that there is no person alive that would have been better at estimating?

I mean, even TexAgs new the numbers were inflated.

The high reported estimates were just another conspiracy that was called a conspiracy theory until the receipts were found.

RoadkillBBQ said:

---

Didn't Trump fire the head of the Bureau of Labor Statistics last month over allegations of not reporting correct numbers? Maybe we're just getting the truth now and this is that adjustment.

---

They did something similar last year. There's a thread about it that I posted up in this thread.

Three of the top 5 largest revisions of the BLS data since 2005 occured in 2023, 2024, and 2025.

Erika McEntarfar was rightly fired.

BusterAg said:

---

HTownAg98 said:

---

BusterAg said:

---

That is a really, really crappy job of estimating. Whoever is in charge of calculating the estimate should be fired, and her work be audited by competent professionals.

---

The data was much more reliable when private companies were filling out the surveys. We didn't see adjustments like this until the survey participation fell through the floor. It's more of a case of garbage in, garbage out.

---

So, are you saying that there is no person alive that would have been better at estimating?

I mean, even TexAgs new the numbers were inflated.

The high reported estimates were just another conspiracy that was called a conspiracy theory until the receipts were found.

---

Given the data they have to work with, probably not. There's only so much an analyst can do with incomplete and in some cases garbage data.

Everyone appeared to throw a wall-eyed fit when McEntarfer was fired. Maybe that was just because Trump did it.

But you're right, the beaurocracy's main goal is to serve and grow itself. Not get better at its job.

BigRobSA said:

---

HumpitPuryear said:

---

Were the original numbers inflated to help Biden or are the new numbers slashed to hurt Trump? Who knows. We are a banana republic doing banana republic things.

---

Or ...



......and follow along with me here......













.............govt numbers are trash. Always.

---

Nope. These numbers are way off. The BLS didn't just suddenly crank government incompetence up to 11.

Sims said:

---

Ag87H2O said:

---

Preliminary estimate = made up number to suit whatever message the DC bureaucracy wants to send.

---

Definitely prelim but also influenced heavily by the QCEW which is an actual count of employment submitted by employers as mandated by law. The initial measures are imputed (made up) numbers based on falling participation in a voluntary survey.

I would put a LOT more stock into the revisions than I would the original numbers that are now being revised.

---

then why report crappy preliminary economic data in the first place?!?!

if it is always wrong (it is) then what is the point of publishing it to the public?!?

The easy fix, at least temporarily, is to encourage more companies to participate in the surveys they send out. You got an idea how to do that? I have one. If you get one of these monthly surveys, and you fill it out and return it on time, you get a tax credit on your corporate tax return. At least someone is getting compensated for filling it out.