Four things: Immigration, Birth-Death Model, Response Rates, Sampling.
As part of its monthly jobs reports, the BLS provides an adjustment series to normalize employment levels to match nonfarm payrolls
The experimental series is LNS16000000, “Employment Adjusted to CES Concepts“.
I was not aware this series existed until yesterday.
For 20 years between 2003 and 2023, the BLS adjustment was remarkably accurate.
But something happened in the late 1990s and again starting in 2023.
Migration Flow 1999–2000
PEW reports The Migration Flow Peaked in 1999–2000
At the start of the 1990s, immigration flows had reached historically high levels of more than a million people a year. In the early and middle parts of the 1990s, the annual numbers held relatively steady at around 1.1 million people a year.
By 1999, immigration levels had increased substantially to the point where the average inflow during the peak period of immigration—a period encompassing 1999–2000 in most estimates and extending a year in either direction in some—reached at least 1.5 million per year. Again, this figure is “an average of averages” that may actually understate the inflows of immigrants, since about 1.8 million arrived in 1999 according to data from Census 2000. The average annual level of immigration during this peak period is about 400,000 greater than during the mid-1990s, or about one third higher.
We know what happened with immigration starting between 2022 and the end of 2024. Biden opened the floodgates.
In both periods, a normally reliable BLS adjustment benchmark went haywire.
For years I pointed out the discrepancy between payrolls and jobs, unaware this BLS experimental series back to 1994 even existed.
2025 Benchmark Revisions
A huge decline in BLS response rates following Covid compounded the problem. And the BLS birth-death model is wrong at turns.
This combination of factors made a remarkable mess of things for the BLS. But the big one is immigration.
In January of 2025, the BLS made a huge adjustment.
The Adjustments to Household Survey Population Estimates in January 2025 were enormous.
The January 2025 adjustment was large relative to adjustments in past years and reflects both updated methodology and new information about net international migration. While the January 2025 adjustment reflects cumulative change back to the 2020 Census blended population base, it largely reflects an increase in net international migration that was concentrated in recent years. To show the impact on the labor force data, BLS uses special tabulations of December data that incorporate the new population controls. When applied to December 2024, the January 2025 adjustment increased the estimated size of the civilian noninstitutional population 16 years and over by 2.9 million people (or 1.1 percent). The adjustment increased the civilian labor force by 2.1 million, including an increase of 2.0 million in employment and an increase of 105,000 in unemployment.
Historical data comparability
The introduction of population controls affects the comparability of estimates over time. In accordance with usual practice, official household survey estimates for December 2024 and earlier months were not revised. Consequently, household survey data for January 2025 will not be directly comparable with data for December 2024 or earlier periods. Although the official estimates were not revised, BLS will produce experimental time series back to April 2020 for the labor force and the employed that account for the January 2025 population control effects. These experimental series will be available in the CPS technical documentation shortly after the publication of the Employment Situation news release on February 7, 2025.
January 2025 Adjustments
The BLS admits that it undercounted employment by 2 million, spread out over a number of years. Instead of parsing that out in the correct months, the BLS plowed the entire adjustment into January of 2025.
This makes all historical comparisons of numbers flawed.
However, there is yet another series in which the BLS parses that 2 million undercount into the correct months.
Benchmark-Adjusted Employment and Nonfarm Payrolls
This is another data series I expect few people are aware of. Again, those are all BLS numbers.
For discussion, please see Labor Force Statistics from the Current Population Survey
The dashed lines show the benchmark revisions to the employment levels and the household survey CPS to the employment levels calculation.
Key Points
- The benchmark revisions to the employment level are much higher than the widely reported BLS employment level.
- Even with benchmark revisions, the adjusted employment level to CES nonfarm payrolls understates jobs.
Immigration largely explains both of these discrepancies.
Employment Level Minus Benchmark-Adjusted Employment
Nonfarm Payrolls Minus Benchmark-Adjusted CPS
For both charts, I subtracted the BLS benchmark-adjusted numbers from the standard reports.
Incredibly Bad Standard Numbers
Many people are reporting unbelievable year-over-year numbers, then attributing the improvement to Trump.
For example, on the Benchmark adjusted chart, Employment in November 2025 was 163,741,000 and November of 2024 was 161,661,000. That’s supposedly a gain of 2.08 million.
But that entire gain occurred in January of 2025 when the BLS threw the whole adjustment into a single month.
If one subtracts the current estimate from the November benchmark-adjusted number, the year-over-year gain is only 359,000.
More benchmark revisions are coming in two more months. I expect negative revisions. Meanwhile, these adjusted charts now put some sanity into year-over-year comparisons.
Foreign-Born Employment Nonsense
The numbers in my charts are seasonally adjusted. Foreign born employment is not adjusted, compounding comparison errors.
And we have no BLS revised data for foreign born employment. So, all such foreign and US-born comparisons with BLS data remain garbage.
A second major problem with foreign-born employment is the BLS makes no distinction between US citizens who were foreign born and genuine foreign workers.
How Fast is the US Shedding Foreign-Worker Jobs?
On December 11, 2025 I asked How Fast is the US Shedding Foreign-Worker Jobs?
Think carefully.
Foreign-born includes US citizens.
There is one more big issue. Since there is widespread disbelief in BLS numbers, why are we supposed to believe this set of BLS numbers?
Are Illegal immigrants suddenly responding to BLS household surveys?
It’s really sad to see a parade of people challenging every BLS report for years, suddenly become believers that that we created 2 million jobs in a single month in January of 2025, all of them US born.
Revelio says foreign-worker employment, not to be confused with foreign-born, is actually on the rise.
Click on the above link for details.
A Word About Percentages
The adjusted numbers in this post are from the BLS. My earlier stab, More Nonsense on the Alleged Surge in US Born Employment, is superseded by this post with actual adjusted numbers from the BLS. (I spread 2 million evenly over 4 years, but the BLS parsing is more refined).
On a percentage basis, 2 million is tiny. For example, a 2-million adjustment to 164 million jobs is only only 1.2 percent.
However, a 2 million adjustment that turns a year-over-year gain of 600,000 into 2,600,000 is a stupefying error. Yet, it’s been repeated over and over again.
Once again, the historical non-revised employment, job, counts, and labor force numbers from both the household and establishment surveys are garbage.
However, percentage numbers such as the unemployment rate are still valid (discounting response rates, sampling errors, etc.)
I crunched the adjusted numbers for unemployment rates with adjusted labor force numbers and there was not a single difference in the rates between the headline numbers.
Addendum
Key Addendum Point: The foreign-born and US-born numbers are still garbage. But with my chart we now have reasonable YOY comparisons for employment.
I expect a smaller benchmark revision in February (January data), this time subtracting employment.
what is the expectation for the February revision now ? the low and the high?
Good question Jake
It could be significant, but I doubt as big as the +2 million.
I am hearing numbers like -900,000 but I have not looked at them.
I have not put my own guess on this.
thank you for your answer, i am expecting -900 k to -1.2 million, when i think about where we at compared to 2007, unemployment was 4.7% then it jumped to 5% in December 2007, also i am having trouble believe the market odds for january 0.25% cut dismissed to this level, a simple tally show miran, williams, waller, bowman, paulson most likely would vote a yes for a jan cut, who can own a no cut followed by a terrible downward revision in february the cost is too high, what would they do , no cut jan then bad revision, unemployment jump to 4.9 / 5 level, emergency meeting ?… i would be very interested on your take on this.
Thank you for bringing this to light.
the truth and the facts are what you see out your car window
drive far and wide with your eyes open and your mouth and ears shut to opinions
the only way to measure value is to compare things to other things, price does not matter money defines the difference it does not establish it
dismiss fear and greed from your personal value equation and enjoy the majesty of the passing of scenery—-that deefines time
If you are gullible enough to believe the government’s latest inflation “estimate,” then you probably believe the verbal vomit that came out of taco’s mouth last nignt.
“It’s really sad to see a parade of people challenging every BLS report for years, suddenly become believers that that we created 2 million jobs in a single month in January of 2025, all of them US born.”
It’s simple, that talking point serves the interest of bigots and racists that have been claiming for years that “illegals” have been “stealing” their jobs and driving down wages.
Ironically Trump took office in January so if the narrative were true, it was Biden that did it not Trump. Was Biden deporting millions of people in January?
Why would anybody place any credence in the BLS Questimates ?
Didn’t John Williams at Shadow Stats years ago indicate that a better indicator of real employment was to look at the employess salary withholding taxes each moth . . . More than likely its closer to the truth than the BLS Rubbish . . .
I have been saying that for years. In order to work in the US legally, the first thing an employer asks is for your social security number. Even if you’re here illegally, you need to provide some tax number.
Why the BLS just doesn’t work with social security admin to count the number of active tax IDs paying FICA is beyond me. Perhaps it would eliminate tons of jobs at BLS since it would be too efficient.
At this point it may be even easier to simply use cell phone tracking to count the labor force. People usually home in the evening so track that as “not working” and during the day, track the cell phone movement. If cell phone moves from a residential area to a work area, it’s likely a work location so count that as employed.
No, it won’t be perfect but it’s far better than these dumb surveys.
The demographics of America, such as over 65 million abortions since Roe v Wade and zero population growth, are having real-life impacts on US society, economics, and national defense survival. A nation with zero population growth is like a business that is “liquidating” all its assets. It may look wealthy and comfortable for a while, but it is effectively consuming its foundation. We are in deep trouble.
Well said. For Europe you can add in their stupidity of allowing an islamic migration that out breeds them 2.5 to 1. Probably higher
For the U.S., before we just had a mexican open border of mostly latinos, and most want a better life and willing to work for it, and assimilate so I could live with that.
Until the Biden Admin with the United Nations help to circumvent congress opened up the border to the entire globe and brought in too many from countries that do not even have a rule of law, Somalia comes to mind, and other islamic countries and that have no intention of assimilation.
Down vote the truth away all you want. It does not make it untrue.
Down voted because you’re a racist, it’s that simple. You call it truth but it’s bigotry, which is not truth whatsoever.
I feel sorry for you, Phil in CT. David is correct, it’s that simple. Maybe you are having your TDS flare up? Dunno. Multi-race lover here.
He is an over the top Trump hater. It really clouds your judgement
Facts are only bigotry to left wing liberals like yourself.
Its kind of like what Jesse Jackson said.
“There is nothing more painful to me than to walk down the street and hear footsteps and start thinking about robbery, then look around and see somebody white and feel relieved.”
How does this fit into your perspective?
https://www.foxnews.com/politics/trump-doubles-down-plan-600000-chinese-student-visas-despite-maga-backlash
President Donald Trump on Monday defended his administration’s plan to offer 600,000 visas to Chinese students — a move that has drawn criticism from within conservative circles.
Trump argued that foreign students — particularly from China — are vital to keeping U.S. universities financially stable, adding that cutting the number of Chinese students in half would cripple the system.
This is what you apologists for the far left liberals do not get.
Dude, I have not watched Fox news in over 30 years give a take a month or 2
So no I am not watching the crap that comes out of there either nor your link.
The problem with people that think like you and phil and all of you apologists for what the democratic party has become is if you do not go along with the standard democrat talking points and go off the plantation, you are a racist a nazi blah blah blah
For the record, I have bi-racial kids and a DACA girl in my family whom I love like my own daughter
What did Malcom X say, He considered the white liberal a “hypocrite” and a “fox” who smiled while the wolf devoured black people
(at that time the wolf was the southern racist, now its 80% of you white liberals globally)
So no real response Trump’s plan for 600,000 Chinese imports. Elon Musk wants 1 million H1B visas which will probably eventually happen, all under republican controlled government.