173,500 payroll jobs added on average per month in 12 months, instead of 242,000, but above 2018-2019 average of 171,900: preliminary annual benchmark revisions.
By Wolf Richter for WOLF STREET.
The Bureau of Labor Statistics today released its preliminary estimate of the upcoming annual benchmark revision due in February, of the number of nonfarm jobs created over the 12-month period from April 2023 through March 2024.
The BLS today downwardly revised by 818,000 the number of nonfarm jobs created in the 12-month period through March 2024, to 2.082 million, from the prior figure of 2.90 million.
This revised figure averages out to 173,500 payroll jobs added per month during the 12-month period, down from the unrevised average of 242,000 per month, but higher than average job creation during the prepandemic two years of 2018 and 2019 (+171,900 per month on average), reflecting a labor market that has normalized to prepandemic Good Times growth.
Up-revisions: Job creation in five major categories was revised up, led by private education and health services (+87,000) and Transportation and warehousing (+56,400).
Down-revisions: Job creation in other categories was revised down, led by Professional and business services (-358,000) and Leisure and hospitality (-150,000).
| 12-Month Benchmark Revisions of Jobs Created through March 2024 | |
| Private education and health services | 87,000 |
| Transportation and warehousing | 56,400 |
| Other services | 21,000 |
| Utilities | 1,700 |
| Government | 1,000 |
| Mining and logging | -11,000 |
| Wholesale trade | -33,600 |
| Construction | -45,000 |
| Information | -68,000 |
| Financial activities | -76,000 |
| Trade, transportation, and utilities | -104,000 |
| Manufacturing | -115,000 |
| Retail trade | -129,000 |
| Leisure and hospitality | -150,000 |
| Professional and business services | -358,000 |
Each year, the BLS benchmarks its employment estimates in Current Employment Statistics (CES) – which produces the nonfarm payroll jobs data from the Establishment Survey in the jobs report – to state unemployment insurance (UI) tax records that most employers are required to file.
The Establishment survey data in the monthly jobs report will not be updated with these preliminary revisions. Instead, it will be updated with the final revisions that include monthly details, to be released in February. We will cover that as we have in prior years with a chart with two lines, one color for the new revised data and another color for the old data. Final revisions have differed significantly from preliminary estimates in the past.
In prior years, final benchmark revisions released in the respective Februarys:
- For 2021 and 2022, final benchmark revisions, in total +840,000
- For 2023, final benchmark revisions: -359,000
Enjoy reading WOLF STREET and want to support it? You can donate. I appreciate it immensely. Click on the beer and iced-tea mug to find out how:
Would you like to be notified via email when WOLF STREET publishes a new article? Sign up here.
This does not mean that a rate cut is imminent or even favorable at this time, but it does add support for a rate cut.
The Fed (Powell, Bowman, etc.) has been mentioning a downward revision for a while, and it’s already baked into whatever decision they’re going to make.
It affects only data for the 12 months through March. So it doesn’t have any impact on current data since March.
Going forward: When government statisticians have missed the last count by 30 %, The stastistics per se are “not impacted” but confidence in them certainly is.
If you want certainty, go to church.
Total payroll jobs of 151.8 million were revised down by 0.5% today, or by 818,000. So it’s a 0.5% revision. You’re talking a rate of a rate.
There once was a report done that said inflation is wrong by like a lot, leaning towards a lot less inflation.
Plus the fact that 3 banks got the data before everyone else as reported by Bloomberg makes it hard to trust the institutions legitimizing the government.
Bloomberg:
After the Bureau of Labor Statistics failed to post its revisions to the monthly payroll figures at 10 a.m. New York time, Mizuho Financial Group Inc. (MFG) and BNP Paribas SA (BNP.PA, BNPQY) both called the department and got the number directly. So did Nomura Holdings Inc.’s economic research team, according to a person familiar with the situation.
Anger quickly mounted as word spread across Wall Street that the BLS had begun giving out the numbers over the phone. A scramble ensued, with other firms and media outlets, including Bloomberg News, trying to obtain the figures, too….
“I don’t wonder that people are upset,” Nancy Tengler, the chief executive officer of Laffer Tengler Investments. “The whole thing reeks of incompetence.”
I realize we’re talking about the government here, but how can their figures be off by 30%. And this is acceptable? Imagine a private company revising their sales or profits or inventory or anything by that amount.
Bonuses for everyone involved
CCCB
Total payroll jobs of 151.8 million were revised down by 0.5% today, or by 818,000. So it’s a 0.5% revision. You’re talking a rate of a rate.
They could have revised it to negative 2.9 Million. So this is pretty good.
cccb?
you mean the same gov’t that hands nearly a trillion dollars a year to the defense department, which has YET in its history to pass an audit?
do you mean the gov’t that abruptly pulled out of afghanistan after 20 years and trillions spent, all for nothing? oh sorry, the taliban got all the billions worth of equipment left behind.. foreign aid is an important part of foreign policy, after all..
perhaps you mean the gov’t that forcibly nationalized the entire student lending industry, pouring even more gasoline and dynamite on exploding higher educational costs? but is now seemingly determined to forgive or substantially write down large portions of the resulting student debt?
could it be the gov’t that continually, without any sign of slowing down, spends ever increasing numbers of dollars it DOESNT HAVE on its obligations..
is it acceptable?!?
i ask you sir, what alternate version of reality have you been living in your entire life?
even trying to compare a private, (or public) company to the gov’t is like comparing a forest fire to a supernova. there are some similarities, but the one is so far removed from the other as to be in an entirely different category, unbound by the ‘rules’ which govern and restrain the lesser…
i suggest you take a good hard look around, because it seems to me you’re missing QUITE ALOT…
a revision of 242,000 jobs monthly to 173,500 for the last 12 months is a 28% monthly or annual revision. Looking at it from the perspective of total jobs it becomes a .5% revision. It is a matter of perspective.
If you want certainty there is always death and taxes.
How does that add support for a rate cut?
“This revised figure averages out to 173,500 payroll jobs added per month…..higher than average job creation during the prepandemic two years of 2018 and 2019 (+171,900 per month on average)”
Still looks like a healthy job market. Of course, wall street wants a rate cut as usual.
Nonfarm job growth is 1600/month higher than 2018-2019, but population has grown, so growth is essentially flat (solid/good), right?
No, you’re mixing apples and oranges.
BTW, here are the prior two “revised v. old data” charts, based on the final benchmark revisions after they’re integrated into the jobs data in February:
at a trillion of deficit spending every 100 days how is this possible?
No kidding. It should still be skyrockets in flight. Jobs and raises to the moon delight.
Maybe the deficit spending is going more directly to the rich now, to heck with trickle-up.
steve/brant lee
it should be skyrockets, raises, etc??
no.. perhaps you confuse this spending with direct fiscal stimulus, as was done during the COVID times. that stimulus resulted in inflated costs everywhere.. many gov’t jobs and social programs have some form of automatic/periodic ‘cost of living’ adjustments. also, the gov’t has to pay alot more to interest than it once did. these are 2 examples of why what you both suggest isnt happening quite like it ‘should’.
for a related allegory, corporate structures are usually divided into ‘cost’ centers, and ‘profit’ centers. most corporate functions are costs, which only increase over time (unless layoffs occur). the company’s product/service line is the only thing producing revenue/profit. the purpose of a business is to discover wants/needs and offer products/services to fulfill them, in a way that generates revenue/profit. business can expand its offerings to ‘capture’ more revenue, or reduce its offerings and focus on high efficiency to do the same.
with the gov’t there is no such distinction. EVERY FUNCTION is a cost, and ‘layoffs’ simply dont happen. the very nature of bureaucracy is to continually expand, which causes the same for budgets. there are only 2 choices to ‘keep up’.. extract more from taxpayers, or create debt and sell it to fund operations. the latter is the preferred choice because it is far easier from a political standpoint.
there is in theory, a third choice.. a mandatory balanced budget. this would mitigate in many ways both bureaucratic ‘creep’, and financial profligacy. however.. it would need to be a constitutional amendment to work right and you can bet it would never pass..
There is a fourth choice: incentivize government managers to run their programs to meet financial goals based on quantity/quality metrics. But citizens would hate the idea of government employees making millions even if they were providing even greater efficiencies. And businesses would hate the idea of citizens thinking government could run things efficiently.
Based on market’s reaction, another nothing burger that I am sure DiMartino Booth or the YT doomers will claim the next great depression is right around of the corner because of downward revision..
Rinse and repeat…
All these doomers made lot of money from their clients as well as money on YT hence they all have fearful click bait titles.
honestly tho, they may be onto something. Literally step outside and ask your network if they are buying any of the big ticket items: houses, cars, home improvement projects, vacations, or any >10,000 ticket items.
All of those are raw materials, jobs, etc. GDP is being juiced by deficit spending but unless they inject that cash into people’s wallets directly the consumer is not going to be able to keep growing the economy.
65% of GDP is services, not goods.
In terms of housing, it’s a big factor in GDP as it is being constructed — it falls under residential fixed investment. But when you buy an existing home, the impact on GDP is almost nil. It’s only when people actually spend the sales commissions they received for the sale that GDP gets a piece of it.
Auto sales are running at the fastest pace since 2019. Vacations are still red hot. Ecommerce sales, the second largest retail category after autos, are hitting huge records. Record Labor Day travel is being projected, along with airport chaos. Restaurants are still going strong. etc.
I know the YT bunch is on to SOMETHING; as others have said it can easily become an echo chamber.
There’s always the extremist: Loud, far out idea etc. Those definitely sell!
However, my conclusion is actually: This time IS different!
There’s no new “stuff” going on, but the combination of these things is a bit novel.
The economists (and commenters here) know that the largest economy in the world is running an unsustainable debt. This is not new, but the size of the economy is.
Also, the series of global disruptions and the fallout/ knock on effects that have occurred (in the biggest, most globalized financial/ social system ever) and are still occurring (conflict, change of power) are creating a disruptive mix.
Add on top the technology and presence of the “gig economy” (times crypto), we see:
When trying to unravel the threads of what’s happening in the national and global economy the tools of yesterday are groaning under the task.
Food and energy resources are generally abundant. It’s more personal and societal pressures that are driving people’s decision making. The ability to accomplish things outside of the traditional channels is unprecedented.
Therefore the ability to gather the data, interpret and draw conclusions is inhibited in “new” ways. Classic economics is being hampered by being an artifact of a government office.
The wealth gap is a major stumbling block, as the concentration of wealth also means a concentration in power, influence and ability.
Tech companies probably have a better line on the pertinent factors in the economy than any bureau, they’re more concerned with what shiny things the users want to consume and even government has forgotten about the people’s wellbeing.
Who is DiMartino Booth? Who or what are YT Doommers?
Feelin’ pretty dumb here…
She owns Quill intelligence and she used to work within the Fed. You can see her videos on YouTube regularly about the economy. She’s smart and up to date(imho).
Don’t waste a second of your time.
I guess the horde of channels on YouTube that people search for if they believe everything is going to $hit. Then the channels reinforce the belief they have.
This is actually how conspiracy theorists work. The old crazy loop
August 2008: those clueless doomers
August 2024: those clueless doomers
When it happens, they will all claim that nobody saw it coming. People claim that economic cycles don’t exist anymore.
So what is the right policy prescription for this? The federal funds futures markets are now pricing in 2.5% of rate cuts by September 2025, from 5.25-5.50% to 2.75-3.00%.
Lower rates can’t create millions of jobs out of thin air. What they can do is inflate asset prices, many of which are conveniently excluded from inflation indices.
Based on their past accuracy, the only thing Fed Fund futures can tell you is what the rate will NOT be.
Inflation is still pretty hot in many categories
Federal funds futures markets are nearly always wrong. Here are my two favorite pictures:
This along with WF’s comments above on the services sector makes me consider no rate drop tomorrow! Still above target a little on inflation too, as far as I can tell. Mebbe one more month delay? Interesting Times…
All of these failures are optimistic takes that didn’t come to fruition. 2008 – 2018 “return to normal” 2019 plunge unexpected, 2022 onwards “this is as bad as it’s going to get” and today “they are going to drop back to 2% but NOT go negative”
Another perspective is that the market is leasing the fed by trying to front run the moves. Eventually when enough bond market participants believe something should happen, the fed must follow or risk blowing up the entire financial system by disrupting debt markets. So ask yourself who really runs the show?
Companies can borrow money more cheaply.
They borrow then hire and buy equipment and ads from Google!
Really nothing to see, they just revised the numbers to pre pandemic “good time” growth.
Not a luxo burger, just a “good time” burger with a small order of fries.
Always good perspective. 👍
Wolf — what causes such massive revisions in the first place and how does this compare with upwards revisions to monthly jobs data that you’ve previously talked about here?
Over the past few months, I’ve been talking about the “household survey” data from the BLS (total number of workers, the unemployed, the labor force, the unemployment rate, etc.). This data has been wrecked by the non-inclusion of at least 6 million new immigrants in 2022 and 2023 (according to the CBO), many of whom are working or are looking for work. That will eventually get fixed, and we will get that adjustment in the future. And it will be big.
This here is the revision to the establishment survey data (nonfarm jobs). The revisions are based on the unemployment insurance tax filings that companies have to file. But this UI data doesn’t include illegal immigrants that are working because they’re not eligible for UI. And this has been pointed out today — that the downward revision ignores the influx of illegal immigrants into nonfarm jobs.
Wolf — what might be helpful, if you happen to know, is a quick explanation of how the establishment survey data is collected and what causes an up or down revision. I’m certain a lot of people reading this blog, myself included, have 0 idea.
You’ve covered this somewhat explaining how the inflation data on housing gets collected, so that was interesting & helpful.
Thanks
There are several brief explanations in this thread, including here:
https://wolfstreet.com/2024/08/21/nonfarm-job-creation-for-12-months-through-march-revised-down-by-818000-to-2-08-million-jobs-created-from-2-90-million/#comment-602146
3/4 of employment growth since 2019 went to immigrants. The trend in the prior 12 months ending in March probably hasn’t changed. Based on that trend, what kind of jobs could have been filled?
Yes, this is a big issue in the data. The revision today is based on unemployment insurance tax filings by companies; and since illegal immigrants are not eligible for UI, they’re not part of this data. People have pointed this out today correctly; today’s revision understates employment by the number of illegal immigrants that have been hired into nonfarm jobs over this 12-month period.
According to the CBO, 6 million immigrants were added to the population in 2022 and 2023. Some of them were legal (H1B, new green-card holders, etc.), and they’re reflected in the revision. But illegal immigrants are not.
How is hiring illegals legal?
You’ve never worked a job getting paid under the table?
“You’ve never worked a job getting paid under the table?”
No, I’ve never illegally entered another country and worked for cash. Your point?
It’s hard to look at this and realistically take future numbers serious aside from the market trading off of initial numbers. This seems like a big miss. How does this stack up to past revisions? There has to be a better way to get this figure.
Total payroll jobs of 151.8 million were revised down by 0.5% today, or by 818,000. So it’s a 0.5% revision. You’re talking a rate of a rate.
Thanks for painting the picture!
@Wolf
Yes, the job numbers might have been revised down. But still our drunken soldiers have spent a lot during those times and continue to spend now as Walmart and Target said. Also, the core inflation is ticking around 3% and perhaps even trying to move up. There seems to be NO compelling reason to reduce the IR now. But perhaps some hedge fund companies in CRE are cracking and Powell has to come to their rescue – like we did in 2008 for Bearn and Stearns.
Are your figures in the comments right? Revised data (red) higher than old data? Or they are from two different sources? TIA
Also, it looks to me all the cheer leading for IR cut seems to miss the point — If inflation stays around 2.5%, the FED rate has to be around 4%; 20 year and mortgage rates around 6%. The FOMO and stonks always go up crowd doesn’t need that small cut anyway. It is not going to help the housing sector either.
“Are your figures in the comments right? Revised data (red) higher than old data? Or they are from two different sources? TIA”
Yes, they’re right. I gave you the numbers in the article at the bottom.
It always feels funny how on the day those monthly numbers are released, wall street is celebrating ‘1000 jobs more than expected’ and stocks surge.
Only for it to be revised down 170k jobs a few months later, to no wall street effect.
Exactly. It makes one wonder if the overstated jobs are intentional for a Wall St. pump. Further, the gloating from the administration looks a bit sus in hindsight.
I can’t help but wonder how much of this rush by central bankers to ease policy is out of genuine concern for the labor market (“their second mandate”), versus the desire to pump Wall Street & juice their own investment portfolios.
While the recent slowing in economic data would appear to suggest some modest easing might be prudent, they started talking about doing this easing in late 2023, when the unemployment rate was in the 3’s, 500K+ jobs were being added monthly, and the economy was showing NO signs of slowing down.
The fact that nearly all FOMC members own millions in investments, and were caught trading their portfolios in 2020-21, makes the public suspicious about their true motivations, even if they now have the political cover to start easing.
Wow! That is a pretty big miss! Does the Department of Labor ever make permanent changes to their procedures as a result of this big of a miss?
“The mystery of the disappearing jobs”
“Inspector wolf” on the scene would know the where and when. Having worked wolf streets for years, he would know where to look, pre-revision and post-revision. Be any weasel play, wolf, a numbers man, would solve the mystery of the disappearing jobs.
This being a short article, I think he will solve the mystery in his next article in “the jobs that never were”
Total payroll jobs of 151.8 million were revised down by 0.5% today, or by 818,000. So it’s a 0.5% revision. You’re talking a rate of a rate.
Wow copy-paste 4 x already. People really don’t read the comments much do they… :'(
Wait till they realize they were counting same 173,000 jobs every month.
OK boys…..we have a job to do…..let’s cut. After all, we’ve reached full employment……going the other way……Wilma might have only one job offer……heaven for bid.
but…..how bout the inflation……
we’ve already explained that once…..its transitory
pass the bottle……or in this case the white powder.
It’s as if Senator Blutarsky is in charge from animal house……as if?……nope……
Oh well…..yellow rocks to $5000 an ounce as a first stop……easily.
Very recently, there has been a jump in new pending sales. The lower rates have really juiced the housing market. This will not show up in data for a little while … perhaps a month or two.
Here’s your jump in pending sales 🤣
And here is the collapse in purchase mortgage applications, hot off the press:
https://wolfstreet.com/2024/08/21/home-buyers-strike-expands-even-as-mortgage-rates-drop-to-lowest-since-may-2023-but-surging-refis-will-speed-up-the-feds-qt/
What color is the sky in your world?
Gold rocking new highs too! :)
“Juiced” the housing market? Sounds like you’re hitting the gin ‘n juice, because you are so full of shid your eyes are brown.
Lots of people writing that a rate cut will juice the market. Maybe those prognosticators aren’t looking at history. First of all, the last two big rate cutting cycles saw the markets crash (2000-2002 and 2007-2009). Secondly, the market has ramped the last 22 months with the highest rates in 16 years. So, maybe it’s better to look objectively at what’s happening.
The Fed govt has $35T in debt, with lots coming due all the time. Large amounts of that are moving from longer term notes/bonds to short term bills that are paying 5.35%. Ergo, retired people, wealthy people, and rich corporations are all getting huge pay increases from safe money investments they’ve had for years. That’s helped to keep the retail, travel, and restaurant biz going gangbusters the past few years now that the “free” money has run out.
The Fed has to lower rates for several reasons and I think that will reduce spending because of what I wrote above which will negatively impact the economy. The reasons are:
1) Fed Govt run rate of interest is approaching $1.5T/year. That’s not sustainable, especially w/ a $2T deficit.
2) Small companies and small real estate developments that began in 2019 or 2020 with high hopes were smacked by the response to COVID and are now being crushed by the high rates. I know this personally. 1000s are at the end of their rope and will go belly up in the next year if rates don’t reduce drastically.
3) The only way to loosen the residential RE market is to either build more homes or make it easier for people to sell. Builders don’t want to build at 8%-10% (the pipeline right now is getting slim) and sellers don’t want to sell with their 3% mortgage in place. Bring mortgages down to 5% and I believe we’ll start to see some serious action in the residential market.
My 2 cents worth anyway.
They started cutting rates AFTER things were going to heck. That’s the difference. Nothing is going to heck right now. Which markets or assets are we talking about? Case-Shiller continued to rise throughout the 2000 rate cuts. It was falling before the 2007 cuts and continued to fall.
So getting like 1-2% over inflation for free is causing the economy to overheat?
lol keep spouting that good stuff. 🐳
MW: Fed minutes flag Treasury-market leverage as a risk to financial stability
So every jobs report was off by an average 68,500 every month for an entire year??!! WTF!!!
That is simply reprehensible when you can’t trust the government’s economic data. What other inaccuracies have they given us?
Still standing firm with my prediction that the S&P is a lock to see 4500 again. More confident now than ever!
Government: US economy added 818,000 fewer jobs than first reported in year that ended in March, worst revision since 2009
-AP headline
WTF is right.
Clearly the statistical bureau is stoned. Who cares though, only a third off.
There was a huge upward revision in 2022 (+840,000) that was bigger than this downward revision in absolute numbers, see the two charts above and bottom of article. So the AP is wrong: it wasn’t the “worst revision since 2009,” it was the “worst downward” revision, but it was a smaller revision that the upward revision in 2022.
Apologies to the Wolf pack for duplicate, not intended. Appreciate the talented, reserved, and objective comments here.
Ace,
Total payroll jobs of 151.8 million were revised down by 0.5% today, or by 818,000. So it’s a 0.5% revision. You’re talking a rate of a rate.
S&P was at 3,500 when FRB stopped raising rates.
Everything since then is basically front running a return, via recession or black swan event, to QE/ZIRP, against a background of QT and mag7-alike action.
What people seem to miss is that the scenarios required to allow that to happen means a big U or V shape in between now and then.
I’d suggest 3,500 is a sensible point to call a bottom of that dip as that’s what people thought it was worth in late 2022 with a modicum of actual fear and risk aversion in their eyes.
BTFD4EVAr1!!1!!
Wrt this revised data.
I’d be keen to understand how you revise data by such magnitudes.
I can only assume early data is guessed/extrapolated and the revisions are actual data being properly processed?
Dont say I didnt warn you
While these numbers show the economy really didn’t create as many jobs as the government claimed, there has not been a huge uptick in jobless claims over the same period, and unemployment has been pretty steady. What spikes that have occurred have mostly been seasonal or weather related.
The slow train wreck continues. Some cars have derailed or detached from the engines, but others are going full speed ahead. The Fed definitely needs to throw at least 50 basis points of coal in the hopper and keep things hurtling towards the bridge out ahead. 😬
It’s impossible that the real job numbers are unknown to the government; the Social Security number of the employee is reported and the employer has to pay a portion of the Social Security beyond withholding. Any “job” without a social security number by an undocumented worker is not a legal job to be counted anyway. The IRS doesn’t release info for any purpose other than taxes, but the agency in question is the Social Security Administration.
SS taxes are paid quarterly. And they’re not deducted from contractors. So the SS data is worthless for employment metrics.
Today’s revision is based on unemployment insurance tax filings by companies. This is tracked separately, and today, the jobs data was benchmarked to the UI data.
But the UI data is incomplete too because it doesn’t include illegal immigrant workers because they’re not eligible for UI. They’re the missing element here. So today’s revision understate the actual jobs, which lots of people pointed out today.
Question. Do the illegal immigrant workers ever get included in the job numbers, and if so, where does number come from?
The revisions are based on UI data that does NOT include illegal immigrant workers.
But the payroll data that was revised comes from surveys sent to companies, and companies that have illegal immigrants on their payroll report the number of their total payroll. The survey doesn’t ask for immigration status. So illegal immigrants are included in the payroll data. That’s where the disconnect comes from.
Payroll data would probably still understate the numbers though. I don’t think you can assume a business that hires illegal immigrants is going to be honest in a payroll survey.
Why don’t they pass E-verify, to eliminate employers hiring illegal aliens. What the hell are they waiting for. Neither party has even mentioned this tool. Someone is making a killing scamming the system and making sure that honest Americans can not earn a living wage.
Swamp – …think the issue might be an acknowledged, but unwritten, larger general resistance to the inevitable cost increases (that would be passed on to the citizenry) to those who currently dodge them (…dodging often ‘justified’ by claims that this is work that ‘Americans won’t do’, ignoring that Americans perhaps ‘won’t do ’em’ because those jobs are under respected/underpaid/under regulated viz worker-safety, etc.).
may we all find a better day.
I think Wolf needs a new graphic similar to the one with Jerome tearing his hair out, except this one shows Wolf tearing his hair out over commenters who DRTGDFA or fail at basic maths.
So the majority of jobs created have been lower-paying, even irregular jobs for the past years?