Following the incredulous May jobs report, I wondered: What are the margins of error assumed by the BLS? Here are the answers.
The above chart is from BLS Current Employment Statistics (CES).
90 Percent Confidence Ranges
Those numbers are for the first release. The numbers vary slightly by revision.
I asked the BLS for the 2023 confidence numbers and received a prompt answer including an Excel spreadsheet.
Thanks!
BLS 2023 CES Monthly Jobs Report 90% Confidence Levels
Those are seasonally-adjusted confidence levels. I do not have unadjusted confidence levels but they should not be wildly different. Because ….
If the unadjusted numbers are not accurate, the seasonally-adjusted numbers cannot be accurate either.
3-Month and 1-Year 90 Percent Confidence Ranges for 2023
- The 3-month 90 percent confidence level is +- 207,300.
- The 1-year 90 percent confidence level is +- 377,500.
I am most interested in the 3-month and 12-month numbers because QCEW numbers are quarterly.
QCEW (Quarterly Census of Employment and Wages) is a count of Unemployment Insurance (UI) administrative records submitted by 11.9 million establishments. That’s about 99 percent of the data.
Nonfarm Payrolls are are from the Establishment Survey (CES). The sample survey was 666,000 individual worksites in 2023. That’s about 5.6 percent of the data.
QCEW will be far more accurate than the BLS monthly convolutions but QCEW lags CES by about 5 months.
How Much Does the BLS Overstate Monthly Jobs?
Note: The numbers in all of my charts are unadjusted. They need to be because QCEW numbers are not seasonally adjusted.
I discussed above QCEW chart on June 6 in How Much Does the BLS Overstate Monthly Jobs? Here’s the Answer
In retrospect, by “overstate” I should say “differ” although in this case I am confident overstate is correct.
CES vs QCEW Full Year 2023
- CES: 155,211,000 to 158,269,000 (+3.06 million)
- QCEW: 152,525,000 to 154,848,000 (+2.32 million)
I do not have confidence levels on QCEW but they should be much more accurate than CES.
For the full year, the difference between QCEW and CES is 735,000.
I struggle to believe the one-year 90 percent confidence range of +-377,500 as stated by the BLS.
Quarter-Over-Quarter Change in Nonfarm Payrolls vs QCEW
What we really need to see is the difference between the reports. Thus …
Quarter-Over-Quarter Change in Nonfarm Payrolls CES Minus QCEW
Quarter-Over-Quarter Notes
- For 2023 Q3, the CES report shows a gain of 3,000. QCEW says the loss is 400,000. The discrepancy is 403,000.
- For 2023 Q4, the CES report shows a gain of 1,424,000. QCEW shows a gain of 820,000. The discrepancy is 604,000.
For the last three years, the quarter-over-quarter discrepancy between the CES change and the QCEW change ranges from -514,000 to +604,000.
The BLS states that the 90 percent 3-month CES confidence level for 2023 is +-307,300.
Once again, I struggle to believe the stated CES confidence levels since QCEW rates to be more accurate.
Let’s look at the year-over-year numbers again. For ease in discussion, I repeat the chart.
Year-Over-Year Change in Nonfarm Payrolls vs QCEW
Unlike quarter-over-quarter changes, that pattern does not seem random.
There appears to be a clear bias at economic turns for CES to undercount jobs coming out of recessions and overstate them headed into recessions.
Otherwise, discrepancies roughly offset.
Directional Bias Questions
- Has something fundamentally changed since Covid?
- Have mass retirements by baby boomers impacted the numbers?
- What about the CES Birth-Death model at turns?
The birth-death adjustment is the BLS estimate of business jobs created by new businesses or lost when businesses shut down.
Mass layoffs have nothing to do with birth-death adjustments, nor do bankruptcies unless companies shut their doors.
Birth-Death Adjustment 2024
The BLS birth-death adjustment added 231,000 jobs to the unadjusted total in May. It added 603,000 to the unadjusted numbers for the year.
Please do not subtract these unadjusted numbers from the reported seasonally adjusted numbers! The Birth-death model did not add 231,00 jobs to the May report as widely reported.
Were there 603,000 new jobs created by new businesses this year ? 231,000 in May?
Birth-Death Math for May 2024
- The unadjusted total payroll for May including the B/D adjustment is 158,918,000 (158.9 million)
- Of that unadjusted total, 231,000 (231 thousand) were from the birth death adjustment
- The unadjusted total before the addition was 158,687,000
The adjustment added 231,000 / 158,687,0000 or 0.15 percent to the unadjusted number.
If the birth-death numbers are randomly wrong, then 0.15 percent does not matter much. But what if the errors are not random?
For the sake of argument, what if the error is half that, a seemingly insignificant 0.075 percent per month, but in the same direction for 6 months?
The cumulative error for six months would then be 0.45 percentage points. Is that enough to matter?
Y-O-Y Percent Percentage Change in Nonfarm Payrolls vs QCEW
In the last six months, the discrepancy between CES and QCEW has gone from -0.10 percent to 0.45 percent.
But 0.45 percent is 735,000 jobs. That’s 122,500 jobs a month.
Curiously, cumulative January-May birth-death adjustment is 603,000. That’s 120,600 a month.
But those birth-death numbers are for 2024. Let’s revisit with 2023 birth-death numbers.
Birth-Death Adjustment 2023
Is it possible that we are headed into recession with far fewer net birth-death jobs than expected?
A historical check, however, shows a recent tendency of the BLS to understate the number of jobs in the birth-death revisions.
Preliminary vs Benchmark Birth-Death Annual Totals
The only negative revisions were in recession years. And those revisions don’t suggest the major discrepancies we now see.
Also, the BLS has revised its methodology in the above timeframe.
For January through May 2024 we have +603,000 preliminary.
For 2023 we have 1,266 preliminary.
2023 Preliminary Birth-Death Numbers
- Q1: 3,000
- Q2: 635,000
- Q3: 264,000
- Q4: 364,000
The birth-death numbers may be a source of the CES-QCEW discrepancy, perhaps a likely source. But unless the benchmark revision is massive, it’s probably not the sole explanation for the discrepancy.
Finally, the 2023 Q4 QCEW number is preliminary. Expect lots of revisions. I believe negative.
Payrolls Rise 272,000 Employment Drop 408,000
The data is confusing and conflicting in many ways.
For discussion, please see Another Bizarro Jobs Report – Payrolls Rise 272,000 Employment Drop 408,000
Nonfarm Payrolls vs Employment Gains Since May 2023
- Nonfarm Payrolls: 2,756,000
- Employment Level: +376,000
- Full-Time Employment: -1,163,000
In the last year, jobs are up 2.8 million while full-time employment is down 1.2 million
Similarly, there is a huge discrepancy between Gross Domestic Product (GDP) vs Gross Domestic Income (GDI)
Real GDP vs Real GDI
GDP and GDI are two measures of the same thing. Income received should match product produced.
However, income is historically low vs GDP.
For discussion of the GDP-GDI discrepancy, please see More Soft Economic Data, Q1 GDP Revised Lower, Q4 GDI Significantly Lower
GDI lends credence to the idea that the household survey (employment) is correct, not the CES establishment survey (nonfarm payrolls) with its assumed birth-death adjustments.
Importantly, QCEW fits in the picture as well.
QCEW, the Household Survey employment numbers, and GDI are in sync. The outlier is the nonfarm payroll report.
On that basis, coupled with weakening trends in hard data, I repeat my unpopular opinion: A Second-Quarter Recession This Year Looks Increasingly Likely
I am passing this post to the BLS to see if I can get some comments.
Addendum
I mentioned QCEW, the Household Survey employment numbers, and GDI are in sync.
I failed to note that the Philadelphia Fed GDPplus indicator and the McKelvey recession indicator are also in sync with the first three.
That’s 5 indicators roughly in sync with nonfarm payrolls the odd man out.
I discuss McKelvey in Is the US in Recession Now? Two Prominent Competing Views
The competing views are Danielle DiMartino Booth and Jim Bianco, both highly respected consultants.
I discuss GDPplus, another great recession indicator, in Philadelphia Fed GDPplus Revised Significantly Lower, But No Recession Yet
Probe into why NHS staff still aren’t getting Covid jabs – as data shows only one in 10 got latest booster in parts of the country https://www.dailymail.co.uk/health/article-12444445/Probe-NHS-staff-arent-getting-Covid-jabs.html
Let me guess – could it be because they are seeing all the maimed Vaxxers coming into the hospitals?
Again – unless vax injuries are acknowledged…these job reports are worse than useless
Say what?
In a recent interview I think it was David Rosenburg (maybe also Jim Bianco, but more likely Rosie) who pointed out that this all starts with the response rates. As i recall he said the raw # of industry surveys received by BLS are falling more than the household surveys. And the household surveys at least come closer to later adjustments made. Anyway- that’s my only recent observation somewhat germane (which is someone else’s).
Probably Rosie
I keep saying that they need to do a 3-month moving average on unemployment and other monthly numbers.
This would give them time to adjust and correct any numbers as necessary while also removing inexact economic numbers from OMG! headline reactions.
Tell me why I am wrong.
Your defective spam filter malfunctioned again. Please free my critical comment from jail.
MISH, an update on the 4th Turning would be TIMELY.
@Mish,
PetroDollar agreement established with Saudi Arabia in the 1970s just ended on Sunday. What is your perspective on this watershed event?
Meanwhile, over in Saint Petersburg, countries who make up a significant fraction of the global GDP are meeting to chart out how to abandon the old, war mongering bully on the block.
https://www.unz.com/pescobar/the-three-key-messages-from-st-petersburg-to-the-global-majority/
Unz is a cesspool of conspiracy theory, misinformation, white/Catholic nationalism, racism and misunderstanding of virtually everything, from politics to economics to history. It has little redeeming value.
For God’s sake Mish… if the first moment (average) is WRONG, why are you now obsessing over the second moment (standard deviation)? Did you take basic statistics?
The problem is the survey data and the idiots collecting it.
They only survey people willing to answer — which rules out almost every productive person in the economy. Seriously, who still answers a random spam phone call or answers the door for an unsolicited door to door salesman? How do you know the guy calling (phone or door) is an annoying government statistician and not an annoying salesman?
Younger Mish was also more aware of global trends like the “lay flat” movement that started in Asia. Turns out that kids who spent years studying and incurred $100K plus in education loans are not excited about delivering food for Uber Eats. Turns out a college degree is not needed to be a glorified pizza delivery driver … not in Asia and not in the USA either.
There are thousands of under-employed college grads who are lounging around depressed. They are not answering their mothers, much less answering the idiot government statistician at the door. Mom doesn’t brag about her loafing child if she answers the door. Between government handouts and Uber Eats earnings, the child is barely able to afford basic necessities in Asia (or if they are in California, one Starbucks coffee and one avocado toast).
It is not in the interest of corrupt bureaucrats to admit that Bidenomics is an epic failure. Pretty sure the government statisticians in France just throw the papers in the trash and go hang out in a cafe. Pointing out that the Japanese PM is old and stupid won’t get a Japanese bureaucrat promoted. And pointing out the CCCP has screwed the pooch is a good way for a Chinese official to get hauled off to a re-education camp.
You are obsessing over the BLS failures, but the “lay flat” movement is global and renders all these statistics even more flawed than they have been for decades. The basic data collection method behind these employment statistics is wrong, which makes the whole thing wrong. Garbage in, garbage out.
A bunch of youth were told to go to college, at great expense, and get a great job. The great jobs don’t exist. Frail, elderly CEOs have diverted all profits to themselves, to corrupt politicians and to illegal aliens — R&D (future careers) has been all but shut down for decades.
Did you see Tim Cook get all excited that the “new” iPad now has a built in calculator? WTF? Apple’s antique Board of Directors is paying this loser tens of millions to add a calculator? Steve Jobs innovations and FoxConn’s labor camps have pretty much run their course, and Tim Cook gives iPad users a calculator!!!
Get a clue Mish. You can’t get meaningful insight from crappy raw data.
You are on drugs. Get better ones. Plonk.
HE KNOWS THAT. This is the point in my mind. I do not pay attention to the NUMBERS, I pay attention to the MANIPULATIONS of the Market, the Interest Rate Trends and try to CUT THROUGH the Bullshit from our Guvment versus what I hear on the Street.
It is ALL Bullshit. Mish is making a living using the numbers that are AVAILABLE and I do not understand what else he CAN use!
The rest of us just look outside, look at empty restaurants and look at LOCK-DOWNS, and VACCINES and FAKE NUMBERS and we KNOW that it is ALL LIES.
Guvmnet agencies are PAID to manipulate numbers to please their masters: THE PROFESSIONAL LIARS known as “Politicians.”
Lets say you want to buy a carpet for your home, and the salesman asks how many square yards you need?
You go home and count the number of bathrooms, and then you wonder why that count isn’t helpful. Number of bathrooms might or might not be manipulated data, but even if you count correctly, it won’t tell you how many square yards of carpet is needed.
Mish and BLS are using survey forms designed for Henry Ford era factory economy, and trying to draw meaningful conclusions about a digital economy. Even if they count the number of factory jobs (non-farm payroll), what does that tell us about a post manufacturing economy?
On top of measuring the wrong things, career bureaucrats have made it clear they have nothing but contempt for the tax paying public. The career politicians are corrupt and only interested in soliciting bribes. But the underlying problem is the crooks are measuring the wrong stuff as well as doing a poor job of it
“They only survey people willing to answer —”
That’s how the process works. People seem to forget that the same limitation holds true for political surveys.
“…the same limitation holds true for political surveys”
True.
I trust the number of people in the photographs of various political rallies.
I do not trust the MSNBC, NPR, Fox, CNN bullsh!t surveys. No matter how accurately they count the “votes” for whatever candidate, its a survey of people dumb enough to watch TV
Sure, their polls aren’t worth conducting in most cases.
No reason to trust digital photos either. It’s pretty remarkable how far the fakes have come.
Excellent analysis. Kudos to Mish. A ton of time & research put into this article. Well done!
Just finished reading a report about how the “new openings” for jobs in the US has dropped 31% in the last 2 years. Something is not connecting correctly if “jobs created” is going up so much but “new openings” has dropped so much. 30 states have seen an increase in unemployment with 19 of those states accounting for 40% of the GDP in country.
Imagine the amount of money that changes hands based on a report that is complete garbage.
End the FED and the BLS.
To end the Fed, you must get the government to balance spending. You need to implement $3.5 trillion in spending cuts per year. The Fed was created to finance WW1 spending (both preparations and then the war itself) – but Washington soon realized they could finance all sorts of bribery after the war ended.
Ending the BLS just involves a spreadsheet with a random number generator function. But the career bureaucrats cannot be fired unless Congress grows a pair and stops supporting corruption.
Washington DC is set up to provide cushy jobs to losers. Thomas Jefferson was quoted saying, “that government which governs least, governs best”. Big government isn’t about competence or good government, its about high paying extortion rackets.
THEY WILL NOT END THE FED or GROW A PAIR. I am not smoking the Good shit, I guess?
I was pointing out the futility of ending the Fed without also ending the reason the fed exists (to fund debt binges)
Got it, Willie.
You cannot end the ones who decide what happens.
Rather than “struggling to believe” the confidence levels, I think the better way to think of it is to believe the confidence level but then to say that the discrepancy, being outside the confidence level, is likely saying that the payroll jobs level is being influenced by something other than random chance.
I think that’s really what you’re getting at, and I agree. I ran the numbers myself off of the raw data files (which are huge and a true pain). The conclusion pretty much has to be that either the payroll survey has a bias (likely), or the QCEW has a bias (not likely) and it’s not random error (highly likely).
RANDOM ERROR IS WHAT THEY DO. Imagine trying to cook up numbers all of this time and keep it from being too obvious that you are COOKING WITH GAS!
We need to factor the record disabilities involving working age people – and the excess deaths…(caused by Pfizer)
Otherwise the jobs reports are useless.
Great analysis, Mish! FYI: The original releases are incredible. No wonder you are incredulous.
I have been agitating at these BLS reports for over a decade. I simply hate all the revisions and even acknowledging them, the data is still problematic. Talking heads run with the ‘Created 377k Jobs’ headlines and all of us paying attention shake our heads. Thanks for the detailed discussion Mish!
Mish, you were one of the first, if not the first, to state that you felt we’d have a recession without a big drawdown in employment. So far it appears you are far more correct then others.
Since Feb 2022, govt employment alone has risen by 1.3mil. Govt spending on “green projects’ and other manipulative ventures has added to this growth in the industrial/power sector. Lets not forget the rise in defense spending with a significant part of the US arsenal sent to Ukraine needing replacement by new hires. The numbers being consolidated are difficult to breakout but the political drive behind employment is not hard to imagine.