Once Again, The (Biden-)Harris Job Numbers Are Proven Fake
Lou Costello Is Alive And Well And Working For The BLS
In the ordinary course of human events, we generally have to contend with three types of lies: lies, damned lies, and statistics.
In the ordinary course of dealing with government bureaucracies, however, we must add a fourth category: Bureau of Labor Statistics data. We are reminded yet again of this essential truth by the latest Early Benchmark Revisions to BLS employment data published by the Philadelphia Federal Reserve.
Estimates by the Federal Reserve Bank of Philadelphia indicate that the employment changes from March through June 2024 were significantly different in 27 states compared with preliminary state estimates from the Bureau of Labor Statistics’ (BLS) Current Employment Statistics (CES). Early benchmark (EB) estimates indicated lower changes in 25 states, higher changes in two states, and lesser changes in the remaining 23 states and the District of Columbia.
What the Philly Fed’s work shows once again is that the BLS data is simply unreliable. What the Philly Fed’s work shows—again—is that the BLS apparently has no idea how to fix either its data or its data collection methods.
Along the way to confirming that the BLS still prefers Lou Costello Labor Math over actual data and sane analytics, the Philly Fed’s work also shows that we truly are in a “jobs recession", and that it is getting worse rather than better.
Longtime subscribers will know that I have more than once been critical of how the BLS handles its employment data. Even before these latest mistakes and mis-steps, it was plain that the official employment data was off kilter.
Nor should we forget the admission by the Bureau earlier in the summer that over 800,000 of its reported jobs created in 2023 never actually happened.
The error the QCEW revisions found this time was in the second quarter the US actually lost jobs.
The forecast that the Philly Fed makes about what the data will show for the rest of 2024 indicates that job loss, rather than job growth, is going to be the dominant employment trend.
Adding insult to injury, the Philly Fed also found that a very large number of states’ jobs numbers have been overstated by the BLS.
Far from having the robust and reslient jobs market that corporate media has been touting throughout the (Biden-)Harris Administration, the United States is—and has been—in an extended “jobs recession”, just as I have been asseting for several months now.
When we look at the Philly Fed’s numbers from the first quarter, it is clear the BLS methodology went significantly off the rails last year, and it has not been corrected.
Historically, if we add the state-level employment figures together for all 50 states plus the District of Columbia, the total approximates the nationwide employment level reported by the BLS each month.
Ironically, this is true even though the BLS cautions against combining the two data sets in this fashion.
Despite this caution, however, what the state totals reveal is that, after June, 2023, the state totals broke from the nationwide aggregate data.
Unlike earlier variances, however, the two charts have not converged shorly after breaking up.
Not only is this confirmed by the first quarter QCEW data, that same QCEW data indicates that there very likely has been job loss during 2024.
That first quarter data certainly confirms that job growth in 2023 was somewhere between slim and none.
All the hundreds of thousands of jobs the (Biden-)Harris Administration presumably created? They don’t exist.
That same June, 2023 break is still present in the second quarter QCEW data.
The BLS has not corrected this nor has it directly spoken tih the discrepancy, apparently believing that its caveat when releasing the state level jobs data is sufficient to dissuade people from looking closely at the state-level data.
However, that same QCEW data confirms that there has been job loss in the US rather than job growth.
This is a particularly damning forecast by the Philly Fed, because it tracks all too well with the Employment Level reported by the BLS Household Survey (the “other” half of the monthly Employment Situation Summary), which has been showing job losses since last November.
The Household Survey has long been the BLS “dirty little non-secret secret”, as it tells an employment tale that is far less optimistic than the propaganda spun by corporate media over the Establishment Survey data.
It is in the Household Survey data that we see not only the recent shedding of jobs but also the near total cessation of job growth for much of 2022.
The Philly Fed’s benchmark revision work confirms that both of these episodes really did happen.
Taken together, what the QCEW analytics prepared by the Philly Fed each quarter confirm is that the BLS jobs data should always be viewed with extreme suspicion. The Bureau of Labor Statistics apparently has no clue how to assemble labor statistics!
Irony abounds.
Does this mean that the BLS is lying when they publish their monthly jobs report?
The parsimonious assessment suggests we should not presume malice whenever stupidity will suffice. That the Philly Fed data stands as a fairly condemnatory appraisal of the BLS data tells us the BLS is almost certainly wrong about jobs markets in the US. That data does not get us all the way to the conclusion that the BLS is intentionally wrong.
However, that data also does not rule out the possibility—nor even the probability—that the BLS is intentionally wrong. As an exercise in logic we cannot say with certainty that the BLS does lie with the Employment Situation Summary each month, but neither can we dismiss the possibility.
What we can conclude is that the Employment Situation Summary is more often than not shot through with errors.
Nor is the Philly Fed data the only indication that something is amiss with the BLS jobs reporting. As I have noted multiple times, the privately-generated ADP National Employment Report has been deviating from the Establishment Survey data quite often of late, even though it nominally is measuring the same overall jobs data. We would not expect two independent samplings to match precisely, but if the data collection is sound for all samplings then we should expect to see similar trends at the very least. When we see significant variation between the ADP report, the Household Survey, and the Establishment Survey, that alone tells us that there is something not right with how employment data is being collected, sampled, and studied.
That this data is then being used by other parts of the federal government either to set policy or reach decisions on things such as whether or not to cut the federal funds rate, when it is demonstrably flawed, is not merely maddening but quite alarming.
Nothing good can come from bad data, and the Philly Fed has reminded us yet again that the Employment Situation Summary is bad data, bad analysis, and simple bad math.
Calling what the BLS does Lou Costello Labor Math is not merely an homage to one of American vaudeville’s greatest stars. The Abbott and Costello routine of how 7 goes into 28 13 times is simply the best and most apt metaphor for the absurdities contained within the monthly jobs report.
Lou Costello would fit right in at the Bureau of Labor Statistics!
Once again, Abbott and Costello are the perfect coda to this exploration of US jobs data.
The BLS needs accountability. At this point this is malicious data manipulation. This should be a felony because of the implications it has on global markets and politics.
Any political party providing false or misleading information should be banned for life from the media and criminally prosecuted. It's sickening the Dems always lie, smear, and show hate to another candidate or political official.