BLS Job Revisions Show Every Job Report In 2021 Was Total Garbage

Submitted by Mish Shedlock of MishTalk

The BLS made huge revisions in the number of monthly job dating all the way back to 2017.

BLS 2021 job level revisions by month, data From the BLS, chart by Mish

Along with its monthly jobs report the BLS produced annual revisions, emphasis mine. 

Effective with data for January 2022, updated population estimates were incorporated into the household survey. 

In accordance with usual practice, BLS will not revise the official household survey estimates for December 2021 and earlier months. 

The adjustments increased the estimated size of the civilian noninstitutional population in December by 973,000, the civilian labor force by 1,530,000, employment by 1,471,000, and unemployment by 59,000. The number of persons not in the labor force decreased by 557,000. Although the total unemployment rate was unaffected, the employment-population ratio and labor force participation rate were each increased by 0.3 percentage point. This was mostly due to an increase in the size of the population in age groups that participate in the labor force at high rates (those ages 35 to 64) and a large decrease in the size of the population age 65 and older, which participates at a low rate. 

Data users are cautioned that these annual population adjustments can affect the comparability of household data series over time. Additional information on the population adjustments and their effect on national labor force estimates is available at www.bls.gov/web/empsit/cps-pop-control-adjustments.pdf.

Job Level Revision Notes

• A positive number in the lead chart indicates the BLS underestimated jobs in the given month 
• A negative number in the lead chart indicates the BLS overestimated jobs in the given month 
• The BLS underestimated jobs for the first 5 months of 2021 plus December.
• The BLS overestimated jobs every month between June and November, 6 consecutive months. 
• The sum of overestimates totals 2,507,000
• The sum of underestimates total 1,890,000
• The final revision in December was up by 211,000 

Every Job Report in 2021 Was Total Garbage

The BLS will counter that the error rate is tiny. It is, if you divide by the total number of jobs. 

For example, the total level of jobs in December of 2021 was initially reported as 148,951,000 and revised to 149,162,000. 

The revision was only (211,000 / 148,951,000) * 100 = 0.142%

Yet, in monthly gain-loss report, which every economist watches, the revisions are not only huge, but ongoing.

The first 5 months of 2021 were hugely understated, followed by 6 months of overstatements, then ending with a 211,000 understatement. 

Let that sink in. The jobs market was weaker than reported every month between June and November. 

Add it all up and there is only one proper conclusion: The monthly jobs reports are garbage but ending with a December and January surprise. 

Published BLS Data is Poisoned 

Just got off the phone with Lacy Hunt who said “The BLS household numbers are now poisoned. I have never seen such huge revisions.”

— Mike “Mish” Shedlock (@MishGEA) February 4, 2022

Revisions needed or historical Y-o-Y comparisons are nonsense.

The BLS left the Household Survey intact

“In accordance with usual practice, BLS will not revise the official household survey estimates
for December 2021 and earlier months”

Lacy Hunt: published data “poisoned”

— Mike “Mish” Shedlock (@MishGEA) February 4, 2022

Nonfarm Payrolls Were Another Big Upside Surprise to Economists, What Happened?

What happened was massive annual revisions. Those revisions do not change my recession opinion, my economic outlook, or my stock market expectations. 

For a discussion of the surprise, please see Nonfarm Payrolls Were Another Big Upside Surprise to Economists, What Happened?

For recession discussion, please see With Nearly Everyone Looking the Other Way, It’s Time to Discuss Recession.

For stock market discussion, please see FAANG Roulette Continues With Huge Swings in the Price of Amazon.