Strong Jobs Report is Suspect


Let's investigate the growing variance between jobs and employment, and the U-3 and U-6 unemployment rates.

Jobs Establishment Survey

  • Jobs November: 261,000
  • Jobs December: 184,000
  • Jobs January: 273,000.
  • Jobs February: 273,000.
  • Jobs Nov-Feb: 991,000

Employment Household Survey

  • Employment November -8,000
  • Employment December +267,000
  • Employment January -89,000
  • Employment February +45,000
  • Employment Nov Feb: +215,000

U3 Baseline Unemployment

  • U3 November: 3.5%
  • U3 December: 3.5%
  • U3 January: 3.6%
  • U3 February: 3.5%
  • U3 Net: 0.0%

U6 Unemployment

  • U3 November: 6.8%
  • U6 December: 6.7%
  • U6 February: 7.0%
  • U6 Net: -0.2%

What's Happening?

  1. In December and January, warm weather padded construction jobs.
  2. In January and February, government hiring padded jobs.
  3. In both cases, seasonal adjustments compounded the improvement.


  • We have seen these kinds of divergences before and they converge over times.
  • However, the reports often converge due to revisions in one direction or another.

What is more likely in this case?

Given the huge seasonal adjustments and escalating coronavirus threat, I suggest the next set of revisions will be a convergence towards lower jobs and increasing unemployment rather than the other way around.

Mike "Mish" Shedlock

Comments (5)
No. 1-5

I have as much faith in these numbers as the Chinese gov't coronovirus numbers. Our gov't is going to do everything possible to calm the markets. Similar to the Chinese doing everything to convince people to go back to work.


The 10 year is now at 0.75%. I expect more announcements from the Fed for loan facilities for corporations to stave off layoffs. The Fed will also start buying stocks and other assets in order to save everyone. This is not that different from the environment we've been living in since the global financial crisis. This is just MMT by another name.

Ted R
Ted R

Of course it is suspect. The report is based mostly on guessing. It is an estimate. The report also does not count long term unemployment. The report is a guesstimate, not a hard, actual counting of new job creation. That would be impossible. The report should be interpreted for what is is, purely an estimate based on guessing. I think this should be obvious.


All govt jobs,prison guards,parole officers,prison custodians ,social workers,cops, wardens,anything that has to do with the prison industrial complex is absolutely booming!

Global Economics