HALF FULL OR HALF EMPTY: A BRIEF NOTE ON THE US JOBS REPORT FOR FEBRUARY 2023

Much if not most of the focus on the announcement of the data from the U.S. Bureau of Labor Statistics (BLS) today for data from February 2023 has been on the gain in the establishment survey on a seasonally adjusted basis.  The reported gain of 311,000 jobs exceeded forecasts made prior to the announcement and is considered to show labor market strength.  However, considering at least some other measures of employment and unemployment presents a different picture.  It reminds me of the old cliche about whether a glass is 'half full or half empty.'

If my understanding is correct, then the above job gain of more than 300,000 jobs from the establishment survey in part reflects the results from a model that makes assumptions about the number of new workplaces added and the number of workplaces that cease operation.  By contrast, the household survey uses a sample of  households to project the total number of people in the labor force, not in the labor force, working, and unemployed.  The household survey by the BLS indicated that the unemployment rate in the U.S. increased from 3.4 per cent to 3.6 per cent in February 2023.  This could largely reflect an increase in the labor force participation rate last month.  I think I heard that on CNBC this (Friday) morning that the labor force participation rate has now returned to about where it was before the start of the COVID-19 pandemic.  Even so, the modest increase in the unemployment rate may be viewed less favorably that the aforementioned jobs gain.

Using the household survey seasonally adjusted, the BLS found a job gain of only about 177,000 jobs added in the U.S. last month as shown in Table A, only somewhat more than half of the jobs gain found by the establishment survey.  But, when looking at the household survey in Table A-1 from the BLS with data not seasonally adjusted, I estimate  employment growth of over one million jobs!  Is some of this growth due to warmer temperatures and sampling issues?  If not, then what is the cause of this discrepancy?  Readers may want to note that based on U.S. BLS data from the FRED website of the St. Louis Federal Reserve Bank, (a) the net gain in jobs from July of 2022 through February 2023 for the household survey not seasonally adjusted was very roughly 650,000 (which could indicate relatively modest monthly average gains) and (b) before February 2023 but after July 2022, the only other month with a higher count of jobs from this survey than in July 2022 was in October of 2022.

I think I heard on CNBC this morning that wage gains recently in the U.S. on an annuallized basis were up by about three per cent.  That's quite a bit lower than recent year over year estimates of U.S. inflation.  Why are many paychecks failing to keep up with inflation?  Combining this with the observation that unadjusted household employment in November 2022, December 2022, and January 2023 was lower than in October 2022 raises questions about how strong the U.S. jobs market really is.  On top of that, with banking problems coming to light, this should give the Federal Open Market Committee of the Federal Reserve a lot to consider when deciding on its target for the federal funds rate soon. 

Note:  Subsequent data revisions may change the above analysis. Please accept my apology for any misstatements, inconvenience, or confusion.

REFERENCES

U.S. Bureau of Labor Statistics, Employment Level [LNU02000000], retrieved from FRED, Federal Reserve Bank of St. Louis; https://fred.stlouisfed.org/series/LNU02000000, March 10, 2023.  However, data were accessed from https://fred.stlouisfed.org/data/LNU02000000.txt

https://www.bls.gov/news.release/empsit.nr0.htm

https://www.bls.gov/news.release/empsit.a.htm

https://www.bls.gov/news.release/empsit.t01.htm