The BLS Jobs Report Is More of the Same BS

I don't know how much trust we want to put into the monthly survey results released by the Bureau of Labor Statistics, given that just last week, that same Bureau released a quarterly report that washed away a large chunk of what had been perceived overall job creation for 2023. 

For the readers who may have just stumbled on to my stuff with this piece, the BLS also released their Business Employment Dynamics (BED) report for Q3 2023 last week and has largely been ignored by the financial media.

What the BED report does is measure employment by quarter, and with a sizable lag. The BED report is far more detailed than the monthly survey results, and makes for a very good analytical tool, but is useful in anything close to real time. 

That said, economists know that this "BED" report is far more accurate than the monthly Establishment Survey that the financial media clings to, as the Establishment Survey is done in a hurry, polling just 670K employers. Those results are then extrapolated across the population, seasoned to taste and then reported to the public. The BED report is done quarterly and polls 9.1M employers. There is no doubt about which of these reports is more accurate.

What was released last week was the BED report for Q3 2023 (see the lag I spoke of). For that quarter, the BED report shows net job losses in the United States of 192K. Those are losses, not job creation. The Non-Farm Payrolls reports for that same quarter totaled 640K, implying robust job creation. That's an overstatement of 832K jobs. We know the 2023 NFP numbers were significantly bloated. The BLS knows this as well. Looking at Q2 2023, we see that the monthly Non-Farm Payrolls reports overstated job creation for that quarter by another 489K positions.

The overstatement in job creation over the six months that make up Q2 and Q3 2023 comes to 1.321M positions. This does show a labor market that did start to deteriorate in 2023. Your young people kept telling you this. The government and the financial media said otherwise.

Of course, the BLS will have to issue a huge revision to its 2023 job creation numbers later this year. Just as obviously, the BEA will have to revise 2023 GDP significantly lower at some point. You'll recall that GDI growth was nowhere close to GDP growth for Q2 and Q3 2023. You know that GDI and GDP are sums of basically the same economic activity from differing perspectives. The numbers are supposed to come close to equal. They did not.

Knowing All That, Happy Jobs Day!

Isn't it funny that one week after the BLS came clean on 2023 job creation, the first release of the monthly employment numbers still showed growth but were well below both recent trends and expectations.

For the month of April, the BLS reported Non-Farm Payrolls growth of 175K positions, well below the upwardly revised 315K March print and well below expectations for something in the 240K's. The Non-Farm Payrolls print is taken from the Establishment Survey (that polls just 670K employers). The Household Survey, which apparently has been far more accurate over the long-term than has been the Establishment Survey, at least according to the BLS' own data, showed April job creation of 25K.

Now, we are put in the awkward position of having to trust the rest of the information in these surveys even though we know the job creation prints had long been MacGyvered in some bizarre way that seems to constantly overstate employment and hires. I am going to take the high road here and say that there must be serious problems either in the way the data is collected or in the seasonal adjustments or in the birth/death model. Clearly, something is badly flawed.

The Nitty Gritty

As labor force only appears to have grown by 87K individuals in April, participation held steady at 62.7% as the employment to population ratio dropped from 60.3% to 60.2%. Demographically, there are some real moving parts in this report not all moving in the same direction. The unemployment rate for adult men ran higher from 3.3% to 3.6%, while the unemployment rate for adult women improved from 3.6% to 3.5%. The teenage unemployment rate (no gender recorded) improved a great deal from 12.6% to 11.7%.

Along racial demographics, the White unemployment rate moved up to 3.6% from 3.5%, while the Asian unemployment rate moved up to 2.8% from 2.5% and the Hispanic or Latino unemployment rate moved up to 4.8% from 4.5%. However, the Black or African American unemployment rate improved dramatically from 6.4% to 5.6%.

Looking into educational demographics, those with less than a high school diploma really took a hit. Their unemployment rate soared from 4.9% to an even 6%. However, those who finished high school, but never started college saw their unemployment rate drop from 4.1% to 4%. Those with some college experienced an improvement as their unemployment rate dropped to 3.3% from 3.4%. However, the group that has a bachelor's degree at a minimum saw a rise in their unemployment rate from 2.1% to 2.2%.

There was actually a reversal in this report in the trend towards part-time work. This is from the household survey, so we are talking about an economy that only created 25K jobs. In that economy, 161K more folks started working part-time for economic reasons, while 649K fewer people worked part-time for non-economic reasons.

Demand For Labor?

Looking at other measures besides job creation that reflect trends in demand for labor, hourly wage growth printed at month over month growth of 0.2%, down from 0.3% in March and below expectations for 0.3%. On a year over year basis, wage growth increased 3.9%, which was well below the 4.1% growth seen in March and also below expectations. 

Additionally, the average workweek for full-time workers (so those downgraded from F-T to P-T by their employers are excluded) dropped to 34.3 hours from 34.4. 34.3 hours equals the second shortest average workweek (to 34.2 this past January) outside of the pandemic shutdowns in early 2020, since the aftermath of the Great Financial Crisis in 2010.

Thoughts

This is an undeniably weaker monthly employment report than the BLS had released in quite some time. Equities are loving it as are debt securities. Treasury yields have come in this morning from the Six-Month T-Bill on out to the 30 Year Long Bond. Why?

Good question. 

There are two things that could get the Fed to act sooner on reducing short-term rates. One would be a sustainable victory over consumer level inflation. That's not happening anytime soon. 

The other would be an overt weakening of the US economy that impacts labor markets. Well, the data across a number of metrics has been obviously mushy of late. This report kind of takes the crown after the April reports for Dallas Fed manufacturing Index, Chicago PMI, and Consumer Confidence all visibly tumbled. Additionally, the April ISM Manufacturing Index tumbled back into contraction, and March Construction Spending fell out of bed as did Q1 Non-Farm Productivity.

In short, the economy appears to be slowing somewhat rapidly and markets are loving it because it likely keeps a rate cut on the table for 2024.

At the time of publication, Stephen Guilfoyle had no position in the securities mentioned.