by Gary Alexander
January 23, 2024
So Why Do We Trade On It?
Next Week, January 31 falls on a Wednesday. Come 8:30 (EST), Friday morning, the Bureau of Labor Statistics (BLS) will throw out a number of the total jobs created in January – that notorious month of part-time labor lost after the Christmas shopping season. That number will quietly be adjusted by an arcane “seasonal” formula that only a PhD economist could dream up, and then the stock market will surge (or fall) based on that phony number. Fittingly, that nonsense will happen on Groundhog Day, a holiday based on a rodent, now more famous as a Bill Murray movie about an endlessly repeating day.
Why do I call that jobs number bogus? For evidence, look at the recent monthly revisions. The number of jobs created each month was revised down in 10 of the last 11 months, and these cumulative revisions were ratcheted down by a huge total of 443,000 jobs, the largest downward revisions outside of a recession year since 2002, and yet those buoyant initial monthly jobs figures tended to move the market up each first Friday of the month. Like Charlie Brown kicking at Lucy’s ever disappearing football, it makes you wonder when young Chuckster will wise up to reality. Consider this reality: Is it possible that the bean counters at the Bureau of Labor Statistics can count all the new jobs in the nation in just a few hours after the close of each month – or that their “seasonal adjustments” aren’t just political happy dust?
In looking at the latest (December) jobs report, I noticed this footnote about the wide range of revisions:
“The confidence interval for the monthly change in total non-farm employment from the establishment survey is on the order of plus or minus 130,000. Suppose the estimate of non-farm employment increases by 50,000 from one month to the next. The 90-percent confidence interval on the monthly change would range from -80,000 to +180,000 (50,000 +/- 130,000).”
That explains the wide variance in future revisions. You simply can’t count on an accurate figure the first or second day – or really the first week – after the end of the month, especially when based on interviews with real human beings, who have their own agenda in answering questions from government agents.
Inside the Official (versus Real) December 2023 Jobs Report
Here’s how to dissect a monthly jobs report. In the latest jobs report, the “headline” number was a +216,000 monthly “non-farm payroll” increase in December. But let’s look at some details for clarity.
• If you take out the 52,000 bump in government payrolls, the gain is whittled down to 164,000. The BLS December report said, “Government added an average of 56,000 jobs per month in 2023, more than double the average monthly gain of 23,000 in 2022.” (Those aren’t productive private-sector jobs).
• The October and November payrolls were revised down by 71,000, so make that a net +93,000 gain.
• There are still 8.8 million job openings (see the chart below). Also, Fed Chair Jerome Powell says that there are 2.5 million more unemployed workers than job offerings (mostly because 49% of small business owners said last month that there were no qualified applicants for their job openings).
Graphs are for illustrative and discussion purposes only. Please read important disclosures at the end of this commentary.
- The unemployment rate was only 3.7% in December, but 845,000 workers disappeared from the labor force in December, since the labor force participation rate declined by 0.3 points to 62.5%, the biggest monthly drop in almost three years (since January 2021). The BLS only counts people who say they are “looking for work” as unemployed persons. Those who have dropped out are not counted as jobless.
This brings up the subject of how jobs (and the jobless) are counted. The BLS has two surveys each month – one being the household survey, to determine how many people are working or are looking for work, and the other is the establishment survey, to ask businesses how many people work there.
This gets into human nature and how people answer surveys: What is a job, and how (or why) would a clever person answer a complex question with a yes-no answer? In the household survey, you may be working part-time for money (and big-time in the underground economy for untaxed money). Would you admit to two jobs? Probably not. And how would a guy like me answer that question? I’m 78 and I work part-time writing about economics for Louis Navellier. Is that a job? You tell me. It sounds like fun.
When it comes to the business survey, a small business owner might employ a handful of part-time migrants. Which are workers, and which are transients? A government phone surveyor calls in: “How many people work there?” The business owner takes out a pencil and tallies up names. How many are taking trips as semi-sabbaticals? In this age of fun-employment Generation Z baristas, it’s hard to tell.
With that in mind, here’s how the BLS finesses the middle ground of non-working non-job seekers in their latest “Household Survey” for the December 2023 jobs report, released Friday, January 6, 2024:
“The number of persons not in the labor force who currently want a job edged up to 5.7 million in December and was up by 514,000 over the year. These individuals were not counted as unemployed because they were not actively looking for work during the 4 weeks preceding the survey or were unavailable to take a job. Among those not in the labor force who wanted a job, the number of persons marginally attached to the labor force changed little at 1.6 million in December but was up by 306,000 over the year. These individuals wanted and were available for work and had looked for a job some-time in the prior 12 months but had not looked for work in the 4 weeks preceding the survey. The number of discouraged workers, a subset of the marginally attached who believed that no jobs were available for them, at 346,000, was little changed in December and over the year.”
Confused? Me too. There is a lot of middle ground between work and no work. In contrast, the ADP jobs survey, published on the Wednesday before the Friday BLS report, is based on actual payroll reports made by the company in charge of processing actual payrolls, Automatic Data Processing (ADP).
I would trust ADP data before placing stock market bets based on premature BLS jobs data. ADP has the added virtue of only covering private sector, not government jobs, which add zero to the economy.