The headlines in the papers and on TV on the September jobs report are that it was “disappointing” but as usual you really have to look more closely and understand the bizarre methodology behind these reports. Again, the monthly report put out by the Bureau of Labor Statistics (and please, go to the website, read the reports, they aren’t long, look at the tables, but especially, read the technical note) are derived primarily from two surveys–one of busineses and one of households. The surveys are bedeviled by low response rates and other issues. The raw data is taken and massaged, for better or worse, by a variety of “adjustments”. One that plays a substantial role in this month’s report is a seasonal adjustment. We also have constant issues with a so-called “birth-death” model, which relates to companies, not people. It basically is about assumptions regarding how many new companies start up and employ people and how many go out of business, thereby eliminating jobs. Supposedly BLS is working on getting a better data source and methodology but it is taking forever. (BLS Report)
For actual numbers this month, from the business survey we supposedly gained 29,000 jobs in September, while the household survey shows over a 400,000 job gain. The good news is about 17,000 government jobs were lost, so the private sector had a bigger gain. The increases for August and July were revised downward, for unexplained reasons. The unemployment rate ticked up a tenth of a percent, which is a rounding issue. Under the numbers is a big seasonal adjustment, as usually occurs in September because of school reopening, but this year that adjustment was much bigger than last year’s adjustment. I fail to understand the need for seasonal adjustments; better to just do year-over-year raw numbers. There really isn’t a lot notable about industry specific numbers, manufacturing has shown steady gains in the last year, as has health care. Insurance jobs are down, that may be AI related.
The labor participation rate rose, which if accurate is good and hopefully the start of a larger trend–we need more adults working. Average hourly earnings were up slightly, not really keeping pace with current inflation, that is not good. And the ADP report for the month showed much higher gains in private employment. ADP processes payrolls for companies so their numbers are likely more accurate. And please note that on a base of 163,000,000 employed persons, the numbers for any monthly change are statistically meaningless. So all in all, from what I can tell, the employment situation is pretty good and the primary concern should be how to boost lower and middle-income workers wage growth to a higher level.
