Private nonfarm payroll employment vs. total nonfarm payroll employment

I am not exactly sure when the transition occurred, but at least in the past few months people have started to ignore the headline payroll employment number, AKA total nonfarm payroll employment and focus on the private nonfarm payroll employment number instead. One possible reason is that ADP offers only a forecast for private employment. Personally, I have not seen any discussion justifying the shift. But, now that I see the Federal Reserve FOMC minutes refer to "private nonfarm payroll employment", I will probably follow suit.

Usually the nuanced distinction does not change the overall picture, but since private nonfarm payroll employment declined in December and has now declined for four consecutive months, as compared to a gain for total nonfarm payroll employment in December and thress consecutive months of decline, the distinction suddenly makes a significant difference for people trying to handicap whether we are in a recession and when the recession started. Clearly the "recession started in November" camp are keying off of that private payroll employment decline from November to December. Fair enough.

For me, I will switch to focusing on private nonfarm payroll employment simply because that is that the Federal Reserve FOMC is focusing on.

One lingering issue is which measure the NBER Business Cycle Dating Committee uses for assigning the starting month of a recession. Most recently, they simply refer to "employment", but they have referred simply to "payroll employment." I would note that NBER president Martin Feldstein, who is also a member of the dating committee, recently suggested the recession started from a peak in December or January, suggesting that at least he personally was not dependent on using the private nonfarm payroll employment number.

Overall, I will still focus on the headline number, but I will look at the private payroll number as well. I see no compelling reason to ignore a significant segment of the working population. In March, total nonfarm payroll employment was 137.846 million while private nonfarm payroll employment was 115.459 million, which ignores over 22 million workers.

Incidentally, people like me do not even get counted as part of payroll employment at all since we are self-employed, but we do get counted in the household survey.

-- Jack Krupansky