Tej Parikh in FT inquires.

He presents growth rates of key indicators followed by the NBER’s Business Cycle Dating Committee (BCDC). Below I show the same indicators in levels (where I’ve replaced the official NFP series with the implied preliminary benchmark NFP using Wells Fargo estimates).

Figure 1: Nonfarm Payroll – estimated preliminary benchmark revision (bold blue), NFP official (thin blue), civilian employment with smoothed population controls (orange), industrial production (red), personal income excluding current transfers in Ch.2017$ (bold light green), manufacturing and trade industry sales in Ch.2017$ (black), and monthly GDP in Ch.2017$ (pink), all log normalized to 2025M04=0. Estimated preliminary benchmark is based on midpoint of Wells Fargo range of downward revision. Source: BLS via FRED, Federal Reserve, BEA 2025Q2 second release, S&P Global Market Insights (nee Macroeconomic Advisers, IHS Markit) (9/2/2025 release), and author’s calculations. 

I normalize on April 2025 because that’s the peak in civilian employment, and there’s some evidence that civilian employment peaks before NFP in real time, just before recessions.

It’s clear that nonfarm payroll employment growth has slowed to a crawl, a slowdown more pronounced if one used the official series. We have the official preliminary benchmark revision on Tuesday (9/9), and the Philadelphia Fed early benchmark on 9/19.

The evolution of civilian employment is shown below.

Figure 2: Civilian employment, smoothed population controls experimental series (orange), and 3 month centered moving average (dark orange), in 000’s, s.a. Source: BLS and author’s calculations.

Since the household survey based employment series is more volatile than the NFP, it makes sense to take a moving average. This transformation confirms that civilian employment is past recent peak…