CES measures show faster growth than others.
Figure 1: Nonfarm payroll employment (blue), first benchmark (tan), CPS estimate adjusted for NFP assumption (green), total QCEW covered by author’s seasonally adjusted employment using geometric motion (sky blue), Business Employment Dynamics cumulative growth in 2019Q4 NFP (pink), and aggregate hours (red), all in log, 2023M05=0. Source: BLS, Philadelphia Fed, and author’s statistics.
Other details: NFP is total workers, production hours and non-supervisory workers, QCEW is the total number of workers covered which is different from that covered by the CES survey. The House Survey (CPS) series is heavily influenced by the use of new population controls, with data through January 2024. CBO has a higher population estimate, which would mean higher employment; see discussion by Edelberg and Watson (2024).
NFP tends to rise faster than other measures, and may be affected by a birth/death method that overestimates the number of firms. However, private NFP comparisons from the BLS and ADP surveys contradict that definition.
Figure 2: Private nonfarm employment from CES (teal), from ADP (purple), from Source: BLS, ADP via FRED, and author’s calculations.
Note that the ADP-Stanford developed series is based on administrative data, and is not designed to predict the BLS series. Not only has the ADP-Stanford series increased over the past year, the growth rates over the past three months are comparable.
Source link