CES measurements indicate faster growth than others.
Figure 1: Non-agricultural wage employment (blue), first benchmark (beige), CPS measure adjusted to the NFP concept (green), total covered employment QCEW seasonally adjusted by the author using the geometric moving average (sky blue), net growth in the dynamics of cumulative company employment on the NFP for the 4th quarter of 2019 (pink) and cumulative hours (red), all in the newspapers, 2023M05=0. Source: BLS, Philadelphia Fed and author’s calculations.
Some details: NFP is for all workers, hours are for production workers and non-supervisory workers, QCEW represents total workers covered which differs from those covered by the CES survey. The Household Survey (CPS) series is greatly affected by the application of new population controls, with data from January 2024. The CBO has a higher population estimate, which would imply higher employment; see discussion in Edelberg and Watson (2024).
NFP increases faster than other measures and could be affected by the birth/death model which overestimates the number of firms. However, the coincidence between the BLS survey’s private NFP and ADP argues against this explanation.
Figure 2: Private non-agricultural salaried employment according to CES (teal), ADP (purple), its Source: BLS, ADP via FRED, and author’s calculations.
Note that the revamped ADP-Stanford series is based on administrative data and is no longer constructed to forecast the BLS series. Not only has the ADP-Stanford series increased over the past year, but the growth rates over the past three months are comparable.
