Series based on CPS adjusted to NFP well below the official NFP, first reference.
Figure 1: CES nonfarm payrolls (blue), CES hours (red), QCEW adjusted by geometric moving average by author (light blue), Business Employment Dynamics (pink), and Philadelphia Fed early benchmark (light green), all in logarithms, 2023M05=0. BED series normalized to Q4 2019 NFP level. Philadelphia Fed series is official CES adjusted by the ratio of the sum of early benchmark states to the ratio of the sum of CES states. Source: BLS via FRED, BLS, QCEW/BLS, BED/BLS, Philadelphia Federal Reserve and the author’s calculations.
Given the distance between the QCEW and BED series and the CES-based NFPs, one might legitimately wonder whether the official NFPs overestimate employment (I would not place much weight on the CPS series adjusted for the NFP concept, given the uncertainty surrounding the reliability of the population controls used to construct civilian employment). As noted earlier, however, the fact that private NFPs are growing less rapidly than the ADP measure (which is not subject to criticisms regarding the birth/death of firms model) gives me some measure of confidence.
Figure 2: Private nonfarm payroll employment from BLS (blue) and ADP (beige), in logarithms 2023M03=0. The numbers on the right are the cumulative increases since 2023M03, in logarithms. Source: BLS, ADP via FRED and author’s calculations.
