The preliminary benchmark estimate points to a massive 911,000 downward revision to the level of non-farm payroll employment in March 2025 and, based on the Quarterly Census of Employment and Wages (QCEW) data, the revision looks like it will be concentrated in the latter six months of the period. Given that the birth-death model adjustment to the payroll data was chronically overestimating employment gains last year, there’s a good chance that distortion has persisted this year too. We suspect that when these revisions are formally incorporated into the data, which won’t be until early next year, the payroll employment data are likely to match the alternative household employment series, which shows an outright decline in employment since April this year.
Subscriber content
Read this in full, free
Start a free trial and use one of your 10 free starter credits to unlock this piece.