Manufacturing Employment, Measured Three Ways
Econbrowser lines up several measures of manufacturing employment and finds them pointing in different directions. ADP's series suggests continued deterioration, while the QCEW measure — which runs through March — fell even as official employment rose. The chart plots manufacturing employment from CES, ADP and QCEW, plus an implied preliminary benchmark for March; the QCEW line is the author's own seasonal adjustment, done with X-13 in logs, drawing on BLS, ADP via FRED and the author's calculations. The lesson is that a phrase like “manufacturing employment” is not one number. Pick up a different series and the recent past looks like decline or like growth, and the measures don't all cover the same months — the QCEW reading here stops in March while ADP's points to ongoing deterioration.
AI-generated discussion · Different perspectives
One line on that chart isn't lifted straight from a statistical release: the QCEW series is the author's own seasonal adjustment using X-13 in logs, so part of the gap could be a difference in method rather than in the industry.
The more basic issue is coverage. The QCEW reading is the one labelled as running only through March, so comparing it with the other lines spans different periods.
Which still leaves the question the chart raises: is manufacturing employment falling, flat or rising? Three measures, three answers, and the reader picks a favourite.