From CNN today:
“Until it is corrected, the BLS should suspend issuing the monthly job reports but keep publishing the more accurate, though less timely, quarterly data,” he told Fox Business. “Major decision-makers from Wall Street to D.C. rely on these numbers, and a lack of confidence in the data has far-reaching consequences.”
BLS does calculate a CENSUS of data, available quarterly. It’s called The Quarterly Census of Employment and Wages. This series is accurate. It’s released quarterly. So Dr. Antoni presumably suggests dispensing with the monthly jobs report, and in order to save money, can just rely on the QCEW. Here’s what the QCEW looks like right now, compared to the CES monthly series (as revised).
Figure 1: Nonfarm payroll employment, July release (bold dark blue), June release (light blue), Quarterly Census of Employment and Wages (QCEW) covered employment, n.s.a. (red). Source: BLS via FRED, BLS.
In fact, the NFP series use as input for revisions the QCEW data, as well as other administrative data. The astute observer will note that QCEW data is only reported up to December 2025.
I believe we will get Q1 QCEW data in about a month. So for data up to 2025Q1, we will get it about 6 months after the fact. Revisions to QCEW are typically very small. However, it is unclear whether reliance on QCEW data will fix the entire problem, if the aim is to measure all employment (not just that taken up by documented workers in the nonagricultural sector); see discussion referring to Goldman Sachs research here.
There is an alternative series, the Cleveland Fed early benchmark. It can be adjusted to conform to national level NFP, but it too lags. The most recent reported months take at face value the BLS series, so it wouldn’t have solved the problem that Dr. Antoni is referring to (see the most recent report).
