Visualizing the controversial nonfarm payroll revisions

Revisions to past US jobs reports are stoking debate in the run-up to a potentially pivotal Federal Reserve meeting -- to say nothing of the presidential election. How strong has the seemingly resilient American economy really been? A recent, "annual" update resulted in the biggest revisions since 2009.

Indeed, data from the Bureau of Labor Statistics (BLS) suggests that the agency underestimated the strength of the post-pandemic job recovery in 2021-22; meanwhile, it seems that the non-farm payrolls (NFP) report significantly and frequently overshot reality during the historic monetary tightening cycle.

Our first chart contrasts these two periods to visualize these tendencies. A negative reading means later revisions resulted in a NFP figure below the initial estimate; a positive reading means revisions resulted in job gains.

The BLS releases the initial NFP figure three weeks after the end of the month in question. The first revision comes a month later, and the second a month after that. Finally, an annual revision is made each year.

To be sure, the BLS' initial readings are still valuable as an indicator of general trends. But as we can see in our second chart, since July 2023, the initial job report has proven to be overly strong in all but one month (December). And the second revision resulted in a worse figure than the first revision for every single month.

Our charts conclude on a potentially upbeat note courtesy of CEIC's proprietary nowcast. August payrolls are reported on Sept. 6; our machine learning-driven model is predicting a snap-back to a gain of 189,000, versus the print of 114,000 in July. As the market assesses whether the Fed will cut rates 25 or 50 basis points on Sept. 17-18, our projection is a reminder that American labor-market conditions remain tight, especially compared to other developed economies.

If you are a CEIC user, access the story here.