Drama, folks. High drama.

And I don’t mean Israel venturing further down the road to becoming a rogue state on Tuesday.

Rather, I mean an enormous downside revision to US payrolls growth in the BLS’s annual benchmarking exercise.

I’ve previewed this oh-so-crucial macro event at least three times by now, including here on Monday, so I won’t recapitulate other than to say that revisions to the BLS jobs count are a (highly) contentious topic, and the short-staffed bureau’s preliminary attempt at the CES revision had the potential to tip the scales in favor an upsized, 50bps rate cut from the Fed next week.

The figure below shows you the print: -911,000. Headline US jobs growth in the year to March was 911,000 weaker than initially reported.

That’s an even larger preliminary revision than the BLS tallied in 2009, which is to say larger than the recount which captured the worst months of the financial crisis.

For context, this revision process typically sees payrolls revised lower by 0.2% of total nonfarm employment. This revision amounted to triple that, at -0.6%. The implication is that average monthly jobs growth in America in the year to March was just half of the pre-revision count.

“Preliminary research, which is not comprehensive and is subject to updates in QCEW data, indicates that the primary contributors to the overestimation of employment growth are likely the result of two sources — response error and nonresponse error,” the BLS said Tuesday, in perfunctory color accompanying the release. The bureau continued as follows,

First, businesses reported less employment to the QCEW than they reported to the CES survey (response error). Second, businesses who were selected for the CES survey but did not respond reported less employment to the QCEW than those businesses who did respond to the CES survey (nonresponse error). Estimates of other errors, such as the forecast error from the net birth-death model, are not available at this time.

In keeping with standard operating procedure, the final version of this exercise will be released in February, with the January 2026 jobs report.

Needless to say, this raises the odds of a 50bps rate cut next week, although it’s hard to say by how much. Traders will probably price in something like even odds of an upsized move. Given the recent history of his public remarks, Chris Waller’s probably leaning 50 now, maybe Miki Bowman too. Stephen Miran will almost surely vote for a 50bps cut in his Board debut.

As discussed in the linked article above, the Fed’s in their self-imposed communications blackout this week, which means if they’re going to go with a half-point cut and they don’t want to leave markets guessing, they’ll need to get on the phone (or on WhatsApp or Signal) to Nick Timiraos at the Journal so he can tip the larger move.

That may have to wait until Thursday, though, or even Friday. Some (most) officials will still want to see the MoM core CPI print for August before supporting a big rate cut given that the YoY pace of underlying price growth is still running a full point warmer than target.

The sheer size of this CES revision will draw a lot of scrutiny from the BLS’s critics inside the Trump administration. Expect this print to be wielded as agitprop. Whether the prevalence of downside QCEW revisions since 2015 is due to the birth-death model, uncaptured (no ICE joke intended) immigrant workers or both is largely irrelevant for The White House’s propaganda purposes.

It doesn’t help that, as Bloomberg noted on Tuesday morning, “a third of BLS leadership jobs are now vacant.”