The Job Market Has Reached Stall Speed

Given the 8% decline in labor demand since March, we are forecasting that June job gains will be revised to -25,000 assuming, of course, that we even see downward revisions ever again.

With the firing of BLS Commissioner Erika McEntarfer and the likely appointment and confirmation of E.J. Antoni, July’s jobs report marks the high-water mark, from here on out, for trustworthiness in U.S. jobs data. Even as chronic underfunding and rapidly declining response rates in the covid era degraded efficacy, no one questioned the integrity of the agency and its commitment to trying to produce the most accurate reports possible under increasingly difficult circumstances. Alas, no longer.

And what a report it was that marks the end of reliable BLS data. More worrisome than the underwhelming job gains last month, gains that we expect will be (or at least should be) revised downward further in subsequent months, were the gigantic downward revision by a combined 258,000 jobs for May and June.

As everyone knows, the difference between initial BLS data and the revised data has been growing for years, but with the July jobs report, the 2025 delta jumped from 21% to 47%, rendering the initial BLS release, once and for all, as completely worthless as a gauge of what is actually happening in the U.S. labor market.

With those revisions, average monthly job gains have dropped to 85,000 for the year and 35,000 since the gong show in the Rose Garden in April.

And therein lies the crux of the catastrophic report that led to the Commissioner’s firing. What had appeared to be, on July 31st, a surprisingly resilient economy motoring steadily along, largely immune to the chaos being inflicted upon it, slammed into the wall of reality on August 1st.

With the report, the weakness of the job market has, at long last, become too glaring for anyone to miss - anemic job growth trending toward the first negative print in 5 years, gains concentrated in a single sector (healthcare), unemployment rising to 4.2%, declining labor supply and falling labor force participation rate that masks even higher unemployment, a white collar recession, especially for recent college grads, massive job losses in Federal government with federal funding cuts still rippling out to state and local governments, long-term unemployment rising, etc., etc.

As TS Lombard’s Chief Economist Freya Beamish writes:

Just 73K were added to non-farm payrolls in July, while June was revised done to a mere 14K, below consensus of 110K and bringing last month's print down below the consensus for that time too. This follows the pattern of downward revisions that we should all be accustomed to by now.

Thematically, this is an economy understandably on pause. Last month's print pre-revision was around 50% government. The revised data show that nobody was hiring across public and private sector. The private sector, however, has at least begun hiring again in July, with private payrolls rising 83K, though that could be revised down, and the government in modest contraction, with uncertainty compounding direct DOGE effects.

We expect employment growth to remain below trend, with more close shaves, but this is broadly an economy on pause rather than in recession at this stage.

While we might replace the phrase ‘an economy on pause’ with ‘an economy that has slammed headfirst into the windshield,’ we share the sentiment that the U.S. economy has, very suddenly and quite obviously, come to grinding halt. But as we’ve made very clear all year, none of this is surprising in the least.

Stagflation was our baseline forecast from the start of the year, and even before but especially after our high quality, real-time labor demand data sourced daily directly from employer websites globally started plummeting in April, our conviction around where all this is headed has grown stronger and stronger.

With the most recent CPI report showing that core inflation has risen to 3.1% and this week’s PPI report showing that wholesale inflation has jumped to 3.3%, not only are we firmly in stagflation territory, things are undoubtedly going to get worse and worse as the the impact of tariffs continue rippling through the economy and the Trump tsunami decimates everything in its path.

And like a tsunami whose delusory slow motion belies its catastrophic force, the destruction here is similarly taking time to roll through the economy and into the data. As Neel Kashkari said on CNBC, “These tariff shocks are unlike anything that we’ve seen in 100 years and it’s taken a while for businesses to try to process it.” They clearly have and, adding to tariffs the ‘Hell of Blunders, Madness, and Deception,’ with no end in sight, it’s about time the impact has finally started showing up in the data.

As the Washington Post noted, “Together, the economic data paints a picture of an economy edging into stagflation territory — slowing growth coupled with stubborn inflation.”

Jared Bernstein and Ryan Cummings were, appropriately, far less equivocal in a recent NYT op-ed:

Mr. Trump’s tariffs are now clearly fueling inflation, particularly in goods such as home appliances, cars and food. In the first six months of the year, real (that is, inflation-adjusted) consumer spending, the main driver behind business cycles and robust economic expansion, barely grew, after rising 3 percent last year. G.D.P. growth slowed by about half, to 1.2 percent this year from 2.5 percent last year. When overall growth falls that sharply, the labor market tends to follow, which is precisely what happened: Job growth, at 35,000 per month on average between May and July, is dangerously close to stall speed.

Stall speed is the perfect description, one that Jan Hatzius, the now even more infamous Chief Economist for Goldman Sachs, used to describe the broader economy in an earlier note published immediately following July’s jobs report:

Friday’s payrolls report reinforced our view that US growth is running below potential and near stall speed—a pace below which the labor market weakens in a self-reinforcing fashion.

Given the declines we’ve seen in labor demand over the past 4 months and the forward-looking nature of accurate, real-time job openings data, we’d dispense with phrases such as ‘near’ or ‘dangerously close.’ We have reached stall speed.

As we wrote last week, with the decline in the millions and millions of job openings we source daily directly from tens and tens of thousands of corporate and employer career portals around the world, labor demand has dropped 8% since March and 13% since the election.

Of special note is the fact that since April 2nd, LinkUp data indicates that labor demand has dropped 7.9% while JOLTS data indicates that labor demand has increased 3.3%. We’ll see if that deviation persists when the BLS releases its July data in September, but if it does, that would mark the 4th consecutive year with material and persistent deviations between LinkUp and BLS data.

Given the very strong correlation between our data and what is actually happening in the job market - always the key driver underpinning the entire economy (see blue line in the chart above in 2023 and again in 2024 and our early and vocal calls (ad nauseam) for the soft landing), we’ll take our data as the better indicator of where the job market is heading and what this economy is going to look like in the coming months.

Since 2022 when the Fed began hiking rates, our job forecasting tracking error against final revised BLS data is just -5%, while consensus estimates is -15%, ADP is -26%, and BLS initial release is +11%. (As an aside, even with our Initial Release Forecast miss for July, our new initial release forecasting model that we launched this year has been correct in 4 of 7 months through July).

So on that note, dusting off our old model to forecast final BLS revised numbers, we are forecasting that June job gains will be revised down to -25,000 and that July job gains will be revised down to 50,000.

If we’re correct, the critical question will then be whether or not the Perkins Rule applies to BLS revisions or just an initial release. We suspect we’ll never know given that the question, in turn, raises another around whether or not we’ll ever again even see a negative job growth number from the BLS. We’re assuming, with good reason, that everything from here on out will be BIG AND BEAUTIFUL.