Revisions in the January employment report may reveal that job growth stalled over the past year, even without a recession.
The economic news this week has been foreboding. Kevin Hassett, a leading economist at the White House, told viewers on CNBC, “one shouldn’t panic,” and advisor Peter Navarro on Fox urged us to “revise our expectations down significantly.” What’s going on? If the data revisions play out as expected today, the United States may have just had a year with essentially no net job growth, yet without being in a recession.
The data argue against simple stories, especially partisan ones. The low-hire, low-fire labor market, paired with solidly expanding economic activity, is unusual and raises tough questions. We may not be careening toward a recession, but we are failing people entering the workforce, and workers’ wages are not broadly sharing in the faster productivity growth.
Zero. Zip. Nada.
The January employment report is special. As with every month, it estimates payroll employment for the current month and revises the two prior months, using a large establishment survey.
But it also includes once-a-year changes: seasonal factors are re-estimated for the prior five years, and most importantly, the level of payroll employment in the prior March is benchmarked to administrative data from nearly all US establishments. Finally, the report updates the birth-death model adjustment that affects estimates in the months following the benchmark.
There are many moving pieces in the January report, and this year most will likely point in one direction: down.1 Large revisions are not a sign of mistakes at the Bureau of Labor Statistics. They reflect genuine changes in the labor market that are difficult to measure in real time.
Today’s data will reinforce and expand a familiar story: job creation hit a wall last year. In the revised data, outright monthly declines in will likely be more common, and net job growth in 2025 may have been close to zero. Also, the weakness began in the summer of 2024, so it’s not a simple story about the current White House.
In the revised data, the summer and fall of 2024 will show notably more weakness in payroll employment. July 2024 was also when the rise in the triggered the Sahm rule recession indicator. At the time, the unemployment rate appeared out of step with payrolls, but that gap largely reflected the long lags in measuring employment.
The unemployment rate remains a critical indicator of labor market conditions and will not be revised in the January report. I expect it to hold steady at 4.4% in January.
The unemployment rate has been rising gradually—by nearly one percentage point over the past three years. Labor demand has not kept pace with labor supply: the unemployment rate is rising, but at a much slower rate than the deceleration in job creation.
The sharp reduction in immigration, beginning in the summer of 2024 under President Biden and intensifying in early 2025 under President Trump, has been a key, though not exclusive, factor slowing growth. Recent shifts in the civilian population are largely attributable to immigration and are unusually large by historical standards. However, reduced labor supply alone cannot explain the weakness in hiring.
Rising unemployment and slowing wage growth point to weakening labor demand. Here is Fed Governor Chris Waller, who dissented at the last FOMC meeting in favor of a , has described the labor market in similar terms
Payroll gains in 2025 were very weak. Compared to the prior ten year average of about 1.9 million jobs created per year, payrolls increased just under 600,000 for 2025. And, last year’s data will be revised downward soon to likely show that there was virtually no growth in payroll employment in 2025. Zero. Zip. Nada.
Let this sink in for a moment—zero job growth versus an average of almost 2 million for the 10 years prior to 2025. This does not remotely look like a healthy labor market. While lower labor supply was surely a factor, it also indicates considerable weakness in labor demand. Employers are reluctant to fire workers, but also very reluctant to hire. I have heard in multiple outreach meetings of planned layoffs in 2026. This indicates to me that there is considerable doubt about future employment growth and suggests that a substantial deterioration in the labor market is a significant risk.
I am less concerned about an imminent recession than Waller, but I have concerns about the labor market. The core problem is low hiring. Layoffs remain low, but avoiding a recession is not ‘good enough’ as an outcome.
I expand on these themes in my Bloomberg interview on Friday, where I discuss my advice to Kevin Warsh and the state of the labor market.
What Does It Mean?
Many readers will interpret the red ink in the January employment report as a warning sign of recession. It’s true that outright payroll declines are rare outside of downturns. But I do not see a recession brewing. Instead, the data point to something more unusual—and harder to diagnose.
The gradual rise in the unemployment rate we’ve seen over the past few years is unprecedented in postwar business cycles. The unemployment rate spiked rapidly during the pandemic, then fell just as quickly back to its pre-recession level. Since then, it has drifted higher without the economy tipping into recession.
What’s going on? One possibility is that the pandemic downturn does not fit the standard definition of a recession. Another is that this is simply what a soft landing looks like—slowing employment growth without a sharp contraction in activity.
In an interview with Fortune, I discussed my concerns about the labor market.
Many economists have been eyeing the “knife-edge” in the labor market. They are watching the “breakeven number” (the job creation figure needed to stop unemployment from climbing) grind lower and lower, offset by significant immigration, which has reduced labor supply.
Sahm isn’t so concerned by the month-to-month shifts. Businesses are finding a steadier footing amid tariffs, according to the Fed’s first Beige Book of the year, meaning employers’ low-fire, low-hire approach is no longer driven by fear. Sahm’s concern is longer term: What it means for people looking for work but who can’t find a job, and whether they’ll be ignored by policymakers who are only alert for the technical numbers that signal a downturn.
“I get concerned when I hear, ‘Well, we don’t have layoffs, so we don’t have a recession,’” Sahm told Fortune in an exclusive interview. “But you do have a very low hiring rate. It might not be an aggregate event; it might not be a broad-based contraction like we see in a recession. However, it certainly has real implications for workers entering the labor market.
“Something’s happening here,” Sahm adds. “It’s clearly bad for people looking for work, but we can’t just have this, ‘Oh, if we avoid a recession, all is good.’ It could be that we’re dealing with much more structural shifts, and those aren’t just hard to forecast, they’re hard to assess in the moment because those structural shifts can be very slow.”
One reason I have long focused on recession prevention is the lasting damage downturns inflict on young workers. Today, conditions for new graduates entering the labor market are not far from recessionary—even without an official downturn. That should concern policymakers. Understanding why hiring has slowed is the first step; determining the right policy response is the harder one.
The data also isn’t illustrating an economy in need of fiscal stimulus to generate activity—though that’s what it’s getting this year anyway in the form of the One Big Beautiful Bill Act. Analysts are also banking on interest rate cuts from a more dovish Fed chairman, but again Sahm feels this won’t kick-start sluggish hiring: Sahm described the behavior as how a government might “traditionally” stimulate a weakening economy, “kind of [a] front-end recession response.”
“But against the backdrop, as best we know from the data, business activity looks pretty okay, consumer activity looks okay. I’m concerned that stimulating more demand isn’t what’s holding back hiring—there’s something else.”
Sahm’s own creation isn’t demanding action: Currently, the recession indicator is sitting at a mild 0.35. She warned policymakers against relying too heavily on the tool in the current cycle, saying their attention should be focused—“maybe even more so”—on the labor market because “it doesn’t hold the typical pattern, which means our typical tools to fight [it] like a recession may not be the right ones.”
In Closing.
Even with stalled hiring, I remain cautiously optimistic about the labor market. But avoiding a recession is not the same as having a labor market firing on all cylinders. Until we understand why hiring has slowed so markedly, policymakers should be cautious about assuming the economy is on a solid footing.
***
1 We already know that the benchmark revision will be substantially negative. The preliminary estimate, published by the BLS in September, was -911,000 (-0.6% of total employment) for March 2025. Under standard practice, the level adjustment is spread out evenly over the prior twelve months, so -76,000 revision per month from April 2024 to March 2025.
The final benchmark revision may be slightly smaller, given more recent data, but it is likely to be large historically. The average absolute revision over the past 12 years has been only 0.2% of employment.
Revisions to the birth-death model, which adjust for newly opened and closed establishments not yet captured in the survey, are also likely to be downward.
