Why jobs revisions that led Trump to fire statistics head were so huge

Show Caption

The massive downward revisions to U.S. job growth that led President Donald Trump to fire the head of the agency that reported the numbers can be explained by a rapidly slowing economy and job market amid Trump’s trade and immigration policies – not data manipulation, forecasters say.

Trump on Aug. 1 fired Erika McEntarfer, the commissioner of the Bureau of Labor Statistics, hours after the agency reported a disappointing 73,000 job gains for July, and more significantly, revised down payroll growth for May and June by an outsize 258,000.

That marked the largest two-month revision ever outside of recessions, Goldman Sachs said. It left monthly job gains averaging an anemic 35,000 from May through July.

“Important numbers like this must be fair and accurate, they can't be manipulated for political purposes,” Trump said on Truth Social. He added that “today's jobs Numbers were RIGGED in order to make the Republicans, and ME, look bad.”

But forecasters said the revisions actually clear up a disparity between surprisingly resilient job gains and other economic indicators, such as gross domestic product, that have been feeble this year.

“I think the revisions have really resolved a discrepancy between GDP data and payrolls,” said Samuel Tombs, chief economist of Pantheon Macroeconomics. “I don't see any reason to suspect manipulation here.”

“Taken together, the economic data confirm our view that the U.S. economy is growing at a below-potential pace,” Goldman Sachs wrote in a note to clients.

Unemployment among Black Americans rises It can be an early sign of economic strain.

What is the state of the economy?

Trump’s tariffs hammered activity in the first quarter by pumping up imports (which are subtracted from growth) as retailers and manufacturers stocked up on foreign goods to avoid the imminent levies. That led to a second-quarter reversal that juiced GDP. After the tariff effects canceled each other out, the economy grew at a meager annual rate of 1.2% the first half of the year.

There are two ways to grow the economy – by adding workers and by increasing productivity, or output per worker. Since the 1.2% average rise in GDP can be entirely explained by strong productivity gains, the economy’s sluggish performance is consistent with virtually no net job gains from May through July, Tombs said.

There are other signs that the economy and job market are losing steam. Just 42% of CEOs of small and midsize companies plan to add to their staffs in the next year, the lowest on record dating to 2003, according to a June survey by Vistage, a CEO networking group.

How does the jobs report work?

Each month, the BLS provides an initial reading of job gains for the previous month and revises figures from the prior two months twice based on follow-up surveys. The downgrade of 133,000 jobs for June marked the first revision for that month, and May's 125,000 drop was its second and final revision.

To come up with its monthly job growth estimates, the agency surveys 631,000 job sites operated by 121,000 businesses and government agencies across the country. The bureau revises the data twice because many employers don’t respond to the first survey or because officials modify the factors it uses to seasonally adjust the figures, Goldman said. For example, since hiring is traditionally weak in the summer, the seasonal adjustment would boost the employment tally to account for that dip.

Even before Aug. 1’s stunning revisions to May and June, job growth for January through April was revised down each month by an average of 52,000 ‒ based on changes from the first to the third estimates. Last year saw an average downward revision as well, but it was just 20,000. Since 1979, the median two-month combined estimate change was an upward revision of 10,000.

Here's why economists believe the payroll revisions for May and June were so large:

Falling response rates, slowing economy

A declining share of employers have been responding to the BLS’s job surveys. In July, 58% responded, similar to July 2024 but down from 68% in the same period in 2022 and 2023 and an average 74% in the 2010s, according to Tombs and BLS figures.

Small businesses, in particular, have struggled to cope with Trump’s tariffs, which must be passed along to consumers through higher prices or absorbed by the companies, squeezing their profits, Tombs said.

Amid those troubles, “Responding to BLS is not a high priority” for small- and midsize firms, said Jonathan Millar, senior U.S. economist at Barclays.

Some firms also may be hesitant to acknowledge they've pared back hiring, even though the surveys are anonymous, Tombs said.

When they ultimately do report their staffing numbers – the response rate for the third payroll estimates has topped 90% – they likely have decreased from the previous month. Companies increasingly have been holding off on hiring due to the uncertainty generated by the tariffs, Millar and Tombs said.

That likely has sparked sharp downward revisions to job gains throughout 2025, but the hiring pullback may have reached “an inflection point” in May and June that more dramatically dimmed the jobs picture, Millar said.

“I don’t think we've faced a situation where we had such a profound supply shock,” Millar said, referring to the constraint on foreign parts and products caused by tariffs.

How the government interprets slowing job gains

When job growth slows, BLS in its first estimate mistakenly attributes the downshift to changes in seasonal trends, Goldman Sachs said. But as subsequent readings also come in softer, the agency's model learns it’s not the result of a seasonal adjustment quirk but rather a genuine pullback in hiring and revises down the jobs totals, the research firm said.

The immigration crackdown

Former President Joe Biden began restricting crossings at the Southern border in June 2024 and Trump stepped up enforcement and launched massive deportations in 2025, Millar said. Since it takes six to nine months for migrants to settle into jobs, the crackdown likely started to show up more prominently in the jobs numbers in May and June, he said.

Little hiring, few layoffs

For many months, employers haven’t hired lots of workers, but also haven’t laid off many after grappling with dire worker shortages during the pandemic. That unusual dynamic could be throwing off seasonal adjustment factors that expect ramped-up hiring and layoffs at certain times of the year, causing officials to revise down job gains more sharply, Goldman suggested.

State and local government jobs

About 20% of the big downward revision in May and June was due to the challenges of making seasonal adjustments to state and local education employment figures at the end of the school year, when many teachers and other employees come off the payroll, Goldman said. Initially, BLS overestimated total job numbers in that sector after seasonal adjustments and then revised down the figure by 51,000 for the two months, Goldman said.

Contributing: Joey Garrison, Paste BN; Reuters