Trump’s Attacks on Jobs Numbers Are Noise But Still Dangerous

By now, the Trump White House has a simple strategy for dealing with what it considers bad news: Attack the messenger. They have pressured TV networks to cancel comedy programs, filed lawsuits against major media outlets, and of course even fired the head of the Bureau of Labor Statistics (BLS) for releasing jobs data they didn’t like.

Trump’s abrupt decision to remove BLS Commissioner Erika McEnterfar was unsurprising, but the absurd reasoning – and the further damage he could still do – deserve a closer look.

The back story is pretty straight-forward: On August 1, the BLS released its July jobs report, which found weaker than expected job creation and a slight increase in unemployment rate. The report also revised job creation numbers from the prior two months. This didn’t sit well with Trump, who claimed the numbers were “RIGGED in order to make the Republicans, and ME, look bad.”

There is no evidence to support the claim of rigging, which would be virtually impossible given the way the data are compiled, and really doesn’t even make sense. If the intent was to make Trump look bad, the jobs data get the most attention the day they are released, not when they are revised one or two months later. Why would someone make Trump look bad, release exaggerated jobs numbers on the day when the data gets lots of attention and then revise them downward when no one is looking?

Trump didn’t help his case when he picked E.J. Antoni, chief economist at the Heritage Foundation, as his nominee for BLS Commissioner, after vowing to appoint a new BLS chair who is “much more competent and qualified.” In fact, Antoni does not seem to have even minimal familiarity with BLS methods or economic statistics. His nomination has been widely criticized across the political spectrum.

It is worth pointing out that just a few weeks ago, the White House was defending BLS from criticism. In July, an administration spokesperson told the Washington Post, “Baseless attempts to undermine confidence in BLS data does not change the fact President Trump’s pro-growth economic agenda has created more than half a million jobs since he took office.” In short, everything is fine so long as the numbers tell the story the White House wants you to hear.

There is Nothing Strange About Jobs Revisions

The revisions that sparked Trump’s tantrum are a totally normal part of the process. The monthly jobs data come from a payroll survey of employers. While employers are supposed to return the survey the month it is sent out, many are not returned until the next month, or even two months later. As more surveys are returned, BLS revises its data. If many employers are slow in returning their surveys, or if the group that returns surveys late is atypical, the revisions can be large relative to the initial job growth number.

In addition, once a year BLS issues what are called ‘benchmark’ revisions that incorporate unemployment insurance data to get a more complete picture of employment. The preliminary benchmark revision arrived earlier this month, and found that the economy added 911,000 fewer jobs than previously thought between March 2024 and March 2025.

This is the second year in a row that BLS issued a large revision. In August 2024, it determined that 818,000 fewer jobs had been created. Trump accused the Biden administration of “fraudulently manipulating job statistics,” though it is unclear how reporting (right before the election) that job growth was less impressive than it had looked would help Biden.

None of this is hidden from view or manipulated by actors seeking to do harm to President Trump. He is not the first president to make such claims; as Dean Baker recently noted, Richard Nixon had similar conspiracy theories about BLS.

Real Concerns, Not Trumpy Ones

While Trump is not able to identify actual problems with how the BLS works, that doesn’t mean there are not any concerns. But those, too, are not secret – indeed, outside economists and folks within the agency have long identified them. We know, for example, that response rates to the household surveys have been dropping. In the case of the Current Population Survey, which is used to calculate the unemployment rate, it has gone from 90 percent to 70 percent. The agency has found it harder to recruit businesses to participate in the payroll survey, and getting timely responses from those that do participate is a challenge, falling from 74 percent to 57 percent in the first month.

These problems are not insurmountable, but like many government agencies, BLS has faced years of tight budgets and understaffing. A letter to Congressional leaders signed by dozens of economists outlined all of this and more, and specifically advocated for adequate funding with an eye towards modernizing existing systems.

The Trump administration, though, has other ideas. The president recently proposed an 8 percent cut to the agency’s budget. Early in his term they disbanded unpaid outside advisory groups that provide feedback on a range of technical matters. The BLS has suspended some data collection it uses to collect inflation data, as well as some of the wholesale pricing data that is part of the producer price index. If the administration wants people to trust the government’s economic data, they appear to be taking steps that move in the opposite direction.

What comes next? Earlier this month, the Labor Department’s Office of Inspector General announced it would be conducting a review of the BLS’s handling of jobs and inflation data. In a normal administration, this might not be cause for concern. But given the repeated attacks on the agency from a president who prioritizes loyalty over accuracy, there are reasons to be alarmed.

This first appeared on Dean Baker’s Beat the Press blog.

The post Trump’s Attacks on Jobs Numbers Are Noise But Still Dangerous appeared first on CounterPunch.org.