BLS data is faulty, not rigged 

The federal government’s Bureau of Labor Statistics (BLS) has struggled in recent years to produce an accurate measure of the national employment situation. President Trump has decided that that is a conspiracy against him. Occam’s Razor provides a much more likely answer: the Bureau is doing the best it can to track the numbers using outdated metrics in a rapidly changing economy. Trump’s own tariff policies have likely added to the turmoil. 

On Friday, Trump fired BLS Commissioner Erika McEntarfer, alleging, without evidence, that the numbers in that day’s dismal jobs report were “RIGGED” to make him look bad. The report found that a mere 73,000 jobs were created in July and revised downward the gains of the prior two months, retroactively eliminating a whopping 258,000 jobs. 

Yes, this did indeed look bad for the current administration, but it is hardly the only one to be hit with a disappointing official jobs report. The Biden administration got sideswiped more than once as well.  

Last November, in a report just days before the fall election, BLS found that a mere 12,000 jobs were created in October and revised downward the gains for September and August by a combined 112,000. It was a gut punch to the faltering campaign of Democratic presidential nominee Kamala Harris and likely contributed to Trump’s eventual win. That followed an August 2024 BLS report that found 818,000 fewer jobs had been created between March 2023 and March 2024 than previously determined. 

So, if the BLS is secretly against Trump, why then didn’t it rig those numbers in Harris’s favor? Maybe the answer is that it was just reporting the numbers that it had honestly calculated using traditional methods. 

As the Wall Street Journal dryly noted, this summer’s jobs slowdown “occurred in the wake of Mr. Trump’s April 2 tariff shock, his rapid backtrack from the highest rates and then his willy-nilly threats and deal-making with the world.” My CEI colleague Ryan Young explained, “Tariffs and continuing policy uncertainty are not encouraging businesses to make long-term investments in either capital or labor.” 

The question remains, however: why were the revisions to May and June’s numbers so steep? There is a pattern here and it doesn’t require any conspiracy. BLS has regularly announced major revisions of jobs numbers over the last several decades.  

BLS’s numbers are based on a voluntary monthly survey of 121,000 businesses and government agencies covering an estimated 26 percent of all non-farm employees. The reports trickle in, with the Bureau getting responses from about 60 percent of respondents in a given month. The Bureau revises those monthly reports as new data trickles in.  

This framework made sense for the 20^th century economy but is dated now. The sheer size of the economy means that small mistakes in the sample are magnified and non-traditional employment can be overlooked. New technology has facilitated new types of employment from ride-sharing to social media influencers, and it is not clear how, or even if, BLS’s monthly reports account for them. On top of that, the agency has faced budgetary pressure and cost cuts that have resulted in it limiting some surveys, a practice that has alarmed many statisticians.  

Perhaps the firing of McEntarfer will spur BLS to rethink and improve its methodology. It is clearly overdue. Until then, the wisest course of action is to treat BLS data (and that of other government agencies) with more skepticism. They are, after all, estimates made by humans and humans, even smart ones, are still fallible.  

In any event, Trump’s decision to fire McEntarfer is a case of killing the messenger. If he wants BLS to have more upbeat data to work with in the future, he should re-think his tariff policies.