News: Republican Administration Fires Head of BLS After ‘Bad’ Jobs Report
President Donald Trump fired the head of the Bureau of Labor Statistics (BLS) immediately after the recent jobs report publication.
Thank you for reading our work! Nominal News is an email newsletter read by over 4,000 readers that focuses on the application of economic research on current issues. Subscribe for free to stay-up-to-date with Nominal News directly in your inbox:
If you would like to support us in reaching our subscriber goal of 10,000 subscribers, please consider sharing this article and pressing the like❤️ button at top or bottom of this article!
Republican President Donald Trump fired the head of the Bureau of Labor Statistics (BLS) after the most recent release of jobs data. Due to the importance of this event, we will go over it in today’s newsletter, instead of our planned article on the impacts of unemployment insurance, which will come out next week.
In Part 1, we will briefly go over what happened, while in Part 2, we will discuss how measurements are misunderstood.
Part 1: The firing of BLS Commissioner
What happened
On August 1, Republican President Donald Trump fired the BLS Commissioner, Erika McEntarfer. This happened soon after the BLS published its latest employment report for the month of July. Moreover, the BLS also updated the employment reports for the months of June and May.
For the month of July, the BLS report stated that the number of jobs increased by 73,000, while unemployment was at 4.2%. For the month of June, the BLS revised the number of jobs created from 144,000 to 19,000 (a 125,000 reduction), while for May, the number of jobs created changed from 147,000 to 14,000 (a 133,000 reduction).
Why it happened
The reason for the firing given by the President was as follows:
"We need accurate Jobs Numbers," Trump wrote. "She will be replaced with someone much more competent and qualified. Important numbers like this must be fair and accurate, they can’t be manipulated for political purposes."
What does it mean
In Part 2, I will explain why the ‘reason’ given by the President is incorrect and unfounded. That is, the BLS provides high quality, fair and accurate data.
The main concern arising from this decision is that new data releases will no longer be fair and accurate. This is a valid concern, although somewhat alleviated by the fact that if manipulation were to occur, the sheer number of people involved in the BLS data collection would result in at least a few of these employees raising concerns. Moreover, BLS data aims to reflect reality – changing the numbers won’t change the reality.
Part 2: Misunderstanding Measurements
The idea that economic data is being ‘manipulated’ has been mentioned in many circles, and not just by politicians. Prominent investors and tech leaders have accused government statistical agencies of intentionally manipulating data. But the truth is that they simply don’t understand how measurements work. Let’s take a look at the BLS data.
What the BLS Actually Measures
The first common misconception about the BLS jobs report is that the BLS attempts to estimate the change in the number of US jobs rather than what is the total number of jobs in the US. This is partially perpetuated by media reports focusing solely on the number of newly created jobs.
In reality, the BLS surveys thousands of employers to count how many positions exist in a given month, not just how many were newly created. This number of positions in the US is almost 160,000,000 as of July 2025. The change is simply the difference between each month’s total number of jobs. Moreover, the BLS states that the 90% confidence interval of the change in the number of jobs is approximately plus/minus 136,000. That means actual job growth is likely to be between -60,000 all the way to 200,000 created jobs in July.
Why Revisions Happen
The second misconception is that revisions are necessary because the initial estimate was manipulated. In reality, the way the BLS collects data to estimate the total number of jobs is based on survey data from employers. Many employers end up delaying responses to the BLS survey, past the deadline the BLS set for them. However, rather than completely ignoring these late responses, the BLS incorporates them later and revises their estimate.
It is also worth understanding how small these revisions are. In the last two months, where the revisions have been numerically larger than usual, each months’ tally was revised by around 130,000 jobs. That means the initial estimate was ‘off’ by 130,000/160,000,000 which is 0.08%.
To put this in perspective – if one were to weigh a 160 pound person (72.57kg) using a home scale, under a similar revision using a better scale, this person would be deemed to weigh 159.87 lbs (72.52kg).
Once we understand what the BLS actually does, it’s pretty clear that their work and accuracy is more than remarkable.
The “Drama” of Measures
None of this would matter if we better understood what is the point of measurements, like the BLS’ measurement of the number jobs. In one of our previous articles, Don’t Tie Yourself to a Number, we discussed why it is important to understand what a particular measure aims to estimate and that just looking at the headline number can often be misleading on its own without understanding the context.
To give a recent example, the US first quarter 2025 Gross Domestic Product (GDP) growth rate came in at -0.5%. Seeing this number on its own could be interpreted as a significant recession in the US started. However, under closer inspection of how this particular official GDP measurement estimates output, economists quickly noticed that the negative number was driven by the front-running tariffs by importers (the act of buying imported goods before tariffs kick in), as well as the fact that these imports were not caught by the GDP measure in inventory. In quarter 2, US GDP growth rebounded to +3%, as both patterns reversed (less import front-running and increased inventory). The first quarter GDP growth number was used by critics to demonstrate why government policies were bad, while the second quarter GDP growth number is used by government supporters to show that economic policies are good. It is quite clear that just using the GDP growth number on its own, absent any context, cannot answer this question
The same pertains to job growth numbers, which is even a more complex data set. When the BLS data came in, many headlines quickly jumped to certain conclusions. But how can we know that a job growth number of +73,000 means the economy is doing good or bad? Turns out, that in depth analysis by labor market economists, including Guy Berger’s great post on the BLS report, shows us that with the fall in immigration in the US, we could expect to see lower job growth numbers than we did in the past. One could theoretically have weak job growth and a strong labor market, if labor supply is reduced (for example, via lower immigration).
The One Takeaway
Having been writing Nominal News for 3 years, if there’s one thing I would like for my readers to take away is to always interpret economic data within its relevant context and analysis. A number, whether it’s GDP, inflation or job growth, does not have much meaning without a broader context. An inflation measure of 4% does not always mean inflation is out of control, a negative GDP reading doesn’t always mean recession (neither does a high GDP reading mean the economy is booming). A particular jobs-number also doesn’t mean the economy is doing well or poorly. That’s why I rely on other labor economists and their analysis that look at far more than just one data point to interpret what is happening in the economy.
Many online commenters, including some high profile billionaires, have stepped in with ‘solutions’ to the jobs data collection, including the use of AI. The reality is that all the proposed solutions and ideas have long been discussed by economists and statisticians. Too often people assume the solution to any economic problem is doing one simple thing – as if there is a large red button at the BLS (or any other institution) that says ‘collect data better’ that no one has pressed. Unfortunately, it is not that easy, and a bit of humility would not hurt.
Interesting Reads from the Week
- goes over the last BLS jobs report. From his write-up:
TL;DR: This was a weak jobs report. However, unlike a typical business cycle, the weakness is currently coming from both the supply side and the demand side.
- discusses a recent study looking at how economists personal views’ on policies are influenced by which economist made a statement. Very cool study.
- , who recently lauched The Peeples Economist, talks about research showing how increases in minimum wages favor big businesses, as they have the financial ability and productivity to pay these wages, while small businesses might not be able to afford the increased minimum wages.
Thank you for suggesting my article!
Thank you for the shout out! And nice post on the issues with these data. I'm thankful for people like you writing articles to educate others on how to think like an economist (and also how to think about data). We really need to push intro stats courses in high school to help people understand context and confidence intervals.