President Donald J. Trump recently escalated his criticism of the Bureau of Labor Statistics (BLS), a key federal agency responsible for collecting critical economic data. After the BLS commissioner was fired, he mainly made statements on social media, claiming that the agency’s employment reports were “manipulated” politically.
This high-profile dispute has cast an unprecedented spotlight on the methods and integrity of the BLS, an institution whose data has long served as a crucial barometer for the nation’s economic health. The president’s accusations challenge the very foundation of trust in government statistical reporting, prompting a deeper examination of how these critical economic indicators are actually compiled and what makes their revisions a routine, rather than suspicious, occurrence.
The president’s recent claims were triggered by the BLS’s Friday Employment Situation Summary, which reported that U.S. employers added only 73,000 new jobs in July, a number considerably lower than many economists had predicted, and this was further compounded by adjustments made to the employment data from May and June, intensifying the president’s dissatisfaction.
1.Specifically, May’s jobs total was revised from an initial report of 144,000 to 19,000 jobs, representing a substantial revision of 125,000 positions. For June, the BLS announced that the U.S. economy had added 14,000 jobs, a stark decrease from a preliminary estimate of 147,000, marking a change of 133,000 positions. These revisions, combined, amounted to an adjustment of approximately 250,000 jobs.
Following the report’s release, President Trump took to Truth Social, stating, “In my opinion, today’s Jobs Numbers were RIGGED in order to make the Republicans, and ME, look bad.” Days later, he escalated his claims, announcing the firing of Erika McEntarfer, who had served as the BLS commissioner since January 2024.
On Sunday, President Trump continued to lambast the agency, claiming without evidence that Commissioner McEntarfer had manipulated data prior to the 2024 election. He posted on Truth Social, “Head of the Bureau of Labor Statistics did the same thing just before the Presidential Election, when she lifted the numbers for jobs to an all time high.”
2.He further asserted, ‘I then won the Election, anyway, and she readjusted the numbers downward, calling it a mistake, of almost one million jobs. A SCAM! She did it again, with another massive “correction,” and got FIRED! She had the biggest miscalculations in over 50 years,’ highlighting the profound nature of his skepticism about the accuracy of federal economic data.
Established over 140 years ago, the Bureau of Labor Statistics functions as an independent agency within the U.S. Labor Department, dedicated to gathering a wide spectrum of economic information, including employment figures, inflation trends, wage data, and other vital labor market indicators, with its commissioner appointed through a Senate-confirmed process for a four-year term that often spans across different presidential administrations.
3.To compile its detailed monthly reports on the U.S. jobs sector, the BLS utilizes two main surveys: the Household Survey, administered by the Census Bureau and surveying around 60,000 households nationwide to assess employment status and labor force participation, and the Current Employment Statistics (CES) survey, which collects data from approximately 119,000 businesses and government agencies.
The second crucial tool is the Establishment Survey, which is conducted directly by the BLS itself. This survey reaches out to around 120,000 non-agricultural businesses, collecting detailed data on payroll numbers, employee earnings, and hours worked. The sample size for this survey is notably robust, representing approximately one-third of all non-farm jobs in the country.
4.Once the raw data is collected through these rigorous survey methods, a meticulous process of verification and analysis begins. Economists and statisticians within the BLS diligently check the data for any potential mistakes, combine the various data points to identify emerging trends, and then proceed with a thorough analysis of the results. This multi-layered approach is designed to ensure the highest possible degree of accuracy and reliability in the reported figures.
The data meticulously produced by the BLS is absolutely essential, serving as the fundamental basis for a vast array of economic decisions made by both governmental bodies and private enterprises, influencing everything from understanding economic trends and informing investment strategies to guiding critical planning for business expansion or contraction.
Despite the sophisticated methodologies employed, revisions to job reports are a common and expected aspect of the data collection process. Katharine Abraham, an economics professor at the University of Maryland who served as BLS commissioner from 1993 to 2001, elucidates this point, stating, “As more information comes in, the numbers get revised. That’s just a part of the process.”
5.She further explains the inherent trade-off in the system: ‘There’s a trade off between getting numbers out quickly and having them be as accurate as possible.’ Early job report figures often depend on data from larger companies that can respond promptly to surveys, while information from smaller businesses, which are often more sensitive to economic fluctuations, usually becomes available later.
This phenomenon of revision is not new or isolated. Historical records demonstrate that sizable revisions have been a recurring feature of BLS reporting. For instance, CNN notes that much larger revisions occurred in 2020 and 2021, when job numbers experienced significant fluctuations due to the unprecedented impact of the pandemic.
Historically, revisions in four specific months during prior years have actually exceeded those observed in the recent report, including a record-breaking 679,000-job revision in March 2020 that was largely attributed to extremely poor survey responses during a national lockdown, and even outside the pandemic, significant adjustments like a 143,000-job decrease in January 2009 have occurred.
The total downward revision of 258,000 jobs for May and June of this year, while substantial, is actually within the historical range of such adjustments; BBC Verify notes that this figure represents the most significant change since record-keeping began, excluding the immediate post-Covid pandemic months, and the BLS reports the average monthly change in job figures since 1979 has been 57,000.
6.Moreover, revisions tend to increase in magnitude during periods of economic turbulence. Instances of the BLS revising down monthly job numbers by over 100,000 have occurred eight times since 2000, with many of these instances coinciding with the 2008 financial crisis. For example, a 143,000 reduction was made to the January 2009 figure during the Obama administration.
It is also noteworthy that job gains for the entire year in 2009 were eventually estimated to be 902,000 lower than the initial figures, marking the largest full-year revision on record. Similarly, job creation under President Joe Biden in 2024 was revised down by 598,000, an update that also generated political debate, although it was a smaller change than the over 800,000 initially estimated in 2023.
In the current economic environment, many analysts expect the June data to be revised as school employment increases abnormally during the month when most schools typically close in the summer, and subsequent survey responses often disproportionately represent small businesses that are more susceptible to economic changes such as new tariffs, leading to downward adjustments.
