By firing one of the country’s top economic research officials, President Donald Trump might have instilled less confidence in the legitimacy of
On Friday, Trump terminated Bureau of Labor Statistics Commissioner Erika McEntarfer, claiming that she “rigged” the data in last week’s employment situation report to hurt the current administration. He also accused her of manipulating data to help his rival in last fall’s presidential election, Kamala Harris.
“We need accurate Jobs Numbers. I have directed my Team to fire this Biden Political Appointee, IMMEDIATELY,” Trump wrote as part of a lengthy post on Truth Social on Friday afternoon. “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.”
But economists, analysts and researchers say the odds of BLS data being intentionally inflated or deflated for political gains are exceedingly slim. Along with facing strict methodological standards and transparency requirements, the bureau’s data gathering operation is highly diffused — giving seasoned researchers more say over collection practices than political appointees.
“I would trust that the internal processes are fairly robust and the U.S. is signed up to international guidelines, so that’s the optimist’s case,” said James Rossiter, head of global macro strategy at TD Securities. “There’s very little way for this data to get manipulated, even at the top.”
For these reasons, there is little concern that a Trump appointee will suddenly be able to concoct more favorable readings on the labor market or inflation. But, that does not mean the move is inconsequential.
Peter Earle, director of economics and economic freedom at the American Institute for Economic Research, said barring dramatic and obvious changes to collections practices, the BLS data will likely remain as reliable as it has been for years. But, he noted, the president’s move inserts a level of doubt that will be difficult for some economic actors around the world to shake off.
“Ironically, the assertion that the data was politicized is going to lead to its inherent politicization over the coming future,” Earle said. “Everything from financial markets to central banks use algorithms that rely upon this economic data. By undermining confidence in its objectivity, it could impair a lot of U.S. soft power.”
By diminishing the country’s role as a global benchmark setter, Earle said the move further diminishes the incentive for foreign investors to buy U.S. stocks and bonds — furthering a trend that has been driven in recent years by aggressive sanctions regimes, the ballooning national debt and, more recently, sweeping uncertainty about the trajectory of the American economy.
Mark Zandi, chief economist for Moody’s Analytics, said a loss of credibility for government data could also have more direct implications for banks, many of which rely on accurate economic readings to structure loans.
“We work with all the banks. If you look at their credit risk models, they’ve got this data in there. They’re using it straight up in the model,” Zandi said. “If they’re getting the data later or the quality of the data is poor and they don’t believe the data, then they’re going to charge a higher interest rate to make that loan, because they just don’t know exactly where they stand relative to the economy.”
The Trump administration has toyed with the idea of changing longstanding data collection practices for months. In March, former White House special advisor Elon Musk and Commerce Secretary Howard Lutnick — who oversees the other key economic research agency, the Bureau of Economic Analysis — both raised issues with calculation of gross domestic product, suggesting that the prevailing model be changed to factor out government spending.
The president’s latest frustration arose from last week’s BLS report on hiring in July, which showed that employers added just 73,000 jobs last month — far below the rate tracked in recent months. The agency also revised down its May and June estimates by 258,000. Trump claimed, without evidence, that this revision was an intentional effort to make his administration look bad.
“Last weeks Job’s Report was RIGGED, just like the numbers prior to the Presidential Election were Rigged,” Trump said in a Monday morning post. “That’s why, in both cases, there was massive, record setting revisions, in favor of the Radical Left Democrats. Those big adjustments were made to cover up, and level out, the FAKE political numbers that were CONCOCTED in order to make a great Republican Success look less stellar!!! I will pick an exceptional replacement. Thank you for your attention to this matter. MAGA!”
Revisions are common in both monthly data series, such as employment and inflation, as well as quarterly data like GDP. The practice is seen as key to accountability and balancing the government’s obligation to be both timely and accurate in its data issuance. Unlike other survey-based reports, the employment figures are ultimately evaluated against real hiring figures reported by business.
While large, the recent revisions were not unprecedented, Earle said. Based on analysis of more than 300 recent employment reports, he said the figure was about 2.5 standard deviations above normal, meaning that he would expect a change of that size once in every 80 or so reports.
“We’ve had four or five downside revisions that came in that strong. We’ve had 6 upside ones that came in that strong,” Earle said. “So improbable? Yes. Impossible? Absolutely not.”
Zandi said the BLS also faces issues that are common across surveyors of all stripes: dwindling participation.
“Anything that requires questioning businesses and consumers has had the increasing problem of lower response rates. People are just not responding to these surveys like they did 20, 30, 40 years ago for lots of different reasons,” he said. “With the response rates down, the quality of the underlying data, at least those that are released early, are more questionable. There’s more volatility in the data, more noise.”
Despite that noisiness, Zandi said the BLS has reliably indicated the direction of travel for various economic factors, including employment. He said some changes could be made to improve the accuracy of this data over time, but noted that they should be implemented deliberately.
In the months ahead, economists will be closely watching who Trump picks to lead the BLS moving forward and parsing that person’s background and past statements to determine how they might run the agency and its marquee data series.
That process will go a long way to indicate how aggressive Trump will be in bringing agencies that release economic indicators to heel. Derek Tang, CEO of Monetary Policy Analytics, said Trump’s willingness to draw typically apolitical parts of the government into the fray of partisanship could have ripple effects for generations to come.
“Let’s fast forward five or 10 years, when the landscape has changed,” Tang said. “Are Democrats going to move it back to a pre-Trump setup? Or are they going to throw up their hands and say, ‘look, this is just the way it is and we have to play this game now.’ If that’s the case, then the bottom line is it increases the chance of a policy error, not just by the Fed, but other government agencies. If they don’t have the high quality data they need to make good decisions … that makes the chance of a policy error and a recession more likely.”