Summarizing his befuddlement with numbers, Mark Twain observed that there were “lies, damned lies and statistics”.
The acerbic phrase later become so deeply embedded in popular consciousness that it once formed the title to an episode of The West Wing, NBC’s portrayal of a fictitious US president played by Martin Sheen.
Now professional economists and number-crunchers fear the aphorism could become a White House theme in real life. Buffeted by global markets and public opinion – both of which show a wary skepticism of Donald Trump’s affinity for trade wars – the president may be about to turn his renowned hostility to truths at odds with what he believes towards public servants charged with producing accurate information.
A proposed rule change making it easier to fire civil servants deemed to be “intentionally subverting presidential directives” could pave the way for the White House to fire statisticians employed to produce objective data on the economy but whose figures prove politically inconvenient, experts warn.
Statistics released by agencies such as the Bureau of Labor Statistics (BLS) and the Bureau of Economic Analysis (BEA) are used by the Federal Reserve Bank to set inflation policy and interest rates. They also form the basis on which businesses and investors take decisions.
The US’s global reputation as a stable economic power and a reliable partner goes hand-in-hand with its long history of producing accurate data, dating back to the establishment of the BLS in 1884. Interfere with the latter and you risk sacrificing the former, experts warn.
But with Trump under pressure to explain shrinking gross domestic product (GDP) figures amid economists’ warnings that tariffs could trigger a recession, the administration could use new employment rules to pressure workers into “cooking the books”.
“There are a number of changes to the civil service that make it much easier for the administration to try to interfere with the activities of the statistical agencies and that worries me,” said Erica Groshen, a specialist in government statistics at Cornell University.
While acknowledging that there is as yet “no evidence” theTrump administrationhas done so, Groshen, a former commissioner at the Bureau of Labor Statistics (BLS), fears a new rule proposed last month by the White House’s office of personnel management threatens the future integrity of federal agencies’ figures.
Thechange, based on anexecutive ordersigned by Trump on 20 January immediately after his inauguration, would reclassify about 50,000 as-yet-unspecified permanent civil servant positions to “policy/career” category, thus enabling their removal for “poor performance or misconduct”.
The precise roles to be so redefined have yet to be revealed but Groshen fears statistic specialists will be in the administration’s crosshairs.
“Bureau of Labor Statistics’ leaders could be fired for releasing or planning to release jobs or inflation statistics unfavorable to the president’s policy agenda,” shewrote in a briefing paperthat urges organizations dependent on BLS figures to submit comments criticizing the proposal.
“By making it easier to remove employees if a president determines that they are interfering with his or her policies, it increases the potential for passivity or political loyalty to be prioritized over expertise and experience.”
Trump regularly cast doubt on the accuracy of economic datawhen in opposition – calling positive BLS jobs figures during the Obama and Biden administrations “fake” but hailing them as accurate when they painted a rosy picture of the economy during his first presidency.
Last month, when GDP figures showed an economic contraction during the first 100 days – partly fueled by tariffs –Trump put the blame on Biden.
“We had numbers that, despite what we were handed, we turned them around and we were getting them really turned around,” he told reporters.
The commerce secretary, Howard Lutnick – who has direct responsibility over many of the statistical agencies – has suggested changing the way GDP is calculated in a way that might provide more upbeat figures but which would mark a departure from established practice and international standards.
Diluting data agencies’ impartiality risks adding the US to the category of countries which have had the veracity of their economic statistics openly doubted, critics say. Groshen citedArgentina, whose official inflation figures were rejected as falseby the International Monetary Fund (IMF), and Greece, wheregovernment statisticians were saidto have miraculously made inflation and disqualifyingly high budget deficits “disappear” to enable it to join the European Union’s single currency, the euro, in the late 1990s.
The sleight of hand had dire consequences. The 2008 global financial crash propelled the country’s economy into a tail-spin, forcing it to seek huge loans from the IMF and the EU, which were only given on condition of harsh austerity measures and cuts to public services.
Popular anger over the conditions in Greece destabilised establishment political parties and led to a rise in support for radical and populist alternatives, including the leftwing Syriza, which won power in 2015. Frequent elections and changes of government since have raised concerns about the health of the country’s democracy.
The IMF also censured Argentina and threatened it with expulsion in 2013 after officials were found to have been grossly understating the inflation rate for the previous six years.
Argentina – historically one of the IMF’s biggest borrowers – did not receive another loan from the organisation until 2018. That loan, followed by another in 2022, failed to stabilise the country’s economy and in 2023, Javier Milei, a far-right candidate and professed admirer of Trump, was elected president pledging drastic spending cuts to address its chronic economic problems.
Last month saw the fund agree to another $20bn bailout for Milei’s government.
Despite these baleful precedents, the Trump administration’s sensitivity to economic figures indicating a tariff-driven slowdown creates a potential spur to follow a similar path, argued Erasmus Kersting, an economics professor at Villanova University.
“I would say that there’s definitely an incentive to cook the books, but I don’t think that it is going to be very easy or feasible to do,” he said, citing the US’s long tradition of producing accurate economic figures.
“The Bureau of Economic Analysis would essentially need to be silenced or defunded and replaced with some other statistical agency, which would then result in different figures. The same would be true of the Bureau of Labor statistics.”
Accurate and unbiased figures are crucial in helping the Federal Reserve form sound policy, Kersting said. In their absence, Trump might have more scope to attack the Fed’s chair, Jerome Powell, who he has already accused of “playing politics” by not bowing to his demand to cut interest rates.
Kitty Richards, a former treasury and White House official under the Biden and Obama administrations, said data collection had been impaired by Elon Musk’s attacks on federal agencies under the auspices of the unofficial “department of government efficiency,” or Doge.
“We should view attacks on government data collection as hand in glove with attacks on journalism,” said Richards, now a senior fellow at Groundwork Collaborative, a thinktank. “Undermining data collection and casting doubt on data that is released is part of a program of undermining the public’s ability to learn the truth.”
Even a temporary interruption of the US’s established data-collecting capacity would be a “real tragedy” and lead to a permanent loss of knowledge, she said. “You can’t go back and fix it. If you have a data series stretching back 50 years, then it gets cut for two or three years, you no longer have that 50-year data series. You’ve lost knowledge forever.”
Greshen, who is calling on users of government statistics to object to the proposed civil service changes before a 30-day window expires on 23 May, said the fate of US democracy could hinge on the continued production of accurate figures.
“In a democracy, you want to be feeding people the right information so they will make the right choices. But if the goal is to destroy democracy, you’d want to control the statistics to fit your story … you want to be promoting your own version of reality.”