Skip to content

U.S data agencies need ‘immediate’ help to do their job, report says

The federal agencies are struggling with staffing, budget constraints, and eroding trust, says the American Statistical Association.

The Suitland Federal Center, which houses the Bureau of Labor Statistics headquarters, in Suitland, Md.
The Suitland Federal Center, which houses the Bureau of Labor Statistics headquarters, in Suitland, Md.Read moreAl Drago / Bloomberg

U.S. data agencies need urgent help from the Trump administration and Congress to ensure they can carry out their basic duties and restore public confidence amid a deepening crisis, according to a new report by some of the country’s top statistics experts.

The agencies are struggling with fragile capacity and eroding trust — as well as diminished safeguards for data integrity — and need more money and staff, says the study led by the American Statistical Association. It cites challenges that have grown more acute since last year’s inaugural version of the report, published before President Donald Trump returned to office.

Government departments such as the Bureau of Labor Statistics, Bureau of Economic Analysis, and Census Bureau are tasked with publishing all kinds of data, which cover the economy and many other topics, and are key to decisions by policymakers, investors, and companies as well as the wider public. Their work has been made harder by longstanding problems such as shrinking budgets and falling response rates for surveys — as well as more recent threats to their independence and integrity.

“Immediate action must be taken to halt the severe decline in the federal statistical agencies’ ability to meet their basic mission and be positioned to keep up with increasing information needs and to address uncertainty in the trustworthiness of federal statistics,” says the report, which was published Wednesday.

In Trump’s second term, the strain on federal statistics has intensified. His administration’s campaign to downsize the government left gaping holes in many agencies, with data products becoming collateral damage of the staffing cuts. Organizations such as the ASA have created dashboards to keep an eye on changes to datasets and highlight any that disappear.

Headcount at the BLS was down 20% last fiscal year compared with the previous one, and the BEA has seen a 25% drop since 2019, the report says. Trump has proposed further cuts in his 2026 budget.

Trump’s most drastic action so far on the data front came when he fired the head of the BLS in August after a weak jobs report — accusing her, without providing evidence, of rigging the numbers to make him look bad. Economists and statisticians have lined up to reject that claim. The administration pointed to large revisions in employment data and said the numbers needed to be “fair and accurate.”

Just a day before all this drama unfolded, the statistics experts behind Wednesday’s study had published an interim report saying they were confident that data could be trusted and there were no signs of meddling by the executive branch. Trump’s move against the BLS forced a rapid rethink. The document was amended to say that the president’s actions “undermine trust in the future by accusing statistical agency heads of past political manipulation.”

The group’s new report cites a survey which found the share of the public expressing trust in federal data had declined to 52% in September, from 57% in June.

It calls out other administration actions this year that undermined official statistics, like the termination of advisory committees, failure to fill leadership roles, and elimination of datasets without consulting Congress or the public. It notes that the positions of chief statistician and Census director have been staffed with political appointees who already held other full-time positions, and argues this could further erode trust.

The report urges the Trump administration to exempt key data-agency positions from the federal hiring freeze, and calls on Congress to fund research and enhancements in IT infrastructure that can help improve the quality of statistics. Such measures would “begin to restore the system’s capacity to deliver the timely, relevant, and trustworthy statistics the nation depends upon,” it says.