A controversial firing by President Donald Trump provided researchers a rare opportunity to study something difficult to measure: the value of trustworthy government statistics.

Even a temporary loss of trust in official data may be costly, found a team including Cornell economist Erica Groshen – a former head of the U.S. Bureau of Labor Statistics (BLS) – with an economic impact many times the budgets of the agencies that report key indicators such as unemployment and the consumer price index.

Providing what they describe as a conservative, back-of-the-envelope estimate, the researchers calculated that the Aug. 1, 2025, firing of then-BLS Commissioner Erika McEntarfer – with the administration claiming jobs data had been “rigged” – caused a spike in uncertainty associated with a $20 billion loss in industrial production.

“Just that reduction in trust from the firing has an economic impact 25 times greater than the budget of BLS,” said Groshen, senior labor market adviser at the ILR School, who led the BLS from 2013-17. “When trust in government statistics is eroded, this is the cost of people flying blind, or with a cloudier windshield.”

The findings are presented in “The Value of Reliable Statistics,” a working paper (still subject to peer review) released May 4 by the National Bureau of Economic Research. Groshen’s co-authors are Nicholas Bloom, the William Eberle Professor of Economics at Stanford University; Duncan Hobbs, senior research associate at the American Enterprise Institute (AEI); and Michael Strain, Ph.D. ’12, director of Economic Policy Studies and Paul F. Oreffice Senior Fellow in Political Economy at AEI.

Part of the U.S. Department of Labor, the BLS collects, analyzes and produces reports on the labor market, workers’ earnings and consumer and producer prices that are among the principal federal economic indicators tracked by the White House Office of Management and Budget. The monthly “Employment Situation,” which reports the unemployment rate and non-farm payrolls, is considered one of the most influential sources of government data used to inform business decisions.

As the BLS commissioner, Groshen said, she often sought to explain the importance of the agency’s products to the economy. For example, she once calculated that a consumer price index off by as little as a tenth of a percentage point could result in a $1 billion overpayment or underpayment to Social Security beneficiaries. But Groshen said that by their nature, the benefits of public goods and statistics, in particular, are not easily measured.

“Even if the benefit is huge,” she said, “it’s hard to measure because it accrues just a little bit to everybody. It’s hard to separate out what a benefit is that everybody gets.”

But McEntarfer’s firing, following a report showing that U.S. hiring had slowed, produced a measurable disruption. Groshen’s team confirmed a dayslong spike in the Economic Policy Uncertainty Index (EPU), a measure Bloom helped develop that monitors uncertainty as reflected in news coverage. That’s not surprising, the researchers said, since although there was no evidence of wrongdoing, allegations that data had been manipulated could cast doubt on the integrity of either recent or future releases.

“We make the case that when trust declines, economic uncertainty goes up,” Groshen said.

To determine how much McEntarfer’s firing eroded trust and increased uncertainty, the researchers removed references to two other events on the same day that could have contributed to the EPU spike: the release of the jobs report – which included downward revisions to prior months’ data – and, unrelated, a surprise resignation from the Federal Reserve Board of Governors.

Counting only news that directly mentioned McEntarfer and “uncertainty” during the week before and after her firing, the team determined, conservatively, that the firing accounted for one-fifth of the index’s spike. Applying a statistical model previously used to measure the economic impact of uncertainty generated by the 2008 financial crisis and the pandemic, the team quantified the impact from McEntarfer’s firing as $20 billion in lost output.

To put that number into context, the researchers said, the BLS budget during the 2025 fiscal year was a fraction as much at $704 million.

“The bottom line is that the trust people have in BLS is extremely valuable to the country – an order of magnitude greater than what taxpayers spend on the agency,” Groshen said.

A drop in the uncertainty index as an event recedes from headlines doesn’t necessarily imply that confidence has been restored, Groshen said. Although news outlets clearly report events that raise uncertainty, they do not tend to cover how long uncertainty is sustained. Economic effects might also linger beyond the spike if businesses that had planned to invest no longer feel the timing and other conditions are right.

“You’ve nudged this train off the track just a little bit,” Groshen said.