On the Prospect website

Kuttner on TAP

Shooting the messenger at BLS

Could Trump’s new commissioner, E.J. Antoni, succeed in disguising the worsening economic realities?

After Trump fired the commissioner of the Bureau of Labor Statistics, Erika McEntarfer, because he didn’t like the bad news on jobs in the revised BLS August report, he nominated E.J. Antoni, the buffoonish director of economics at the Heritage Foundation. Last week, I wrote a short post pointing out that BLS is overdue for some reforms, though it is unlikely to get the right sort of reforms from Antoni. Since then, I’ve interviewed several experts and done a deeper dive, and the story gets even more interesting.

For starters, one of Antoni’s complaints, echoed by Trump, is that BLS often makes extensive revisions in its jobs numbers after the fact. In early August, the revisions were extreme and totally changed the headline story.

May’s initial report of a 144,000 gain in payroll employment was revised down to 19,000. June’s initial report of 147,000 has been corrected to just 14,000. In July, according to the preliminary number, the economy created 73,000 jobs, but that is also likely to be revised downward in September.

It turns out that there were two main reasons for the extensive revisions. One, ironically, is the impact of the DOGE cuts on the agency’s chronic understaffing.

BLS collects jobs numbers from employers, and the submissions are voluntary. BLS allows firms three months to report. Normally, about 70 percent of firms report in the first collection month, but BLS staffers often have to chase employers to provide the data.

Thanks to DOGE, there were extensive layoffs at BLS, resulting in worse-than-usual delays. BLS also is under a hiring freeze. In addition, says David Hiles, a former senior career BLS economist, lots of experienced people, seeing where the agency was headed under Trump, decided to retire early. “It was a kind of self-deportation by senior staff,” Hiles told me. “When those people go, their embedded knowledge is not transferred, and the people who remain are overwhelmed.”

So, in reality, much of the extreme revision necessitated by late reporting is Trump’s own fault.

Also, whenever the economy is heading in or out of recession, says Heidi Shierholz, former chief economist of the Labor Department and now president of the Economic Policy Institute, the jobs numbers will be less stable and subject to more frequent revision. This is one of those times when the economy is at an inflection point. The revision of the May and June jobs numbers, according to economist Heather Boushey, was the largest two-month downward revision since 1968.

Michael Horrigan, president of the Upjohn Institute and former associate commissioner at the BLS in charge of employment data, points out that in September 2008, at the start of the Great Recession, the first jobs report showed a decline in payroll employment of 159,000, but that was later massively revised to a decline of 401,000.

This economy is clearly generating fewer jobs, a sure sign of recession. GDP growth has declined from an annual rate of 2.5 percent in 2024 to just 1.2 percent in the first half of 2025.

What about inflation? The impact of Trump’s tariffs has not yet shown up very much in the Consumer Price Index, but the uncertainty of the ever-changing rates and exceptions is playing havoc with producer and retailer planning, and contributing to a downturn nonetheless.

The Wall Street Journal recently cited an extensive study by Barclays, in which Barclays economists dug down and calculated what importers actually paid in May. Barclays found that more than half of U.S. imports were duty-free. That included over 90 percent from Canada and over 80 percent from Mexico, as well as almost 80 percent from Taiwan. As our colleague David Dayen has pointed out, the average tariff rate, weighted by the distrubtion of actual imports, is only 9 percent; and imports are only 11 percent of GDP.

Even so, how does an importer decide what to order from, say, India, when it’s not clear from day to day if the tariff will be 25 percent, or 50 percent, and which goods will be exempt? And even if the actual tariffs are lower than the headline numbers and the Consumer Price Index showed modest increases in prices for July, the Producer Price Index signaled a much sharper increase in recent inflation—0.9 percent in one month, or an annualized rate of 10.8 percent. Some of that was due to producers and players in the service economy (which is not subject to tariffs) taking advantage of market power and inflationary expectations to hike prices opportunistically. All of this suggests why we need data on prices, wages, and jobs to be as accurate as possible.

Given the complexity of BLS data collection and the professionalism of the remaining staff, most of the labor economists I interviewed agreed that it would be difficult for Antoni to rig the numbers after the fact—difficult but not impossible. The first time he tried to cook the data, there would be mass resignations in protest, says Jeffrey Fuhrer, former research director of the Boston Federal Reserve and now a fellow at Brookings.

At that point, Antoni could try to bring in a small cabal of DOGE-style 19-year-old computer geeks to come up with new programs to generate the desired numbers by massaging the raw data. But the problem is that the BLS does not have a monopoly on data.

As Fuhrer points out, the Federal Reserve often compares BLS data with the jobs data generated and published by payroll companies such as ADP. And the Fed has its own economists. With AI, it’s possible to scrape the internet for real-time data on prices as well as jobs. So the Fed would rely on its own data to make policy, and businesses and investors would take their cue from the Fed, not the BLS.

Antoni’s nomination will come before the Senate Health, Education, Labor and Pensions (HELP) Committee, most likely in September. The Republicans have only a one-vote majority on the committee, and the members include the last two relatively independent Republicans, Lisa Murkowski of Alaska and Susan Collins of Maine, who faces a difficult re-election. His confirmation is not assured.

If Antoni did get confirmed and did try to cook the numbers, the main effect would be to discredit the BLS, not to persuade anybody of rigged happy news. The better strategy would be to prevent his confirmation in the first place.

~ ROBERT KUTTNER

Follow Robert Kuttner on Bluesky

Click to Share This Newsletter

The American Prospect, Inc., 1225 I Street NW, Suite 600, Washington, DC xxxxxx, United States
Copyright (c) 2025 The American Prospect. All rights reserved.

To opt out of American Prospect membership messaging, click here.

To manage your newsletter preferences, click here.

To unsubscribe from all American Prospect emails, including newsletters, click here.