Commuters cross K Street NW outside of the Farragut North Metro station in Washington, D.C., on November 13, 2025.
Here’s a shocker: If you give businesses more time to respond to a survey, chances are you’ll get more data back in return. That’s exactly what transpired in the latest U.S. jobs data from the Bureau of Labor Statistics (BLS). After the government shutdown pushed back the release dates of the September and November reports, firms were given more time to report their payrolls information, and the agency collected more responses as a result—a development that stands to boost the accuracy of the numbers.
The BLS collects payrolls data over the course of three months on a rolling basis. As it gathers more information, initial payrolls numbers can get revised—sometimes so much so that the update changes the perception of the state of the labor market. While the agency says revisions make the figures more accurate over time, critics wonder whether that tradeoff for speed is really worth it, especially as revisions are getting increasingly politicized.
“With any program that has revisions, it’s because you’re trying to balance timeliness with accuracy,” said Erica Groshen, who was BLS commissioner from 2013 through early 2017. “You want to give people both, but you can’t give both at the same time.”
Economists pay close attention to collection rates, which measure how much data was retrieved in a given month. For the first collection period of the payrolls survey, the rate was 80.2 percent in September and 73.8 percent in November—some of the highest readings in the past five years. The rate for October, which was reported with the November release out today, was similarly high at 73.9 percent. The BLS attributed the rise in recent months to the extended collection period.
While the BLS halted data collection and most agency operations during the shutdown, the agency said businesses continued to electronically report payrolls independently during the lapse in funding. And when the government reopened, it gave firms more time to respond.
The BLS release of the November jobs report revealed that September payrolls growth was revised down by 11,000. That’s on the smaller side of initial revisions this year. But because of the shutdown, the initial estimate of September employment wasn’t reported until November 20, well after the originally scheduled October 3 release date. By that time, the data were fairly stale, and economists and investors had already moved on to parsing private data sources for a more current snapshot of the job market.
“It’s nice that the revisions are low, but I don’t think the sacrifice is worth it by waiting another five, six weeks,” said Omair Sharif, president of Inflation Insights LLC.
Jobs numbers have garnered greater attention since President Donald Trump fired the head of the BLS in August, following substantial revisions to jobs data in prior months, which he called “a major mistake.” His nominee to fill the position—which the administration ultimately pulled from consideration—had suggested suspending the monthly report and publishing only quarterly numbers until issues with data collection were corrected.
In addition to the monthly revisions, the BLS also conducts an annual revision that benchmarks the payrolls numbers to a more accurate but less timely series based on unemployment insurance tax records. That process has resulted in sizable annual revisions in recent years. A preliminary estimate in September suggested that payrolls in the year through March would be marked down by the most on record. The final number is due early next year.
Last year’s initial projection was similarly large and drew condemnation from key Republican senators at the time. They pressed the BLS on whether it’s obligated to release the monthly reports on the first Friday of the month, or whether it can wait longer—and whether it’s considered pushing back publication until it has more reliable data. The BLS didn’t respond to Bloomberg requests for comment on those questions Tuesday.
While it’s too soon to say what the revision to the November jobs counts will be, as the December employment report isn’t due until next month, there’s reason to believe that the adjustment should be relatively minor given the high collection rate, Sharif said.
“There’s less sense of scope for bigger revisions because you’re capturing most of the responses already,” he said. “If there’s any silver lining to all these delays, you’re probably getting a more accurate first read on payrolls than we have most of this year.”
————————————————————
Copyright 2025 Bloomberg. All rights reserved. This material may not be published, broadcast, rewritten, or redistributed.
© Arc, All Rights Reserved. Request academic re-use from www.copyright.com. All other uses, submit a request to TMSalesOperations@arc-network.com. For more information visit Asset & Logo Licensing.