New applications for U.S. unemployment benefits fell by 6,000 to 189,000 in the week ended Oct. 31, matching the lowest reading since mid-July. But the number of continuing claims rose by 23,000 to 1.718 million in the week ended Oct. 24, the latest period for that measure.
The two figures, published Nov. 5 by the Labor Department, show why a low pace of new filings can coexist with an increase in ongoing claims. Initial claims measure people entering the unemployment insurance system. Continuing claims track ongoing use of benefits and are reported with a one-week lag. Neither series directly measures the pace of hiring.
The broader picture was less abrupt than the weekly swings suggest. The four-week average of initial claims fell by 2,250 to 194,750. The average of continuing claims edged up by just 1,000 to about 1.71 million. Both weekly readings were lower than a year earlier.
The raw count also fell. State programs recorded 176,698 new applications before seasonal adjustment, down from 179,649 a week earlier and 216,026 in the comparable week of 2025. The adjusted series accounts for recurring changes in filing patterns, which is why its level and weekly change differ from the raw tally.
New filings remain historically low
Thursday’s initial claims figure came in below the 196,000 expected by economists polled by Reuters. It also matched a July reading that was the lowest since 1969, according to the outlet. That is a comparison of the number of applications, not a layoff rate. Initial claims are a useful early signal of job losses, but some workers who lose their jobs are ineligible for benefits or do not apply.
The latest numbers are preliminary. The Labor Department revised the previous week’s initial claims up by 1,000 to 195,000 and revised the prior continuing claims figure down by 6,000 to 1.695 million. Those changes illustrate why the four-week averages help put the latest report in perspective. The Associated Press reported that no week in 2026 had recorded more than 230,000 new claims through the latest release.
Continuing claims tell a different story
The rise in ongoing claims does not establish that people who lost work in October are taking longer to find jobs. The count can increase when more eligible people begin drawing benefits or when fewer existing claimants leave the rolls. People can also leave the rolls for reasons other than finding work, such as exhausting their benefits. A single weekly change cannot distinguish among those possibilities.
The year-over-year comparison puts that increase in context. Continuing claims were 244,000 lower than in the comparable week of 2025, and the insured unemployment rate held at 1.1 percent, compared with 1.3 percent a year earlier. The insured rate covers employment subject to unemployment insurance. It is a narrower measure than the national unemployment rate, which comes from a household survey and does not depend on whether a worker claims benefits.
Hiring and layoff plans add context
Other reports show why claims should be read alongside measures of hiring. The Bureau of Labor Statistics’ September turnover survey estimated about 7.5 million job openings at the end of the month. Employers made about 5.4 million hires during September, little changed from August, and reported about 1.6 million layoffs and discharges, also little changed. Openings are positions employers are trying to fill; hires count people who actually started work.
In a separate October tally, outplacement firm Challenger, Gray & Christmas counted 43,994 announced job cuts, up 2 percent from September but down 71 percent from October 2025. It also counted 444,692 announced hires, with a single large seasonal hiring plan from Amazon accounting for much of that total. The figures describe employer announcements, which may be implemented later and cannot be read as payroll gains or losses in October.
Output grew faster than hours worked
A separate productivity report released Thursday showed that nonfarm businesses produced more while hours worked barely increased in the third quarter. Output rose at a 3.7 percent annual rate, hours worked rose at a 0.1 percent rate, and output per hour increased at a 3.6 percent rate. The estimates are preliminary and cover the July-to-September quarter, rather than the late-October weeks in the claims release.
Labor costs per unit of output fell at a 0.7 percent annual rate in the quarter, even as hourly compensation rose. That can happen when output per hour rises faster than compensation. The same costs were still up 2.4 percent from a year earlier. The figures explain how businesses could expand production without a comparable rise in hours, but they do not identify what drove the change or tell employers how much staff they will need later in the year.
The October employment report, due Nov. 6, will provide a broader view through payroll growth and the national unemployment rate. Thursday’s figures answer a narrower question: relatively few people filed new claims at the end of October, even as more claims remained active in the previous week. The distinction matters for workers looking for a job as much as it does for businesses deciding whether to add staff.
