The Employment and Training Administration at the U.S. Department of Labor reported an advance estimate of 197,000 seasonally adjusted initial unemployment insurance claims for the week ended 26 September in its weekly report, released on 1 October 2026. That was 1,000 below the previous week's revised level of 198,000. The four-week moving average fell to 200,000. Because the prior week's average was revised up from 202,250 to 202,500, the new average was 2,500 below the revised comparison. The release date is 1 October; the measured week ended in late September. These indicators describe unemployment insurance claim activity, not payroll employment or the total number of unemployed people.

An initial claim is filed by a person separated from an employer who asks a state unemployment insurance programme to determine basic eligibility. It is therefore not a count of everyone laid off in a week. People who do not file, are not eligible, or file at another time are not captured by that weekly count. The Department's technical notes say the data come from weekly administrative reports supplied by state programmes and that advance weekly figures are later revised using subsequent reporting. A prior value first published as 197,000 was revised up to 198,000 in this release. That example shows why readers should distinguish an initial estimate from the later revised level when assessing small week-to-week movements.

The four-week average helps smooth short-term volatility. For the week ended 26 September it stood at 200,000 claims, 2,500 below the revised 202,500 average a week earlier. The report's weekly table lists seasonally adjusted initial claims of 198,000 for both 19 September and 12 September; the latest four-week average puts the most recent reading in a wider frame. But a moving average still depends on administrative claims and revisions. It does not, by itself, explain changes in hiring, layoffs, vacancies or the unemployment rate. Initial claims are best read as an early indicator about one part of labour-market activity, alongside other evidence rather than as a complete labour-market account.

Continuing insured unemployment is a separate measure for a different week. The seasonally adjusted level for the week ended 19 September was 1,701,000, down 11,000 from the revised previous week. Its four-week average fell by 18,500 to 1,723,750, while the insured unemployment rate remained 1.1%. These are not measured for the same week as initial claims: new claims refer to the week ended 26 September, while continuing insured unemployment refers to the week ended 19 September. Continued claims reflect weekly ongoing benefit claims in the insured system; they should not be described as a census of everyone without a job.

The report also recorded 156,738 initial claims without seasonal adjustment, a fall of 7,979 or 4.8% from the previous week. Seasonal factors had expected a 4.1% decrease for that period. This raw count should not be compared directly with the seasonally adjusted estimate of 197,000. The Department says states use seasonal factors updated at the beginning of each year, and historical administrative data may also be revised alongside the new factors provided by the Bureau of Labor Statistics. The technical notes stress that weekly administrative claims are difficult to adjust seasonally and can be volatile. The series helps monitor short-term conditions, but it is not a comprehensive or definitive measure of employment.