Jobless claims at 197,000: 36,000 fewer than a year earlier
Initial claims were 197,000 in the week ending October 3, 2026, the 4-week average 198,000, continuing claims 1,716,000. What each counts and what it cannot.
- What happened
- The Department of Labor reported on October 8, 2026 that seasonally adjusted initial claims for unemployment insurance were 197,000 in the week ending October 3, 2026, down 2,000 from a revised 199,000. The 4-week moving average fell to 198,000. Continuing claims for the week ending September 26, 2026 were 1,716,000.
- Why
- An initial claim is filed by a person who has just lost a job, so the series counts layoffs as they happen rather than a month later. It is published every Thursday, which makes it the fastest regular reading on the labour market, and also the noisiest, which is why the 4-week average exists.
- The numbers that matter
- 197,000Initial claims, week ending October 3, 2026, seasonally adjusted198,0004-week moving average of initial claims, week ending October 3, 20261,716,000Insured unemployment (continuing claims), week ending September 26, 2026233,000Initial claims in the comparable week of 2025
- Why it matters
- Claims measure the layoff side of the labour market. Payroll growth of 29,000 in September 2026 with claims under 200,000 says employers are not cutting staff, but are also not hiring much. For a household, that is the difference between the risk of losing a job and the difficulty of finding one.
- What happens nextas we saw it on 13 Oct 2026
- The claims report is issued every Thursday at 8:30 a.m. Eastern; the next one, covering the week ending October 10, 2026, is due on Thursday, October 15, 2026. The Employment Situation for October 2026 follows on the Bureau of Labor Statistics schedule in early November 2026.
Six weeks of readings sit between 197,000 and 207,000; the week-to-week moves are small next to the 36,000 gap against the same week of 2025.
| Aug 29 | 207 |
|---|---|
| Sep 5 | 207 |
| Sep 12 | 198 |
| Sep 19 | 198 |
| Sep 26 (rev.) | 199 |
| Oct 3 (adv.) | 197 |
The number of Americans filing a first claim for unemployment benefits in the week ending October 3, 2026 was 197,000. In the comparable week of 2025 it was 233,000. The labour market that produced a September payroll gain of only 29,000 is, on this measure, laying off fewer people than it was a year earlier, not more.
Below is what the weekly claims report counts, why its headline is smoothed into a 4-week average, and how it fits beside the September jobs report published on October 2, 2026.
The numbers released on October 8, 2026
The Department of Labor's release of Thursday, October 8, 2026 (USDL 26-1561-NAT) put seasonally adjusted initial claims at 197,000, “a decrease of 2,000 from the previous week's revised level.” The prior week, ending September 26, was revised up by 2,000, from 197,000 to 199,000. The 4-week moving average was 198,000, down 2,500 from a revised 200,500.
Seasonally adjusted insured unemployment, the count of people who had already filed and were still claiming, was 1,716,000 in the week ending September 26, 2026, an increase of 17,000 from a revised 1,699,000. Its own 4-week average was 1,711,000, down 12,250 from 1,723,250. The seasonally adjusted insured unemployment rate was 1.1%, unchanged on the week; a year earlier it was 1.3%.
| Measure | Week ending | Level | Change on week | Comparable week of 2025 |
|---|---|---|---|---|
| Initial claims (advance) | Oct 3, 2026 | 197,000 | −2,000 | 233,000 |
| 4-week average, initial claims | Oct 3, 2026 | 198,000 | −2,500 | 227,500 |
| Insured unemployment | Sep 26, 2026 | 1,716,000 | +17,000 | 1,929,000 |
| 4-week average, insured unemployment | Sep 26, 2026 | 1,711,000 | −12,250 | — |
| Insured unemployment rate | Sep 26, 2026 | 1.1% | unchanged | 1.3% |
Initial claims: 233,000 − 197,000 = 36,000, a fall of 15.5% (36,000 ÷ 233,000). Insured unemployment: 1,929,000 − 1,716,000 = 213,000, a fall of 11.0% (213,000 ÷ 1,929,000). Inputs are the October 8, 2026 release's own current and prior-year columns.
How the series is built
The release defines its own terms. “An initial claim is a claim filed by an unemployed individual after a separation from an employer.” Continued claims, it says, “reflect a good approximation of the current number of insured unemployed workers filing for UI benefits.” The first series counts people entering unemployment; the second counts the stock of people still in it and still eligible.
The release states that “advance claims are reported by the state liable for paying the unemployment compensation” and that “the following week initial claims and continued claims are revised based on a second reporting by states.” It adds that “advance claims are not directly comparable to claims reported in prior weeks.” That is why the September 26, 2026 figure moved from 197,000 to 199,000 between the October 1 and October 8 reports.
Unadjusted initial claims in the week ending October 3, 2026 were 170,333, up 11,994 (7.6%) from 158,339 the week before. The release says the seasonal factors “had expected an increase of 13,725 (or 8.7 percent).” Because the actual rise was smaller than the expected seasonal rise, the adjusted figure fell. The unadjusted count in the comparable week of 2025 was 207,124.
The adjustment removes the pattern that repeats every year so that, in the release's words, “trend and cycle developments [are] easier to spot.” It does not remove weather, strikes, a single large employer's layoff, or a state's processing backlog. Those show up as one-week spikes and dips.
The release prints its denominator: “Most recent week used covered employment of 153,732,307 as denominator,” a total it attributes to the Bureau of Labor Statistics. Unadjusted insured unemployment of 1,504,800 for the week ending September 26, 2026 divided by that figure reproduces the printed 1.0% unadjusted rate. The seasonally adjusted rate is 1.1%. Both are far below the 4.2% unemployment rate in the September 2026 Employment Situation because they measure different things: claimants drawing insured benefits against everyone the household survey counts as jobless and looking.
Why the 4-week average exists
The 4-week moving average is not defined in the release; it is simply the arithmetic mean of the latest four weekly figures. Its purpose is to stop a single week's noise from being read as a turn in the cycle.
The four seasonally adjusted weeks ending September 12, 19, 26 and October 3, 2026 were 198,000, 198,000, 199,000 and 197,000. Their sum is 792,000; divided by four, 198,000. The previous average of 200,500 used the weeks ending September 5 to 26: 207,000 + 198,000 + 198,000 + 199,000 = 802,000, ÷ 4 = 200,500. The week that dropped out (207,000) was higher than the week that came in (197,000), which is the whole reason the average fell by 2,500.
Because each week stays in the average for four reports, a single spike is diluted by three-quarters on the day it arrives and then lifts the average for three more weeks after it has passed. The average lags; that is the price of smoothing. It smooths timing effects, such as a holiday falling in one week rather than the next. It does not smooth revisions: when the September 26, 2026 week was revised up by 2,000, the previous average moved with it to the 200,500 the October 8 release reports.
Claims beside the September jobs report
The Employment Situation for September 2026, released by the Bureau of Labor Statistics on October 2, 2026, reported that total nonfarm payroll employment “changed little in September (+29,000)” and that the unemployment rate was 4.2%, with 7.1 million people unemployed. July was revised from +21,000 to −10,000 and August from +162,000 to +133,000, a combined 60,000 lower. The labour force participation rate was 61.8% and average hourly earnings rose 5 cents (0.1%) to $37.81.
Claims count separations that lead to a benefit filing. Payrolls count the net of hiring and separations. A month in which few people are laid off but few are hired produces exactly the pattern on display: claims under 200,000 for four consecutive weeks and a payroll gain of 29,000. The claims data cannot tell you about hiring at all. That is what the JOLTS openings and hires data are for, and they arrive with a longer lag.
Insured unemployment of 1,716,000 (week ending September 26, 2026) set against 7.1 million unemployed in the September 2026 household survey gives 24.2%. The periods differ, a single week against a monthly survey reference week, and the counts come from different systems, so treat the ratio as rough scale. It shows why a 1.1% insured unemployment rate and a 4.2% unemployment rate are not in conflict: most unemployed people are not drawing insured benefits, because they have exhausted eligibility, never qualified, or are new entrants.
Continuing claims rose 17,000 in the week ending September 26, 2026 while initial claims fell. Fewer people are entering, but those inside are not leaving as fast. One week is not a trend, as an earlier piece on this site argues for monthly data; the point is sharper for weekly data.
What it means for your money
For a household, the claims report answers one question: how many people lost a job and filed for benefits in the week just gone. It says nothing about how long they will take to find the next one. The two questions have different answers in the data above.
- Layoff risk is the initial-claims number. At 197,000 a week it is 15.5% below the same week of 2025. The aggregate says nothing about any one job, but it is lower than it was a year earlier.
- Duration risk is the continuing-claims number. At 1,716,000 it is 11.0% below a year earlier, but it rose on the week while initial claims fell. Set alongside payroll growth of 29,000, this is the figure that bears on how long an emergency fund would need to last.
- The weekly figure is a draft. A number quoted on a Thursday may be 2,000 different the following Thursday, as the September 26, 2026 week was.
- The safety net varies by state. The release lists the highest insured unemployment rates by state for the week ending September 19, 2026, from New Jersey at 2.0% to Illinois at 1.3%. Eligibility, benefit level and duration are set state by state.
The series to keep reading is the 4-week average alongside continuing claims. The first says whether layoffs are rising; the second says whether the laid-off are being re-employed. The monthly unemployment rate is published weeks later than either.
Sources
- Unemployment Insurance Weekly Claims (USDL 26-1561-NAT), week ending October 3, 2026, released October 8, 2026US Department of Labor, Employment and Training Administration · Oct 12, 2026 · Primary source
- The Employment Situation — September 2026, released October 2, 2026US Bureau of Labor Statistics · Oct 12, 2026 · Primary source
Related analysis
September added 29,000 jobs, and July turned negative again
Payrolls rose 29,000 in September 2026 and unemployment edged up to 4.2%. Revisions cut July and August by 60,000, putting July back below zero.
August 2026: one job opening per unemployed person, quits at 1.9%
JOLTS counted 7,079,000 openings on the last business day of August 2026, against 7,031,000 unemployed. The quits rate held at 1.9%.
August added 162,000 jobs. The average of the prior twelve months was 31,000.
The strongest hiring month in a year sits on top of a twelve-month average five times smaller. Both numbers are true, and only one of them is a trend.
The unemployment rate has six versions. Here is August 2026 in each
The headline 4.1% is U-3. The same BLS release also put the rate at 1.8%, 1.9%, 4.4%, 5.1% and 7.7%. Who each version counts, and what the gaps have meant.