Unemployment Rate Was 4.1% in August 2026: What the Rate Isn't Telling You
The U.S. unemployment rate was 4.1% in August 2026, the same as in July 2026. U-6, the broadest U.S. Bureau of Labor Statistics measure of labor underutilization, was 7.7% in August 2026, little changed from 7.9% in July 2026. For August 2026, U-6 minus U-3 is 3.6 percentage points. Nine labor market indicators from BLS, the U.S. Department of Labor, Challenger, and Indeed follow, each with its own period and source-file date.
These are saved source readings, not a live release check. Observation periods differ; source files are dated 2026-09-22–2026-09-29. Newer releases may revise these values.
What Is the State of the U.S. Job Market in 2026?
The unemployment rate was 4.1% in August 2026, according to the U.S. Bureau of Labor Statistics. A year earlier it was 4.3%. It is the lowest rate since June 2025. Its complete 2019 average was 3.68%. A headline unemployment rate cannot by itself show that households are financially secure.
The broader U-6 measure was 7.7% in August 2026. For August 2026, U-6 minus U-3 is 3.6 percentage points. Continued weeks claimed were 1.72 million for the week ending Sep 12, 2026, the same as the week before. That counts weeks of unemployment claimed in state programs, not unique people or how long anyone stays unemployed.
The quits rate was 1.9% in August 2026, the same as in July 2026. Challenger, Gray & Christmas counted 52,900 announced job cuts in August 2026, down from 86,000 in August 2025. The American Distress Index includes both unemployment and initial claims in Labor. American Default Research reports an initial-claims/unemployment association with 1 quarter with r = 0.85 in its analyzed sample; that does not establish causation or a guaranteed forecast.
Key Statistics at a Glance
The American Distress Index currently reads 47.0 (Typical). The composite itself sits higher than 44% of all published quarters since 2005. The unemployment rate and initial jobless claims are the two inputs to the index's Labor domain — claims are the higher-frequency of those two inputs, not a count of all income loss. Job loss can increase payment risk without determining an individual household's outcome — if you've experienced an income disruption, our guide on what to do when you're behind on payments covers immediate steps. Separately, the personal saving rate was 4.1% in August 2026, and 37% of adults in the 2025 SHED said they would not pay a $400 emergency expense entirely with cash or its equivalent. Those measures do not identify the same households or determine how any one household would absorb an income shock.
What Is the Real Unemployment Rate?
U-3 is the official unemployment rate, not a false measure that U-6 replaces. It includes people without work who are available and actively searched in the past four weeks, plus people on temporary layoff who expect recall and need not be searching. U-6 adds all marginally attached workers, including discouraged workers, and people working part-time for economic reasons. Its denominator also adds marginally attached workers to the civilian labor force. See the BLS definitions and labor force participation.
For August 2026, U-6 minus U-3 is 3.6 percentage points. The gap between the two 2019 annual averages was 3.5 percentage points. Because their populations and denominators differ, subtracting the rates is not a count of additional unemployed people. The chart aligns identical months. Missing months remain gaps; it does not interpolate a reading.
U-3 vs. U-6, Saved Monthly Readings (2005–2026-08)
Source: Bureau of Labor Statistics data retrieved via FRED (UNRATE, U6RATE). Monthly, seasonally adjusted.
Full data: Unemployment Rate (U-3) · U-6 Underemployment
How Many People Are Filing for Unemployment Each Week?
Initial claims are applications for unemployment insurance in state programs, not all job losses or all unemployed people. Along with U-3, they feed the index's Labor domain. Claims are released weekly; the monthly jobs report measures a different population and reference period. See the DOL data retrieved via FRED series notes.
The seasonally adjusted count was 197,000 for the week ending Sep 19, 2026, little changed from 198,000 the week before. Its complete 2019 average was 217,712. The chart averages available weekly observations in each calendar month, not people over a month. Partial months are included and can change as observations arrive. For September 2026, 3 of 4 calendar week-ending dates are present.
Initial Claims: Available-Week Monthly Means (Since 2019)
Source: Department of Labor data retrieved via FRED (ICSA). Seasonally adjusted. Means use available week-ending dates; partial months are included, missing months remain gaps.
Full data: Initial Unemployment Claims · Continuing Claims
Labor Market Dashboard: All Indicators Compared
Each change is the comparison stored with the indicator: the period before for seasonally adjusted series, and the same period a year earlier for Challenger and Indeed. BLS calls a move smaller than its significance threshold little changed, and so does this table. Payrolls are jobs, not people. Rate changes are percentage points. The 2019 average needs every calendar observation in 2019 and is shown for reference. The last column says where the latest reading stands in the stored history, with the year the series starts.
| Indicator | Saved Reading | Change | Complete 2019 Average | Where It Stands |
|---|---|---|---|---|
| Unemployment Rate (U-3)ADI | 4.1% August 2026 · file 2026-09-22 | The same as in July 2026 | 3.68% | Lowest since June 2025 |
| U-6 Underemployment | 7.7% August 2026 · file 2026-09-22 | Little changed from 7.9% in July 2026 | 7.15% | Lowest since June 2025 |
| Initial Claims (weekly)ADI | 197,000 Sep 19, 2026 · file 2026-09-24 | Little changed from 198,000 the week before | 217,712 | 72nd lowest of 3,116 stored readings since 1967 |
| Continued Weeks Claimed (weekly) | 1.72 million Sep 12, 2026 · file 2026-09-24 | The same as the week before | 1.68 million | 358th lowest of 3,115 stored readings since 1967 |
| Nonfarm Payrolls (jobs; monthly level) | 159.08 million August 2026 · file 2026-09-22 | Up from 158.91 million in July 2026 | 150.91 million | Highest in the series, which starts in 1939 |
| JOLTS Quits Rate | 1.9% August 2026 · file 2026-09-29 | The same as in July 2026 | 2.32% | 88th lowest of 307 stored readings since 2000 |
| Part-Time (Econ Reasons) | 4.39 million August 2026 · file 2026-09-22 | Down from 4.8 million in July 2026 | 4.41 million | Lowest since December 2024 |
| Indeed Job Postings Index | 103.5 Sep 18, 2026 · file 2026-09-23 | Little changed from 102.7 a year earlier | Unavailable: incomplete 2019 coverage | 683rd lowest of 2,422 stored readings since 2020 |
| Challenger Announced Cuts (monthly) | 52,900 August 2026 · file 2026-09-25 | Down from 86,000 in August 2025 | Unavailable: incomplete 2019 coverage | Not ranked: the stored series is sparse |
All labor indicators: Nonfarm Payrolls · Part-Time for Economic Reasons · Challenger Layoffs
How Have Job Postings and Quits Changed?
The Indeed Job Postings Index was 103.5 on Sep 18, 2026, little changed from 102.7 a year earlier. It tracks postings on Indeed, not hires or all U.S. job openings; the official openings count is the BLS JOLTS survey. The index is set to 100 on Feb 1, 2020.
The quits rate was 1.9% in August 2026, the same as in July 2026. Its complete 2019 average was 2.32%. These series do not identify why workers leave jobs or show that fewer quits caused household hardship. For a household facing rising essential costs, the practical question is whether its own income can cover its payments. The chart selects the last available daily observation in each month, not a monthly average; the final point may precede month-end.
Indeed Job Postings: Last Available Daily Reading per Month
Source: Indeed data retrieved via FRED (IHLIDXUS). Feb 1, 2020 = 100. Seasonally adjusted 7-day trailing average. Each point keeps its actual selected date; missing months are gaps.
Full data: Indeed Job Postings Index · JOLTS Quits Rate
Continuing Claims: How Many Weeks of Unemployment Are Being Claimed?
Continuing claims count the weeks of unemployment claimed in state programs during the report week, after an initial claim. The Department of Labor says they do not count unique people, and they include weeks claimed but not paid. The seasonally adjusted count was 1.72 million for the week ending Sep 12, 2026, the same as the week before. Its complete 2019 average was 1.68 million. Entries, exits, eligibility and benefit exhaustion all affect the count. This series alone cannot show how long job searches last or how quickly the labor market absorbs job losers. See the DOL data retrieved via FRED definition.
The chart uses available-week monthly means, including partial months. For September 2026, 2 of 4 calendar week-ending dates are present. The headline card is one weekly observation, not that monthly mean.
Continuing Claims: Available-Week Monthly Means (Since 2019)
Source: Department of Labor data retrieved via FRED (CCSA). Seasonally adjusted. Available-week means include partial months; missing months remain gaps.
Hidden Slack: What the Unemployment Rate Doesn't Show
The measures answer different questions: unemployment status, unemployment-insurance claims, jobs on payrolls, voluntary quits, advertised postings, and announced cuts. Their periods and release schedules differ. Reading them together adds context, but no single change shows that all conditions are getting worse.
Other pages report an aggregate personal saving rate of 4.1% in August 2026, credit card balances of $1.26 trillion in Q2 2026, and 6% of eligible participants in Vanguard-administered retirement plans taking a hardship withdrawal in 2025. Those measures do not identify the same people, show that one event followed another, or establish a path from thin savings to default.
Explore current indicator-pair research and its validation boundaries →Data Sources and Methodology
CPI measurement basis: Every CPI reading and historical comparison on this page uses its source-owned basis unless explicitly identified otherwise: Cumulative percent change since January 2020 computed from the seasonally adjusted BLS Food-at-Home CPI (CUSR0000SAF11); seasonally adjusted values may be revised.
Bureau of Labor Statistics
Unemployment rate (U-3), U-6 underemployment, nonfarm payrolls, part-time for economic reasons, and JOLTS quits rate. Monthly releases have different lags and are revised. The household survey measures people in the reference week; payrolls measure jobs in the pay period including the 12th; quits cover the month. Seasonally adjusted. Retrieved via FRED (UNRATE, U6RATE, PAYEMS, LNS12032194, JTSQUR). See BLS payroll technical notes and JOLTS definitions.
The values on this page keep their saved observation periods and file dates. Later releases can revise a published comparison.
Department of Labor
Initial claims and continued weeks claimed under state unemployment insurance programs in the 50 states, DC, Puerto Rico and the Virgin Islands. Weekly releases normally arrive Thursday, and continuing claims cover an earlier week than initial claims. Counts cover unemployment insurance, not everyone out of work. Seasonally adjusted, subject to revision. Retrieved via FRED (ICSA, CCSA). Chart means use available weeks; a missing week is not a zero.
Challenger, Gray & Christmas
Monthly job cut announcement report tracking planned cuts by company, sector, and stated reason. Challenger's reports do not say the counts are seasonally adjusted, so it compares each month with the same month a year earlier, and so does this page. It captures announcements rather than completed terminations. The saved series is sparse and does not establish annual totals or historical records.
Indeed / FRED
Indeed Job Postings Index (IHLIDXUS) measures job postings on Indeed.com, indexed to February 1, 2020 = 100, seasonally adjusted, as a 7-day trailing average. The source is daily; this page selects the last available daily reading per month. It covers one platform, not a census of U.S. job openings.
Frequently Asked Questions
What is the current U.S. unemployment rate?
The U.S. unemployment rate was 4.1% in August 2026, the same as in July 2026. A year earlier it was 4.3%. It is the lowest rate since June 2025. The source is the U.S. Bureau of Labor Statistics, retrieved via FRED (file dated 2026-09-22). U-6, the broadest BLS measure of labor underutilization, was 7.7% in August 2026, little changed from 7.9% in July 2026. For August 2026, U-6 minus U-3 is 3.6 percentage points. U-6 adds all marginally attached workers and people working part time for economic reasons, not just discouraged workers.
How many people file for unemployment each week?
The seasonally adjusted count of initial claims was 197,000 for the week ending Sep 19, 2026, little changed from 198,000 the week before. That counts claims filed with state unemployment insurance programs, not people laid off or all job losses. Its complete 2019 average was 217,712. The unemployment rate and initial claims are the two inputs to the American Distress Index's Labor domain.
Are layoffs increasing in 2026?
Challenger, Gray & Christmas counted 52,900 announced job cuts in August 2026, down from 86,000 in August 2025. Challenger does not seasonally adjust its counts, so it compares each month with the same month a year earlier, and so does this page. Announcements are planned cuts, not completed layoffs, and one large announcement can move a month's total a long way.
What does the JOLTS quits rate tell us?
The JOLTS quits rate was 1.9% in August 2026, the same as in July 2026. Its complete 2019 average was 2.32%. BLS measures quits during the month as a share of nonfarm payroll employment; quits exclude retirements and transfers. The rate alone does not show why people quit, whether they found better work, or a household's income prospects.
How does unemployment connect to the American Distress Index?
The unemployment rate and initial jobless claims are the two inputs to the American Distress Index's Labor domain. Income loss can make payments harder to sustain, but these aggregate indicators do not show which household will miss a payment. The index currently reads 47.0 (Typical). The composite itself sits higher than 44% of all published quarters since 2005.