US Jobs Report September 2026: Time, Date and What to Watch

US jobs report September 2026: release time, date and official route

The direct answer is simple: the **US jobs report September 2026** is scheduled for **Friday 2 October at 08:30 ET**. The Bureau of Labor Statistics, or BLS, calls the release the Employment Situation. It covers September rather than October, and the scheduled time is in Eastern Time because that is the time zone used on the BLS release calendar.[1]

For readers following the release live, the official BLS Employment Situation page is the useful starting point at 08:30. The publication brings together the news summary, household- and establishment-survey tables, technical note, frequently asked questions and supplemental material. Reading those items together gives the definitions, industry detail and revisions that a headline alone cannot supply.[2] [4] [5]

This remains an upcoming-event guide, not a report of results. At the 26 September verification cut-off, BLS had not published September payrolls, the September unemployment rate, September wage measures or revisions to August and July. A release date establishes when information is due, not what the figures, financial-market response, policy decision or employer action will be, so none is assumed here.[1]

What the US jobs report September 2026 will include

The Employment Situation combines results from two separate monthly surveys. The Current Population Survey, often called the household survey, provides labour-force status, employment and unemployment measures. The Current Employment Statistics survey, also called the payroll or establishment survey, provides nonfarm employment, hours and earnings by industry.[2] [3]

That split is important when asking **what is in the US jobs report**. The household survey counts employed people, while the establishment survey counts nonfarm wage and salary jobs; a person holding more than one payroll job can therefore be counted more than once in the payroll total. Household-survey coverage also includes categories such as agricultural and unincorporated self-employed workers that do not appear in the nonfarm payroll measure.[3] [4]

The surveys also use distinct samples and reference periods. BLS says the household survey covers about 60,000 eligible households and generally refers to the calendar week containing the 12th, while the establishment survey draws on payroll reports from about 119,000 businesses and government agencies representing roughly 622,000 worksites and uses the pay period including the 12th. Different coverage, timing and methods mean that movements in the two surveys can differ without either result automatically being an error.[4]

September payrolls report 2026: how to read the headline jobs number

The headline nonfarm payroll figure is the over-the-month change in jobs on nonfarm payrolls. It includes employees who worked or received pay in the establishment-survey reference pay period, and it counts each payroll job held rather than every person only once. It does not include every kind of work measured by the household survey, including unincorporated self-employment and agricultural work.[3] [4]

A measured first read asks where a reported change occurred, rather than treating the total as a complete labour-market verdict. Table B-1 separates total nonfarm employment into sectors and selected industries, and the release also includes private-sector hours and earnings measures. Readers can use those tables to see whether changes are widespread or concentrated and then compare the payroll result with the household-survey measures published in the same release.[2] [4]

August is the latest published baseline before the pending September release. BLS reported that total nonfarm payroll employment rose by 162,000 in August, following revised gains of 31,000 in June and 21,000 in July. Those are already published prior-month figures; they provide context for reading the new tables but do not establish or forecast the September result.[2]

The release process also matters for timing. The establishment survey reports estimates from its sampled payroll records, and the initial estimate is subsequently updated as more reports are received. A reader comparing news coverage should check that each account is referring to the same release vintage, especially when it quotes June or July figures after the October publication has revised them.[3] [4]

September 2026 unemployment rate release: participation matters too

The **September 2026 unemployment rate release** will be produced from the household survey. BLS defines unemployed people as people with no employment in the reference week who are available for work and have actively looked for work in the preceding four weeks; people on temporary layoff who expect recall are also included. The unemployment rate is the number of unemployed people as a share of the labour force.[4]

That definition does not mean that everyone who says they want a job is in the official unemployed count. People who want work but are not currently searching are outside that count, while BLS separately publishes information about people not in the labour force who want a job and alternative measures of labour underutilisation. Checking the relevant tables helps keep the headline rate in its defined context rather than extending it beyond what it measures.[4] [5]

The **US labour force participation rate September 2026** is another useful companion measure because it is the labour force as a share of the population. The employment-population ratio instead shows employed people as a share of the population. For the latest published month, August, BLS reported a 4.1% unemployment rate, a 61.6% participation rate and a 59.1% employment-population ratio; September values were not available at the cut-off.[2] [4]

A change in unemployment can be read alongside participation and the employment-population ratio because all three are household-survey measures with related but different denominators or numerators. That comparison does not supply a single mechanical explanation for a monthly move, but it makes the report’s underlying labour-force measures visible to readers instead of reducing the household survey to one percentage.[4] [5]

Nonfarm payrolls and average hourly earnings: wages, hours and sectors

Readers following **nonfarm payrolls and average hourly earnings** have more to examine than the payroll headline. The establishment survey’s B tables report average hourly and weekly earnings, average weekly hours and overtime measures alongside employment by industry. BLS describes these hours and earnings series as private-sector measures available for all employees and for production and nonsupervisory employees.[2] [4]

Average hourly earnings and average weekly hours answer different questions in the published tables, so neither is interchangeable with the payroll count. A practical reading sequence is to identify the series label, employee group and period shown in the table before comparing it with another month. This keeps the distinction between earnings, hours and employment intact and avoids attaching a September label to an August value.[2] [4]

For context only, BLS said average hourly earnings for all employees on private nonfarm payrolls increased 10 cents, or 0.3%, to $37.75 in August and were 3.1% higher than a year earlier. BLS also said the average workweek edged up to 34.4 hours. Those are the latest published values, not September data, and the October release may report different figures.[2]

Industry tables help explain the composition of a payroll total. In the August release, BLS identified gains in food services and drinking places, local government education, manufacturing and health care, while information employment declined. The September report will contain its own industry estimates, so the August pattern should be treated as prior context rather than an assumption about which sectors will add or lose jobs next.[2]

Why revisions can change the jobs report story

The first payroll estimate is not a final historical verdict. BLS routinely revises the previous two monthly estimates as additional reports arrive from businesses and government agencies and as seasonal adjustment factors are recalculated. The agency says the estimates are generally final after two successive revisions, apart from the separate annual benchmark process that uses unemployment-insurance tax records.[3] [4] [5]

The latest release illustrates the practical importance of that routine. BLS revised June payroll growth up by 11,000 and July by 44,000, leaving the combined level 55,000 higher than previously reported. When the September release arrives, readers should look for the stated revisions to August and July as well as the new September estimate; otherwise, an account may describe only part of the change in the published employment history.[2]

Sampling uncertainty is another reason not to describe one monthly movement as a definitive labour-market turn. BLS gives an approximate 90% confidence interval of plus or minus 122,000 for the monthly change in total nonfarm employment. Its technical note gives an approximate plus or minus 0.3 percentage-point interval for a monthly unemployment-rate change when the rate is around 6.0%.[4]

Seasonal adjustment is intended to account for recurring patterns such as holidays, weather and the school calendar so month-to-month comparisons are more useful. It does not make a sample estimate certain, and BLS says it recalculates relevant factors concurrently as new data arrive. The technical note and revision lines are therefore part of the report’s evidence, not fine print to ignore after the top-line figures appear.[4]

What to check before and after the 2 October release

Before Friday, it is useful to distinguish supporting indicators from the Employment Situation itself. BLS scheduled August Job Openings and Labor Turnover Survey data for Tuesday 29 September at 10:00 ET. JOLTS measures openings, hires, quits and separations for an earlier month, so it can supply context but cannot substitute for September’s payroll and household-survey figures.[6]

The weekly unemployment-insurance series is also a separate measure. The Department of Labor reported 197,000 seasonally adjusted initial claims for the week ending 19 September and a four-week average of 202,250. Its technical note identifies claims as weekly administrative data and cautions that seasonal adjustment can leave the series volatile, which is a reason to keep the series distinct from the monthly employment estimates.[7]

Reuters independently reported the 24 September claims figures and noted economists’ caution about seasonal-adjustment issues around moving holidays such as Labour Day. That reporting reinforces a basic pre-release discipline: claims and JOLTS may be useful context, but neither can be converted into a promised September payroll, unemployment-rate or wage result. The pending BLS release remains the source for those September measures.[6] [7] [8]

At 08:30 ET on 2 October, begin with the official BLS summary and confirm the reference month and any revision lines. Then read payrolls with unemployment, participation, the employment-population ratio, earnings, hours and the industry tables, followed by the technical note if a one-month change appears unusually large or small. That sequence uses the material BLS publishes in the Employment Situation and leaves room for later revisions rather than presenting a first reading as final.[1] [2] [4] [5]