The jobs report: what it is, how it’s calculated, and what it means
How many jobs were added, and who’s out of work?
On the first Friday of most months, the government publishes the jobs report: how many jobs employers added, the unemployment rate, and how fast pay is rising. It’s the most-watched number on the economy, and one of the most misread.
Latest data marks the parts that update with each report. Everything else explains how it works.
Is the economy still adding jobs?
Jobs added each month
Change in payroll jobs, thousands, since January 2022. The shaded band is the estimate’s margin of error (±122K): bars inside it are “little changed.”
Employers added about 377,000 jobs a month in 2022 and 210,000 in 2023, as the economy refilled jobs lost in 2020. Gains slowed to 122,000 a month in 2024 and about 10,000 in 2025, with several months below zero. In 2026 so far they have averaged about 68,000.
Which industries are adding jobs?
Over the past 12 months, health care added about 372,000 jobs and social assistance about 149,000: together more than the 496,000 the whole economy added. Government shed about 216,000, information about 120,000 and financial activities about 107,000.
| Industry | Sep 2026 | Bar | Past year |
|---|---|---|---|
| Health care and social assistance | +23K | +520K | |
| Construction | +11K | +109K | |
| Leisure and hospitality | +10K | +62K | |
| Manufacturing | +9K | +40K | |
| Transportation and warehousing | +8K | −10K | |
| Other services | +6K | +34K | |
| Retail trade | +6K | +27K | |
| Wholesale trade | +5K | +27K | |
| Utilities | +1K | +10K | |
| Mining and logging | −2K | +2K | |
| Private education | −3K | −4K | |
| Financial activities | −7K | −107K | |
| Professional and business services | −9K | +123K | |
| Information | −10K | −120K | |
| Government | −17K | −216K |
Is unemployment going up?
Unemployment, pay growth and who’s working, monthly since 2000.
Unemployment is low, and drifting up
Unemployment rate and U-6 (adds people stuck in part-time work or no longer looking), percent, monthly. Shaded bands are recessions.
Unemployment peaked at 14.8% in April 2020 and fell to 3.4% in April 2023, its lowest in this series. It has drifted up since and held between 4.1% and 4.3% since March 2026. U-6, the broader measure that adds people working part time who want more hours and people who have stopped looking, was 7.6%. The gap in October 2025 is real: the household survey wasn't collected during the government shutdown.
Seasonally adjusted. Source: BLS Employment Situation via the BLS Public Data API.
What does this mean for my job, or my job search?
Signals, not verdicts. These hold whatever the month: the jobs report describes the whole country, not your job.
When job gains are small, searches take longer
Fewer new jobs means more competition for each one. If your search is dragging, plan for a longer runway.
When unemployment drifts up, employers get choosier
It can be a good time to strengthen where you are and keep your network warm before you need it.
When pay growth slows, raises take a case
Smaller budgets mean the people who come with evidence of their impact tend to get more of them.
When your industry is shrinking
National numbers can hide a sector that’s losing jobs. Know your industry’s direction before it shows up in your team.
None of these tell you whether to stay or go. They describe the field you’re playing on, so your next move starts from what’s real.
What is the jobs report, and how does it work?
The jobs report is two surveys published together. One counts jobs by asking employers. The other counts people by asking households. That’s why one report can say employers added jobs while the number of people working fell: they’re measuring different things, from different samples.
Where do the jobs number and the unemployment rate come from?
The payroll survey
119,000 businesses and agencies, about 622,000 worksites
Counts jobs on payrolls, so someone with two jobs counts twice. Leaves out the self-employed and farm workers.
Gives you: jobs added, jobs by industry, hours, average hourly pay.
The household survey
60,000 households, interviewed by the Census Bureau
Counts people: who worked, who looked for work, and who did neither. Includes the self-employed.
Gives you: the unemployment rate, participation, who’s working by age, sex, race and education.
Who counts as unemployed?
Not everyone without a job. To count as unemployed, you have to pass all three tests.
No job
You didn’t work at all during the survey week (the week including the 12th).
Available
You could have started a job that week.
Looked
You actively searched in the past four weeks: applied, contacted employers, used a job site.
Miss a test and you’re “not in the labor force”, not unemployed. In September 2026, 5.8 million people outside the labor force said they wanted a job, and 4.5 million were working part time because they couldn’t get full-time hours. The broader U-6 rate counts both groups: 7.6%.
How is the jobs report calculated?
- Two surveys2every month
One asks employers about jobs. The other asks households about people.
- Employers report119,000businesses and agencies
About 622,000 worksites report payroll jobs, hours and pay for the pay period including the 12th.
- Households answer60,000households
Census Bureau interviewers ask who worked, who looked for work, and who didn’t.
- Estimate±122Kmargin on job gains
Samples are scaled up to the whole economy and adjusted for seasonal swings.
- Publish1stFriday, usually
At 8:30 a.m. Eastern, about a week after the month ends.
- Revise2more looks
The next two reports revise each month, and every February brings a benchmark revision.
Covers the 50 states and DC. The payroll survey covers nonfarm jobs; the household survey covers the civilian population aged 16 and over.
How to read the jobs report
Read the trend, not the month
A single month’s job gain can be off by 122,000 either way. Three-month averages tell you more.
Expect revisions
Each month is revised twice as late reports come in. The first number is a first draft.
The two surveys can disagree
Payroll jobs and household employment are measured differently, so they can move apart for a while.
Unemployment isn’t everyone without a job
You only count as unemployed if you looked for work in the past four weeks. Others are “not in the labor force.”
Pay growth isn’t a raise
Average pay rises when lower-paid jobs disappear, too, and it says nothing about prices.
It’s national
State and metro unemployment come out later in the month, from a different program (LAUS).
The jobs report vs. JOLTS and local unemployment
JOLTS
The jobs report gives the net change in jobs. JOLTS shows the movement underneath: openings, hires, quits and layoffs. Out about a month later.
Local unemployment
Unemployment for every state and metro comes from a related program (LAUS), released later in the month.
Inflation
Pay growth only means something next to prices. The Consumer Price Index comes out the following week.
Private reports
ADP publishes its own private-payroll count two days earlier. It often differs from the official number.
Common questions about the jobs report
What is the jobs report?
The jobs report, officially the Employment Situation, is the U.S. Bureau of Labor Statistics’ monthly release on the job market. It combines a survey of about 119,000 employers (how many jobs, hours and pay) with a survey of about 60,000 households (who is working, unemployed or not looking). It usually comes out on the first Friday of the month at 8:30 a.m. Eastern.
When is the next jobs report?
The next jobs report is scheduled for Nov 6, 2026 at 8:30 a.m. Eastern. It will cover October 2026.
How is the unemployment rate calculated?
It’s the number of unemployed people divided by the labor force (everyone working or looking). In September 2026 that was 7.1 million out of 170.3 million, or 4.2%. To count as unemployed you must have no job, be available to work, and have looked for work in the past four weeks.
Why do the job numbers get revised?
The first estimate is based on the employers who have reported by the deadline. As more reports arrive, BLS revises each month twice, in the next two reports, and once a year it benchmarks the whole series to near-complete tax records.
Why do the two surveys disagree?
The payroll survey counts jobs (a person with two jobs counts twice) and leaves out the self-employed and farm workers. The household survey counts people, including the self-employed. Different samples and definitions mean they can move apart for months at a time.
Is the job market good right now?
It depends what you’re looking for. In September 2026 unemployment was 4.2%, low by historical standards, and 80.7% of people aged 25–54 had a job. But employers added only +45K jobs a month over the prior year, so people looking for work face a slower market.
Why is it so hard to find a job when unemployment is low?
Low unemployment says few people are looking, not that many are being hired. When job gains are small, openings get filled slowly and each one draws more applicants. JOLTS, the monthly report on openings and hires, shows that side directly.
Sources
- BLS, The Employment Situation, September 2026 (released Oct 2, 2026)
- BLS, Employment Situation Technical Note
- BLS Public Data API (the monthly series on this page)
✦ Editorial, written from BLS releases and methods documentation. Updated each month when the jobs report is released.
Cite this page
Truest. “The jobs report: how many jobs were added, and who’s out of work?” Updated Oct 2, 2026 with September 2026 data. https://www.truest.me/careers/data/jobs-report