On October 2, 2026, the Bureau of Labor Statistics reported that the unemployment rate for September was 4.2 percent. Within minutes, that single figure was on every news site, and most readers took it to mean “4.2 percent of Americans can’t find work.”
That is close, but not quite right. The rate answers a narrower question than people assume, and the gap between the question it answers and the one you care about matters when you are job hunting. This guide walks through how the number is built, who falls outside it, how much weight a single month deserves, and how to use it without letting it mislead you.
Key takeaways
- The headline rate is the share of the labor force that is jobless, available and actively searching. It is not the share of all adults without a job.
- People who want work but have not searched recently, and part-timers who want full-time hours, are not in it. The BLS publishes broader measures for them.
- It is a survey estimate. Small monthly changes can be noise, and survey response has been falling.
- For your own search, local and occupation-level data matter more than the national headline.
Where the number comes from
The unemployment rate does not come from unemployment insurance claims or from a count of people who sign up for benefits. It comes from the Current Population Survey (CPS), a monthly survey of households. The Census Bureau’s interviewers do the fieldwork for the BLS, either in person or by phone.
According to the BLS, the sample includes about 60,000 eligible households, or roughly 110,000 people, spread over about 800 areas chosen to represent every state and the District of Columbia. About a quarter of the households rotate out each month, so roughly three-quarters of the sample carries over from one month to the next.
Interviewers ask about one specific week, called the reference week, which is usually the week that includes the 12th of the month. That is why a layoff on the 20th often does not show up until the following month’s numbers.
The survey covers civilians age 16 and older. People living in institutions, such as prisons and nursing homes, and active-duty military members are left out.
The formula, step by step
Two definitions do most of the work.
The labor force is everyone age 16 and older who is either employed or unemployed. The unemployment rate is the number of unemployed people divided by the labor force.
Employed means doing any paid work at all in the reference week, even a few hours, plus being temporarily away from a job because of illness or vacation. Unemployed has three tests, and a person has to pass the first two and one part of the third.
- No job during the reference week.
- Available to start work, apart from temporary illness.
- Actively searched at least once in the prior four weeks, or is on temporary layoff and expects to be recalled.
Everyone else is “not in the labor force.” That group includes students, retirees, full-time caregivers and anyone who wants a job but has not searched recently.
What counts as an active search
The BLS handbook says an active search is something that could lead to an offer without further action by the job seeker. Contacting an employer, sending an application, or interviewing counts. Reading job postings or taking a training class does not.
That has a practical consequence. Someone who spends a month quietly reading listings and rewriting a resume, but sends nothing, can be counted as outside the labor force that month.
A worked example
Say a small town has 1,000 adults. Of them, 600 have jobs, 40 are available and searching, and 360 are not working and not looking. The labor force is 640. Divide 40 by 640 and the rate is about 6.3 percent.
Now suppose 15 of the 40 searchers get discouraged and stop applying. Nobody found a job, but the labor force shrinks to 625, the unemployed count falls to 25, and the rate drops to 4 percent. That is an illustration, not real data, but it shows why a falling rate does not automatically mean more hiring.
Who the headline rate leaves out
Three groups are easy to miss.
Discouraged workers. These are people who want a job, are available, and searched in the past year, but are not searching now because they believe nothing is available to them. The BLS definitions page lists reasons such as past failures to find work, lacking the right skills or education, or facing age or other discrimination. They are not counted as unemployed.
Other marginally attached people. This is a wider group that wants work and searched at some point in the last 12 months, but not in the last four weeks. Discouraged workers are a subset. Someone who stopped looking because of a child care gap or a health problem would be here.
Part-timers who want more hours. They have a job, so they count as employed. The BLS calls them “part time for economic reasons,” meaning slack work or an inability to find full-time hours.
None of this is a flaw in the strict sense. The headline rate is built to measure active job search, consistently, over decades. It just was not built to measure everything about how comfortable the labor market feels.
The six measures, U-1 through U-6
The BLS publishes a family of six “labor underutilization” measures every month. The official rate is U-3. The others slide the definition narrower or wider.
| Measure | What it counts | September 2026 (seasonally adjusted) |
|---|---|---|
| U-1 | People unemployed 15 weeks or longer, as a share of the labor force | 1.8% |
| U-2 | Job losers and people who finished temporary jobs | 1.9% |
| U-3 | The official unemployment rate | 4.2% |
| U-4 | U-3 plus discouraged workers | 4.4% |
| U-5 | U-4 plus other marginally attached workers | 5.0% |
| U-6 | U-5 plus people working part time for economic reasons | 7.6% |
The figures come from the BLS Table A-15 for September 2026.
Read it left to right and the story is clear. Only a small share of the labor force has been jobless for more than 14 weeks. The official rate sits in the middle. Add discouraged and marginally attached workers and it rises to 5.0 percent. Add involuntary part-timers and it reaches 7.6 percent.
The U-6 number is often called “the real unemployment rate” online. That label is not accurate. U-6 measures something different, which is labor underutilization, and it mixes people with no job at all with people who have one but want more hours. Neither number is wrong.
Compare across time, not across measures
A U-6 of 7.6 percent next to a U-3 of 4.2 percent does not mean the government is hiding anything. The gap between the two is useful when it changes. If U-3 holds steady while U-6 climbs, hours are being cut or people are drifting out of the search, even though the headline looks calm.
Other numbers that belong beside it
The monthly Employment Situation report carries several companion figures. For September 2026 the BLS reported a labor force participation rate of 61.8 percent and an employment-population ratio of 59.2 percent.
| Indicator | What it tells you | Why it helps |
|---|---|---|
| Labor force participation rate | Share of the population 16 and older that is working or looking | A drop can pull the unemployment rate down without any new jobs |
| Employment-population ratio | Share of the population 16 and older that has a job | Not affected by who is or isn’t searching |
| Duration of unemployment | How long current jobless people have been looking | Shows whether spells are lengthening |
| Reasons for unemployment | Job losers, job leavers, reentrants, new entrants | Separates layoffs from people entering or returning |
| Part time for economic reasons | Involuntary part-timers | The extra piece inside U-6 |
The participation rate counts retirees on the outside, so it reflects the age of the population as well as the economy. When it moves, check how the BLS describes the change before drawing a conclusion.
If you want a refresher on where these figures sit in the monthly release, see our walkthrough of the monthly U.S. jobs report.
How precise is one month’s reading?
Here the story has changed in the past few years, and it is worth knowing.
The CPS is a sample, not a head count. The BLS states that its monthly estimate of the number of unemployed people has a 90 percent chance of falling within about 300,000 of what a full census would show. That is a small share of the labor force, but it is enough to make a one- or two-tenths move in the rate hard to read.
Falling response rates
In an April 2026 report to Congress on modernizing the survey, the BLS said the CPS sample is still 60,000 households, the same number as in 1981, even though the population it represents has grown 61 percent. It also said the survey’s response rate has declined from the low 90 percent range to the upper 60 percent range over 15 years.
The Atlanta Fed reached a similar picture in an April 2026 analysis. It put the response rate at 90 percent in 2013 and below 70 percent in 2024, and found it dropped nearly 5 percentage points between the September and November 2025 surveys, during the federal government shutdown.
The BLS spelled out what that means for the headline. In its report, it said a net change of fewer than 50 survey responses could move the headline unemployment rate by 0.1 percentage point. It also said a month-to-month change as small as 0.18 percentage point was statistically significant 20 years ago, while today it takes two months of data to reach the same confidence.
Those numbers are not a reason to dismiss the rate. They are a reason to hold a single reading loosely.
Don’t build a story on 0.1 point
A headline that says “unemployment ticks up to 4.2 percent” may describe a change inside the survey’s margin of error. Check whether the BLS calls the change significant, and look at three months of data before deciding the trend has turned.
Why the rate can fall for the wrong reasons
Go back to the town example. A falling rate has three common causes, and only one is good news.
- More people found jobs. Employment rises and the labor force holds steady or grows. This is the good version.
- People gave up searching. Discouraged workers leave the labor force. The rate falls and nobody is better off.
- People left for other reasons. Retirements, caregiving or going back to school shift people out of the labor force.
The quickest way to tell these apart is to check the employment-population ratio and the participation rate. If the rate fell while both of those also fell, cause two or three is probably in play.
The reverse happens too. When a strong labor market pulls people who had been on the sidelines back into searching, the unemployment rate can rise at first, because those newcomers are counted as unemployed before they land a job. A higher rate is not always bad news.
The rate is an average, and you are not
Even a perfectly measured national rate is a blend. It averages across states, occupations, ages and education levels, and your situation sits somewhere inside that blend.
The BLS Local Area Unemployment Statistics program publishes monthly employment, unemployment and labor force data for states, counties, metro areas and many cities. The numbers for smaller areas lean more on modeling, so expect more uncertainty than in the national figure.
Here is a short list of ways the national number can mislead a job seeker.
- Geography. A metro area can run well above or below the national figure.
- Field. A low overall rate can sit next to weak hiring in your industry, especially for entry-level roles. Our entry-level job search guide covers what to do when openings are thin.
- Job quality. The rate has no information about pay, benefits, schedules or stability.
- Fit. Employers can have openings that do not match the skills of the people searching, and the rate cannot show that mismatch.
- Posting quality. Not every advertised opening is real, a problem we cover in what ghost jobs are and how to spot them.
A five-minute routine for your own search
You do not need an economics degree. Once a month, after the BLS releases its report, do this.
- Write down the headline rate, the labor force participation rate and the U-6 figure from the release summary and Table A-15.
- Note whether the BLS describes the change from last month as little changed. If so, treat it as flat.
- Look up your state or metro area on the BLS local data page and compare it to the national figure.
- Check the JOLTS openings report for openings and hires in your industry, since openings tell you more about your odds than the rate does.
- Adjust your plan. Slow hiring suggests spending more time on networking and referrals. Strong hiring suggests applying widely and asking for what you are worth.
On the last point, the labor market backdrop affects your leverage. When employers compete for workers you have more room to ask, and our guide to salary negotiation shows how to do it without overreaching.
A sample way to describe the market to yourself
A one-paragraph market note
“As of the September report, the national rate is 4.2 percent and U-6 is 7.6 percent. My state is a little above the national rate. Postings in my field are steady but entry-level openings are thin, so I will send fewer, more targeted applications and ask three people for referrals this month.”
The point is not the exact words. It is that a few numbers become one decision.
Common mistakes when reading the rate
- Treating it as a share of all adults. The denominator is the labor force, not the population.
- Calling U-6 the “real” rate. It is a different measure with a different purpose.
- Reacting to one month. A move of a tenth of a point is often within sampling noise.
- Mixing seasonally adjusted and unadjusted numbers. Seasonally adjusted and unadjusted figures differ, as Table A-15 shows: in September 2026 U-3 was 4.2 percent adjusted and 4.0 percent unadjusted. Compare like with like.
- Assuming a low rate means a good market for you. Local conditions, your field and your experience level can tell a different story.
- Assuming a high rate means no one is hiring. Employers hire in every economy. They just have more candidates to choose from.
Looking ahead
The BLS has proposed adding an internet response option and running a larger parallel survey to modernize the CPS, while warning that a change in how people answer could affect results. That means the survey you read about next year may be built a little differently from the one described here. The definitions of employed and unemployed are the part to keep in your head, because they are the part that changes least.
This article is general information about public statistics, not financial or career advice.
The bottom line
The headline unemployment rate is a careful, consistent measure of active job search. It is not a measure of how hard it is to find a good job, and it is a survey estimate with real uncertainty.
Next steps:
- Look up the latest BLS Employment Situation release and read the first page, not only the headline.
- Compare U-3 and U-6 in Table A-15, and watch the gap over several months.
- Check your state or metro rate through the BLS local data tools.
- Pair the national picture with openings data and postings in your own field.
- Set your search plan by what you see locally, then revisit it next month.
