Skip to main content

Independent information site. Not affiliated with or endorsed by any government agency. For informational purposes only.

HomeAll guides

How the Unemployment Rate Is Calculated and What It Actually Measures

What the unemployment rate counts and what it leaves out

The unemployment rate is a single number — usually between 3 and 10 percent — that represents the share of people actively looking for work who cannot find it. The Bureau of Labor Statistics (BLS), a division of the U.S. Department of Labor, calculates this number every month using data from two separate surveys. It does not count all jobless people. It counts only those who have looked for work in the past four weeks and are available to start a job when ready.

This distinction matters because millions of people without jobs are not counted in the unemployment rate. Someone who stopped looking for work six weeks ago, someone in school full-time, someone retired, or someone unable to work due to illness does not appear in the number. The rate also does not capture people working part-time who want full-time work, or people so discouraged they have stopped searching altogether. Understanding what the rate measures — and what it ignores — is the first step to reading unemployment news accurately.

Key Takeaways

  • The unemployment rate counts only people without a job who have actively searched for work in the past four weeks and are ready to start when ready.
  • The BLS collects unemployment data through two monthly surveys: the Current Population Survey (household data) and the Current Employment Statistics (employer payroll data).
  • A person must meet all four criteria — no job, active job search, availability, and recent effort — to be counted as unemployed; missing any one disqualifies them.
  • The official unemployment rate excludes discouraged workers, part-time workers wanting full-time jobs, and anyone who has not searched in the past month, making it lower than some alternative measures.
  • The BLS releases the monthly unemployment rate on the first Friday of each month, covering the previous month's data.

The four requirements to be counted as unemployed

To appear in the unemployment rate, a person must meet all four conditions at the same time. The first is having no job at all — not even part-time or self-employment work. The second is having actively looked for a job within the past four weeks. "Active" means specific actions: submitting applications, interviewing, contacting employers directly, or registering with a public employment agency. Passively hoping for a job or reading want ads does not count.

The third requirement is being available to start work when ready if offered a job. Someone caring for a sick family member with no backup childcare, or someone scheduled for surgery next month, would not meet this test. The fourth is being part of the labor force — meaning the person is either working or actively seeking work. Students not looking for jobs, retirees, and people with disabilities not seeking employment fall outside the labor force entirely and are not counted as unemployed.

All four must be true in the same week. If someone stopped job searching two months ago but is willing to work, they are not unemployed in the BLS definition — they are out of the labor force. If someone is actively searching but has a part-time job, they are employed, not unemployed, even if they want more hours.

How the BLS collects unemployment data each month

The BLS uses two separate surveys to build the monthly unemployment picture. The Current Population Survey (CPS) contacts about 60,000 households by phone and in person. Trained interviewers ask detailed questions about employment status, job search activities, and availability. This survey produces the official unemployment rate and also captures information about who is searching, how long they have been unemployed, and what kinds of jobs they are seeking.

The second survey, the Current Employment Statistics (CES), contacts about 140,000 business establishments — factories, offices, hospitals, stores — and asks how many people are on their payroll. This employer survey does not measure unemployment directly; instead, it counts jobs added and lost. The two surveys sometimes tell different stories: the CPS might show unemployment rising while the CES shows employers still hiring, or vice versa. Both are correct because they measure different things.

The BLS collects data during a specific week each month called the reference week. For the CPS, this is the week that includes the 12th of the month. Data collection continues for several weeks after the reference week ends, so the agency is still gathering information even as it begins analyzing the previous month's data. The official unemployment rate is released on the first Friday of the following month.

Why the official rate differs from alternative unemployment measures

The BLS actually publishes six different unemployment rates, labeled U-1 through U-6. The one reported in news headlines is U-3, the official unemployment rate. The others exist because reasonable people disagree about who should be counted as unemployed. U-1 counts only people unemployed for 15 weeks or longer — a narrower group. U-5 includes people who searched for work within the past year but not in the past month, capturing some discouraged workers. U-6, the broadest measure, includes part-time workers wanting full-time jobs and people who want to work but have not searched recently.

U-6 is typically two to three percentage points higher than U-3. If the official rate is 5 percent, U-6 might be 7 or 8 percent. Neither number is wrong; they answer different questions. U-3 tells you the share of people actively competing for jobs right now. U-6 tells you how many people are underemployed or have given up searching. During recessions, the gap between U-3 and U-6 widens because more people stop looking after months of rejection.

The BLS publishes all six rates monthly on its website and in the monthly employment report. News outlets focus on U-3 because it is the official measure and allows year-to-year comparison, but understanding that alternatives exist helps explain why unemployment can feel worse than the headline number suggests.

Seasonal adjustment and why January looks different from July

The raw unemployment data collected each month is "seasonally adjusted" before release. Retail hiring spikes in November and December, construction slows in winter, and agriculture peaks in summer. Without adjustment, the unemployment rate would swing wildly every month just because of these predictable seasonal patterns, making it hard to spot real economic changes. The BLS uses historical patterns to estimate what the unemployment rate would be if these seasonal swings did not exist.

The adjustment is not perfect. Unusually warm winters, early holiday shopping, or supply chain disruptions can throw off the historical pattern. The BLS also publishes the unadjusted rate alongside the adjusted one, so you can see both numbers. Most economic analysis uses the seasonally adjusted figure because it is easier to compare month to month. If you are reading unemployment news and wondering why January's number looks different from December's, seasonal adjustment is usually part of the explanation.

What happens when unemployment data is revised

The unemployment rate released on the first Friday of each month is preliminary. The BLS collects more responses to the Current Population Survey over the following weeks, and employers submit more detailed payroll data. The agency revises the previous month's unemployment rate when new information arrives, usually in the second month's report. These revisions are typically small — a tenth of a percent — but they can shift the narrative. A rate reported as 4.0 percent might be revised to 4.1 percent, changing whether the month showed improvement or decline.

The BLS also conducts a major annual revision each January, adjusting the previous 12 months of employment data based on more complete employer records. These annual revisions can be larger, sometimes shifting the unemployment rate by several tenths of a percent for months that seemed settled. If you are comparing unemployment rates across years, check whether you are using the originally reported numbers or the revised ones — they may differ.

How unemployment rates vary by state and demographic group

The national unemployment rate masks significant variation. Some states consistently have lower unemployment than the national average; others run higher. The BLS publishes state-level unemployment rates monthly, usually with a one-week lag behind the national release. These state rates use the same definition and methodology as the national rate, so they are directly comparable.

Unemployment also varies sharply by age, race, education level, and gender. The BLS breaks down the national rate by these groups in its monthly report. Typically, unemployment is lowest among college-educated workers and highest among teenagers and workers without a high school diploma. These differences persist across economic cycles. Understanding your own demographic group's unemployment rate — not just the national headline — gives a more accurate picture of your own job market.

Frequently Asked Questions

Why does the unemployment rate sometimes go down when jobs are lost?

The unemployment rate can fall if people stop looking for work faster than jobs disappear. If 100,000 people leave the labor force and only 50,000 jobs are lost, the unemployment rate falls because the denominator (people in the labor force) shrinks. This happened during the COVID-19 pandemic, when the unemployment rate fell even as total employment remained below pre-pandemic levels.

Is the unemployment rate the same as the job loss rate?

No. The job loss rate measures how many jobs disappear in a month. The unemployment rate measures the share of people actively seeking work who cannot find it. A month can have significant job losses but a falling unemployment rate if enough people leave the labor force, or rising unemployment with job growth if more people enter the labor force seeking work.

How long does someone stay in the unemployment statistics?

Someone remains counted as unemployed as long as they meet all four criteria: no job, active job search, availability, and being in the labor force. The moment they stop searching, they drop out of the unemployment rate and become classified as "not in the labor force." This can happen after weeks or months, depending on how long they continue looking.

Does the unemployment rate include self-employed people?

No. Self-employed people are counted as employed, even if their business is struggling or generates very little income. The BLS counts anyone with self-employment income, however small, as employed. Someone who lost a job and started a side gig for a few dollars a week would no longer appear in the unemployment rate.

Why is the unemployment rate released so late in the month?

The BLS waits until the first Friday of the following month to allow time for data collection and verification. The reference week ends around the 12th of the month, but responses continue arriving for weeks. Releasing the rate too early would mean missing responses and producing less accurate numbers. The one-week delay ensures the data is as complete as possible.

This guide is for general information only and is not medical, financial, legal, or other professional advice. For decisions specific to your situation, consult a qualified professional. See our Editorial Policy.