The unemployment rate is the percentage of people actively looking for work who cannot find a job, measured monthly by the U.S. Bureau of Labor Statistics.
The rate sounds straightforward—it's not. The unemployment rate only counts people who are actively searching for work right now. If you stopped looking last month, you're not in it. If you're working part-time but want full-time work, you're not counted as unemployed. The rate is built on a specific definition that leaves out millions of people without work, which is why economists often look at other numbers alongside it.
The Bureau of Labor Statistics publishes the rate on the first Friday of each month, based on a survey of about 60,000 households conducted the week before. That survey asks whether each person worked, looked for work, or did neither. The unemployment rate comes from dividing the number of people actively job-hunting by the total labor force—everyone either working or actively looking. A rate of 4% means 4 out of every 100 people in the labor force are searching for work without success.
Key Takeaways
- The unemployment rate only counts people actively searching for work in the past four weeks, not everyone without a job.
- The Bureau of Labor Statistics calculates it monthly from a household survey and publishes it the first Friday of each month.
- The rate excludes discouraged workers who stopped looking, part-time workers who want full-time jobs, and people not in the labor force.
- A lower unemployment rate does not mean everyone has found work—it can mean people left the labor force entirely.
- Unemployment rates vary significantly by state, industry, education level, and race, so the national number masks real differences in job markets.
Who counts as unemployed
To be counted as unemployed, you must meet three conditions: you have no job, you have looked for work in the past four weeks, and you are available to start work when ready. "Looking for work" means specific actions—submitting applications, contacting employers, attending interviews, or registering with a public or private employment agency. Passively hoping for a job or checking job boards without explore does not count.
This definition excludes several large groups. Discouraged workers—people who want work but stopped searching because they believe no jobs are available for them—are not counted as unemployed. Students not looking for work are not counted. People who left their job voluntarily and are not searching are not counted. Someone working one hour per week is counted as employed, not unemployed, even if they desperately want full-time work. These exclusions mean the unemployment rate understates the number of people without adequate work.
The labor force and why it matters
The unemployment rate is a ratio: unemployed people divided by the total labor force. The labor force includes everyone 16 and older who is working or actively looking for work. It does not include retirees, students, people with disabilities not seeking work, or anyone else outside the job market. When someone stops looking for work, they leave the labor force—and the unemployment rate can fall even though that person still has no job.
This creates a counterintuitive situation. If 100,000 people stop searching for work in a given month, the unemployment rate can drop even if no new jobs were created. The denominator shrinks, so the same number of unemployed people represents a smaller percentage. During recessions, this effect is real: some people give up and leave the labor force, which can make the unemployment rate look better than conditions actually are.
How the Bureau of Labor Statistics collects the data
Every month, the Census Bureau conducts the Current Population Survey on behalf of the Bureau of Labor Statistics. Surveyors contact about 60,000 households and ask detailed questions about employment status for everyone 16 and older in the household. The survey covers the week that includes the 12th of the month. Households remain in the survey for four consecutive months, then rotate out for eight months, then return for four more months. This rotation ensures continuity while bringing in new households regularly.
The survey asks whether each person worked for pay, looked for work, or did neither during the reference week. It captures hours worked, industry, occupation, and reason for unemployment (job loss, quit, new entrant, re-entrant). The Bureau then weights the responses to represent the entire U.S. population and publishes the results as the official unemployment rate. The margin of error is typically around 0.2 percentage points, meaning a reported rate of 4.0% could actually be between 3.8% and 4.2%.
Unemployment rates vary by state, industry, and demographic group
The national unemployment rate masks significant variation. In any given month, some states have rates well above the national average while others are below it. Nevada, for example, historically has higher unemployment than New Hampshire. Unemployment also varies sharply by industry: construction and hospitality typically have higher rates than professional services or government. Education level matters: people with a bachelor's degree usually face lower unemployment than those with a high school diploma.
Demographic differences are substantial and persistent. Black unemployment is typically 1.5 to 2 times higher than white unemployment. Hispanic unemployment usually falls between Black and white rates. Asian unemployment is often the lowest. These gaps reflect differences in access to networks, discrimination, and geographic concentration in industries with higher job loss. The Bureau of Labor Statistics publishes breakdowns by state, industry, age, race, ethnicity, and education level, all released on the same day as the national rate.
Other unemployment measures that tell a different story
The official unemployment rate is called U-3 in government statistics. The Bureau of Labor Statistics also publishes five other measures, U-1 through U-6, each capturing a different slice of joblessness. U-1 counts only people who lost a job or completed a temporary assignment. U-5 includes discouraged workers. U-6, called the "underemployment rate," includes part-time workers who want full-time work and all marginally attached workers—people who want work and have looked in the past year but not the past month.
U-6 is typically 2 to 3 percentage points higher than U-3. When the official unemployment rate is 4%, U-6 might be 7% or higher. During recessions, the gap widens. Many economists argue U-6 better reflects actual labor market distress, but U-3 remains the headline number that policymakers and media report. Understanding which measure is being discussed matters: a politician citing the unemployment rate and an economist citing underemployment may be describing very different labor market conditions.
What unemployment rate changes actually mean
When the unemployment rate falls, it can mean jobs were created and people found work. It can also mean people stopped looking and left the labor force. When the rate rises, it can mean jobs were lost. It can also mean discouraged workers re-entered the labor force to search again—which is actually a sign of improving confidence, even though the rate went up. The monthly change in the number of jobs (the employment level) and the change in labor force participation often tell a more complete story than the rate alone.
Economists typically look at three numbers together: the unemployment rate, the number of jobs added or lost, and the labor force participation rate. A falling unemployment rate paired with rising labor force participation suggests genuine improvement. A falling rate paired with falling participation suggests people gave up. A rising rate paired with rising participation suggests people are re-entering the job market. The unemployment rate is real and important, but it is one piece of a larger picture.
Frequently Asked Questions
Why doesn't the unemployment rate include everyone without a job?
The rate measures labor market tightness—how hard it is for employers to find workers and for job-seekers to find jobs. Someone not looking for work is not competing for jobs, so they don't affect that tightness. Including everyone without a job would measure poverty or underemployment, which are different questions. The trade-off is that the rate can look better than conditions feel, especially when people have given up searching.
Can the unemployment rate go down when jobs are being lost?
Yes. If more people stop looking for work than lose their jobs in a month, the labor force shrinks and the unemployment rate can fall. This happened in 2020 during the pandemic: millions left the labor force, and the unemployment rate fell even as employment remained below pre-pandemic levels. The employment level and labor force participation rate reveal what the unemployment rate alone does not.
How long does someone stay counted as unemployed?
As long as they are actively searching. There is no time limit on how long someone can be counted as unemployed. However, if someone stops searching for four weeks or longer, they are no longer counted as unemployed—they become "not in the labor force." The average duration of unemployment varies with economic conditions; during recessions it can exceed six months.
Is the unemployment rate the same everywhere in the country?
No. Every state and major metropolitan area has its own unemployment rate, published monthly by the Bureau of Labor Statistics. Rates vary based on local industry mix, economic conditions, and population demographics. Rural areas often have different rates than cities. The national rate is a weighted average of all these local rates.
What's the difference between unemployment rate and underemployment rate?
The unemployment rate (U-3) counts people with no job who are actively searching. The underemployment rate (U-6) includes unemployed people plus part-time workers who want full-time work plus marginally attached workers. U-6 is typically 2 to 3 percentage points higher and captures more of the actual labor market distress, but U-3 is the official headline number.