Real news.A stronger Maryland.

MARYLAND HERALD


Maryland Edition

LATEST
Economy

How the Unemployment Rate Is Calculated, and Who Gets Counted

The unemployment rate is the number of unemployed people divided by the labor force, then multiplied by 100. Here is who counts as unemployed, how the survey works and why the headline number leaves some people out.

Historic storefronts line a charming small town street

The unemployment rate is the number of unemployed people divided by the total labor force, multiplied by 100. The labor force is everyone who is either employed or unemployed, so the rate is not a share of the whole population and not a count of everyone without a job.

The figure comes from a monthly government survey of households, not from unemployment insurance records. This guide walks through who is counted, how the arithmetic works, a worked example and what the headline number leaves out.

What is the formula for the unemployment rate?

According to the Bureau of Labor Statistics (BLS), the unemployment rate is the number of unemployed people as a percentage of the labor force. Written out:

Unemployment rate = unemployed ÷ (employed + unemployed) × 100

The denominator, the labor force, is the sum of the employed and the unemployed. People who are neither working nor looking for work are outside the labor force and do not appear in either part of the fraction.

Where do the numbers come from?

The data come from the Current Population Survey (CPS), which the Census Bureau conducts for the BLS every month. The BLS describes the sample as about 60,000 eligible households, covering roughly 110,000 people.

Several design features matter for how the numbers should be read:

  • Reference week. Interviewers ask about the week that includes the 12th of the month.
  • Rotation. Each household is interviewed for four consecutive months, leaves the sample for eight months, then returns for four more. One-fourth of households rotate out each month.
  • Weighting. Responses are adjusted to Census Bureau population estimates by age, sex, race, Hispanic ethnicity and state, so the sample stands in for the whole civilian population.
  • Population base. The survey covers the civilian noninstitutional population age 16 and older. People in institutions, such as prisons, and people on active-duty military service are excluded.

Because it is a sample, the result carries sampling error. The BLS says that, with 90 percent confidence, the monthly unemployment estimate is within about 300,000 of the figure a full count would produce. That is why small month-to-month moves are usually described as “little changed.”

Who counts as employed?

Under BLS definitions, people are employed if they did any work for pay or profit during the reference week, including part-time and temporary work. People who have a job but were temporarily absent, for reasons such as illness, vacation or a labor dispute, are also counted as employed. So are unpaid family members who worked 15 or more hours in a family business or farm.

Each person is counted once, even with more than one job. And work takes priority over job searching: someone who worked a few hours in the reference week and is also looking for a better job is counted as employed, not unemployed.

Who counts as unemployed?

The BLS says a person must meet all three of these conditions to be counted as unemployed:

  1. They did not work during the reference week.
  2. They were available for work, apart from temporary illness.
  3. They made a specific, active effort to find a job in the prior four weeks, or they were on temporary layoff and expecting to be recalled.

Active job search includes contacting employers or employment agencies, sending resumes, answering job ads, and contacting friends or relatives about openings. Passive steps, such as reading postings or attending a training course, do not count.

Eligibility for unemployment insurance is not part of the test. A person who never applied for benefits, or who has used up their benefits, is still counted as unemployed if they meet the three conditions.

How does a worked example look?

The numbers below are illustrative arithmetic, not real data. Suppose a country has a civilian noninstitutional population age 16 and older of 250 million, of whom 150 million are employed and 6 million are unemployed.

Measure Calculation Result
Labor force 150 million + 6 million 156 million
Unemployment rate 6 ÷ 156 × 100 about 3.8%
Labor force participation rate 156 ÷ 250 × 100 about 62.4%
Employment-population ratio 150 ÷ 250 × 100 60.0%
Not in the labor force 250 − 156 94 million

Now suppose 1 million of the 94 million people outside the labor force start looking for work and are available. The unemployed count rises to 7 million and the labor force to 157 million, so the rate is 7 ÷ 157 × 100, or about 4.5 percent. Nobody lost a job, yet the rate rose. This is why analysts read the unemployment rate alongside participation and the employment-population ratio, both of which the BLS also publishes.

What are the most common misconceptions?

“The rate counts everyone without a job.” It does not. Retirees, full-time students, people caring for family and anyone not searching are outside the labor force. The BLS notes that students who are not available for work, for example, are not counted as unemployed.

“The rate is based on people collecting benefits.” It is not. The BLS says unemployment insurance claims are not used to measure total unemployment because they exclude groups such as the self-employed, people who have exhausted benefits, new entrants to the labor force and eligible people who never file. The agency notes that only about one-third of the unemployed received regular benefits over the decade covered by its explanation.

“A falling rate always means more people are working.” Not necessarily. The rate can fall because people stop searching and leave the labor force. Maryland’s September report is a recent case of two measures moving differently, as Maryland’s 3.9 percent jobless rate alongside a smaller payroll count showed.

“The household survey and the payroll survey are the same.” They are different. The unemployment rate comes from the household survey. The monthly jobs-added figure comes from a separate survey of employers, which is discussed in the September jobs report coverage.

What does the headline rate leave out?

The official rate is one of several measures. The BLS publishes alternative measures of labor underutilization, labeled U-1 through U-6, and the official rate is U-3.

Measure What it adds or isolates
U-1 People unemployed 15 weeks or longer, as a share of the labor force
U-2 Job losers and people who finished temporary jobs, as a share of the labor force
U-3 The official rate: total unemployed as a share of the labor force
U-4 Adds discouraged workers to both the unemployed and the labor force
U-5 Adds all marginally attached people to both
U-6 Adds marginally attached people and those working part time for economic reasons

The BLS defines marginally attached workers as people who are not in the labor force, want a job, are available and searched for work within the past 12 months, but did not search in the last four weeks. Discouraged workers are a subset who did not search because they believe no suitable work exists, among other reasons.

Using the earlier illustration, suppose 1 million people are marginally attached and 2 million work part time because they cannot find full-time hours. U-6 would be (6 + 1 + 2) ÷ (156 + 1) × 100, which is 9 ÷ 157, or about 5.7 percent, compared with 3.8 percent for the official rate. The gap shows why a single number never tells the whole story.

Does the rate differ by place and by date?

Yes. National figures come straight from the CPS. State rates work differently. According to the BLS Local Area Unemployment Statistics program, state estimates are model-based, combining current and historical CPS data with employer-survey data and state unemployment insurance records, and they are controlled so that they sum to the national CPS totals. State workforce agencies prepare the estimates under BLS standards. County and city figures are built with other methods, so they should not be compared casually with the national headline.

Timing matters as well. The BLS reported that the national rate was 4.1 percent in October 2026, with 7.1 million people unemployed, in its release of November 6, 2026. Those figures are time-sensitive and will change with each monthly report. For related labor-market context, see how job openings compare with unemployment and how another monthly index is built in the explainer on measuring inflation.

What should readers remember when the next report arrives?

The headline rate answers a narrow question: of the people who are working or actively looking, what share are looking but not working? It is a reliable, consistently defined measure, but it carries sampling error and says nothing about people who have stopped searching, who work fewer hours than they want or who earn low pay.

When a new monthly report arrives, a sensible reading looks at the rate, the participation rate, the employment-population ratio and the broader U-6 measure together, and treats changes of a tenth of a percentage point as noise.

This article is for general information only and is not financial or employment advice.

Sources: U.S. Bureau of Labor Statistics; U.S. Census Bureau