On this page
The US unemployment rate can fall even when the number of employed people stays the same. If some unemployed people stop meeting the job-search criteria, they can move out of the labour force. That changes both the number counted as unemployed and the group used to calculate the rate.
This guide uses the US Bureau of Labor Statistics’ Current Population Survey (CPS). In that survey, the unemployment rate compares unemployed people with the labour force, while the labour force participation rate compares the labour force with a broader population. Looking at both, alongside the employment–population ratio, helps you see what a falling rate does—and does not—show.
Start with three groups of people
The population used here is the US civilian noninstitutional population aged 16 and over. It excludes active-duty members of the US armed forces and people living in institutions such as prisons or residential care facilities. It is neither the whole population nor a group restricted to ages 16–64. These are US CPS definitions, not definitions for every country. BLS concepts and definitions
Within that population, people fall into three groups: employed, unemployed or not in the labour force. The labour force is the employed and unemployed groups combined. Someone outside the labour force has not become employed simply because they are no longer counted as unemployed.
For CPS purposes, being unemployed generally requires no employment during the reference week, availability for work, and at least one specific active job-search effort during the four weeks ending with that week. BLS allows an exception to active search for people on temporary layoff who expect recall. Availability also has a temporary-illness exception.
Active search can include submitting an application or contacting an employer. Simply browsing job advertisements without further action is a passive method under the BLS definition. These details matter: “has no job” and “is unemployed in the survey” are different descriptions.
Unemployment and participation use different denominators
The unemployment rate asks what share of the labour force is unemployed. The participation rate asks what share of the specified population is in the labour force at all. A third measure, the employment–population ratio, asks what share is employed. BLS definitions of the three measures
| Measure | Calculation | Question it answers |
|---|---|---|
| Unemployment rate | Unemployed ÷ labour force × 100 | Of the people in the labour force, what percentage are unemployed? |
| Labour force participation rate | Labour force ÷ specified population × 100 | What percentage of the population are employed or unemployed under the survey’s definitions? |
| Employment–population ratio | Employed ÷ specified population × 100 | What percentage of the population are employed? |
A person leaving unemployment for employment stays in the labour force. A person leaving unemployment for a status outside the labour force does not. Both changes can reduce the unemployment rate, but they have different effects on participation and employment.
A fictional population: leaving search versus finding work
Imagine a fixed population of 1,000 people who all meet the population definition above. At the starting point, 600 are employed, 40 are unemployed and 360 are outside the labour force. The labour force is 600 + 40 = 640.
Now compare two alternative outcomes with that same starting point. In Alternative A, 20 unemployed people have made no active job-search effort during the relevant four weeks, remain without employment, and do not qualify for the temporary-layoff exception. They are now outside the labour force. In Alternative B, those 20 people become employed instead.
| People or rate | Starting point | Alternative A: 20 leave the labour force | Alternative B: 20 find work |
|---|---|---|---|
| Specified population | 1,000 | 1,000 | 1,000 |
| Employed | 600 | 600 | 620 |
| Unemployed | 40 | 20 | 20 |
| Not in the labour force | 360 | 380 | 360 |
| Labour force | 640 | 620 | 640 |
| Unemployment rate | 6.25% | About 3.23% | 3.125% |
| Participation rate | 64% | 62% | 64% |
| Employment–population ratio | 60% | 60% | 62% |
At the starting point, unemployment is 40 ÷ 640 × 100 = 6.25%. In Alternative A, it becomes 20 ÷ 620 × 100 ≈ 3.23%. The rate falls, yet employment stays at 600. Participation falls from 64% to 62%, and the employment–population ratio stays at 60%.
In Alternative B, unemployment becomes 20 ÷ 640 × 100 = 3.125%. This time employment rises to 620. Participation stays at 64%, while the employment–population ratio rises to 62%.
The two alternatives show why the unemployment rate alone cannot tell you whether more people found work. They do not reconstruct the movements behind any published month. Real comparisons also need to account for population changes, survey estimation and the adjustment basis.
Outside the labour force does not automatically mean discouraged
“Discouraged worker” has a specific BLS meaning. Discouraged workers are a subset of people marginally attached to the labour force: they want a job, meet the relevant availability and earlier-search conditions, but have not actively searched during the most recent four weeks and give a job-market-related reason for not searching. BLS definitions of marginal attachment and discouragement
Not everyone outside the labour force meets those conditions. Even among people marginally attached, reasons for no recent search can include family responsibilities, education or health. A lower participation rate does not by itself tell you why people’s status changed, or whether all of them want a job.
We have not called the 20 people in Alternative A discouraged workers. The example supplies enough information to place them outside the labour force, but does not specify all the conditions for that narrower label.
Read the three measures together
For a release comparison, match the country, population, age group, reference period, seasonal adjustment and publication version. A rate for one age group cannot explain a rate for another without checking their composition. The fixed population in the example is a teaching assumption, not a feature you can assume in real data.
Keep people and payroll jobs separate too. CPS employment counts people; nonfarm payroll statistics come from a different survey and count payroll jobs. The nonfarm payrolls guide explains that distinction and the treatment of revisions.
The three rates help describe different aspects of labour-market conditions. They do not form an automatic score for economic health, explain every person’s circumstances or determine a policy decision. Nor does one unemployment reading date a recession; the US recession-indicators guide distinguishes observations from dating and model estimates. If you are comparing another type of headline, the US CPI guide shows why series and period labels matter there too.
Common questions
Is everyone without a job counted as unemployed?
No. Under US CPS definitions, people must meet the unemployment criteria, including availability and the active-search requirement or temporary-layoff exception. People outside the labour force are a separate group. Receiving unemployment benefits is not what determines the classification.
Does stopping job search for one day remove someone from the labour force?
No. The active-search test concerns the four weeks ending with the survey reference week, alongside the other criteria. Someone who applied for work within that window may still qualify as unemployed. The example assumes no active search throughout the relevant window and no temporary-layoff exception.
Are all people outside the labour force discouraged workers?
No. Discouraged workers must meet specific conditions and give a job-market-related reason for not searching. They are a subset of the marginally attached, who are themselves a subset of people outside the labour force. A participation-rate decline alone does not establish discouragement.
Can unemployment fall while employment is unchanged?
Yes. In Alternative A, employment remains at 600 while 20 people move from unemployment to outside the labour force. Unemployment falls from 6.25% to about 3.23%, participation falls, and the employment–population ratio stays at 60%. That is one fictional mechanism, not a claim about a real release.
Does every country use these population and search definitions?
Do not assume so. This guide uses US CPS definitions, including the civilian noninstitutional population aged 16 and over. To interpret another country’s statistics, check its statistical agency’s population, age limits, reference period and status rules. An English-language explanation does not make the US definitions the rules for India or any other country.
Explore HypaForecast markets
Browse the market list or read the linked BTC market questions and rules. These general navigation links do not establish that an unemployment contract is offered or available to you.
Sources and scope
The BLS CPS concepts page supports the US population, status and rate definitions used here. Its relevant text was consulted on 20 October 2026; the check date is not the article’s publication date. Both figures and all population counts are original teaching examples. This guide provides no current unemployment or participation reading and does not infer actual flows, personal circumstances or future market movements from the fictional calculations.
HypaForecast publishes educational guides and operates a prediction market. This guide explains US labour-market measures; it does not make a current economic assessment or recommend a transaction.
Share this guide