What Is the Unemployment Rate? How Economists Measure Joblessness

Economists usually divide working-age people into broad labour-market groups. Employed people have work under the survey definition. Unemployed people do not have work, are available to work, and have taken recognised steps to seek a job within a recent period. Together, employed and unemployed people make up the labour force.

The unemployment rate is calculated by dividing the number of unemployed people by the labour force and multiplying by 100. Someone who is retired, studying without seeking work, caring for family without looking for a job, or otherwise outside the labour force is not counted as unemployed in this measure.

This distinction is the key to the unemployment rate. Imagine 1,000 working-age people, with 600 employed, 30 unemployed and looking, and 370 outside the labour force. The labour force is 630, so the unemployment rate is 30 divided by 630, not 30 divided by 1,000.

Unemployment can rise during an economic downturn when businesses sell less, reduce production, and need fewer workers. It can fall when demand strengthens and firms hire. Economists often call unemployment connected with the business cycle cyclical unemployment.

Other unemployment exists even in a healthy economy. Frictional unemployment occurs while people move between jobs or enter the workforce. Structural unemployment can arise when workers' skills, industries, or locations do not match available jobs. Seasonal patterns can also affect industries such as tourism, agriculture, and construction.

These categories make the unemployment rate more informative. The same headline rate can come from different underlying situations. A temporary wave of job switching is not the same problem as a long-term mismatch between workers and available jobs.

The headline figure is useful, but it does not describe every form of labour-market difficulty:

• •People who want a job but have stopped searching may be outside the official unemployed group.
• •A part-time worker who wants more hours is still usually counted as employed.
• •The rate does not directly show job quality, pay, security, or working conditions.
• •Two places with the same unemployment rate can have different participation rates.
• •Short-term and long-term unemployment can have very different effects on workers.

Labour force participation gives another piece of the picture by showing what share of the relevant population is working or actively looking for work. Employment rates, hours worked, wage growth, job vacancies, and long-term unemployment can also reveal changes that one percentage misses.

When you read the unemployment rate in a chart or news report, check the definition, country, age range, and time period. Statistical agencies may use slightly different survey rules. The number is best treated as one indicator in a larger labour-market dashboard rather than a complete judgement about how easy it is to find good work.

Most countries do not count unemployed people one by one. Instead, statistical agencies run large household surveys, asking a sample of people about their work and job searching during a recent reference period. Many follow definitions set out by the International Labour Organization, which makes comparisons between countries more meaningful. Because the figure comes from a sample, it has a margin of error. A small change from one month to the next may be statistical noise rather than a real shift, so analysts usually look at trends over several releases.

Some countries also publish a separate measure based on the number of people claiming unemployment-related benefits. This claimant count can move differently from the survey figure, because not everyone who is unemployed is eligible to claim, and changes to benefit rules can alter the count without any change in the job market. A simple worked example shows why definitions matter. If all 10 unemployed people in a labour force of 100 give up searching, unemployment falls to zero and the labour force shrinks to 90, yet nobody has found a job.

The unemployment rate sometimes moves in a direction that seems backwards. Imagine a labour force of 100 people with 90 employed and 10 unemployed, a rate of 10 percent. The economy improves and firms create 5 new jobs. Encouraged by the better news, 8 people who had given up searching start looking again and rejoin the labour force. Now 95 people are employed and 13 are unemployed in a labour force of 108, so the rate has risen to about 12 percent, even though employment grew.

Unemployment also tends to lag behind the wider economy, because firms often wait to be sure a recovery is lasting before hiring. Historical figures need care too. The often quoted estimate that around a quarter of the American labour force was unemployed in 1933, during the Great Depression, was largely reconstructed afterwards from scattered records, since the United States did not run a regular monthly household survey of employment until around 1940. Comparing today's rates with those of the distant past therefore means comparing a modern survey with a careful historical estimate.

In 1958 the economist A. W. Phillips, a New Zealander working at the London School of Economics, published a study of British data stretching back almost a century. He found that when unemployment was low, wages tended to rise quickly, and when unemployment was high, wage growth slowed. Other economists soon generalised this into the idea that governments faced a menu of choices: accept a little more inflation in exchange for lower unemployment, or the reverse. For a time the relationship seemed to hold, and it shaped how policymakers in several countries thought about managing the economy.

The 1970s broke the simple version. Many countries suffered high inflation and high unemployment at the same time, a combination nicknamed stagflation. Milton Friedman and Edmund Phelps had already argued in the late 1960s that once people expect inflation, they build it into wage demands, so pushing unemployment below a certain level produces ever-rising inflation rather than a lasting gain. That idea survives in the concept of a natural rate of unemployment, closely related to a measure called NAIRU. Central banks still try to estimate it, though it cannot be observed directly and estimates shift over time.

Headlines about youth unemployment are easy to misread. Suppose there are 100 people aged 16 to 24, of whom 50 are full-time students not looking for work, 40 have jobs and 10 are unemployed and searching. The labour force is 50, so the youth unemployment rate is 10 divided by 50, or 20 percent. That does not mean one young person in five is unemployed. Only 10 of the 100, or 10 percent, are. Some statistical agencies publish this second figure, the unemployment ratio, alongside the rate, because a large student population can make the rate look far more dramatic than the share of young people affected.

Economists also link unemployment to the size of the economy. In 1962 the American economist Arthur Okun described a pattern in United States data: each percentage point by which unemployment rose above its normal level went along with national output roughly 3 percent below what the economy could otherwise produce. Many later estimates put the figure closer to 2 percent. Known as Okun's law, it is really a rule of thumb rather than a law, and the ratio varies between countries and over time. It still gives a quick sense of the lost production that sits behind each rise in joblessness.

The unemployment rate is unemployed job seekers as a percentage of the labour force, not of the whole population. Learn the denominator first, then look at participation, job quality, and the reasons unemployment is changing. That small bit of context makes the headline number much easier to interpret.