But on average a few percent of people will be unemployed at the moment the survey week is conducted. And that is what the unemployment rate is measuring.
Let's look at this mathematically. If the average person changes jobs every four to five years, and spends two-three weeks unemployed doing so, as they already have a plan for their next job, you would expect a job-related unemployement rate of around 1% due to job changes.
In practice it makes up around 2%. The remainder in good times is people who were just fired, a few layoffs (since even in booms businesses can fail) and a few people entering the workforce.
