Whenever there’s an economic downturn, there’s a debate about the limits of the much-watched unemployment rate: People who give up looking for work are considered out of the labor force rather than unemployed; those who have to take part-time work when they’d rather be full-time are still considered employed; etc. Such points are made especially fervently by the party out of power.

But this is 2020, so the debate has to be crazier and weirder this time around. Yesterday, upon hearing the news that the unemployment rate actually declined in May, Paul Krugman half-seriously floated the theory that the administration was manipulating the statistics and later apologized. Now the Washington Post is out with a story titled “A ‘misclassification error’ made the May unemployment rate look better than it is.  Here’s what happened.”

The story itself takes the shock out of the headline. Nothing went wrong with the May unemployment rate in particular. There’s just an ongoing, and incredibly boring, technical difficulty stemming from the pandemic.

You Might Like

Basically, the unemployment rate is calculated from surveys the government conducts; people are classified as employed, unemployed, or out of the labor force depending on the answers they give to a slew of questions. But these questions were not written with a pandemic in mind. For instance, if someone is not working because their business was idled, are they considered temporarily laid off (and thus unemployed), or are they employed but not working, like someone who’s on leave?

These workers are supposed to be counted as unemployed, but they’re not always getting entered into the system that way:

The BLS instructed surveyors to try to figure out if someone was absent because of the pandemic and, if so, to classify them as on “temporary layoff,” meaning they would count in the unemployment data. But some people continued to insist they were just “absent” from work during the pandemic, and the BLS has a policy of not changing people’s answers once they are recorded. It’s how the BLS protects again[st] bias or data manipulation.

How does this affect the numbers? Basically, it just makes the official unemployment rate a little lower during this period. After adjusting for the sudden rise of “absent” workers, the March, April, and May unemployment rates are about 5.4 percent, 19.7 percent, and 16.3 percent, respectively, instead of the official figures of 4.4 percent, 14.7 percent, and 13.3 percent. The unemployment rate still fell a bit in May when everyone expected it to rise, and it’s still good news.

Hey, I warned you it was boring.