Explainer: Why Friday's U.S. jobless figures won't capture the true state of the coronavirus economyExplainer: Why Friday's U.S. jobless figures won't capture the true state of the coronavirus economy
A “Now Hiring” sign sits in the window of Insomnia Cookies in Cambridge

By Ann Saphir

SAN FRANCISCO (Reuters) – The U.S. economy is expected to have shed 22 million jobs in April, tripling the nationwide unemployment rate to 16%, when new government data is published Friday morning.

To put that into perspective, the U.S. economy has never lost more than 2 million jobs in a single month. And although the unemployment rate reached 25% in 1933, it got there much more slowly.

But even these grim estimates, from economists polled by Reuters in recent weeks, don’t capture the staggering impact of the coronavirus pandemic on the workforce in the world’s largest economy.

The unemployment rate is part of a monthly report from the federal government’s Labor Department, showing how many people don’t have jobs as a percentage of the overall American workforce. The “jobs report,” as the release is known, provides two important labor market yardsticks: that unemployment rate, generated by a survey of households, and nonfarm payrolls, from a survey of businesses.

The unemployment rate has long been an indicator of the health of the economy, shrinking when jobs are plentiful and rising when times get hard.

A 16% unemployment rate means that 16 out of every 100 people who want to be in the nonfarm workforce don’t have jobs. That is a lot more than the 4.5% rate in March.

But some economists believe the “true” unemployment rate for April could be double that – meaning that more than a third of Americans who want or need to work cannot do so.

Here is what the “jobless” report does and doesn’t show:


Many U.S. workers who lose their jobs are eligible for weekly payments to help tide them over until they find their next job.

These payments, which workers and employers fund, are known as unemployment insurance benefits. Some 33.5 million people have filed for them since mid March, when states first began imposing stay-at-home orders, weekly data published on Thursday shows.

That is about one out of every five workers in the United States.

So does that mean the U.S. economy now has 33.5 million jobs fewer than it did then? No, because even during a severe recession, some employers – think supermarkets or delivery truck drivers right now – are still hiring. There have been three jobs added for every 10 layoffs, a recent Atlanta Fed survey found.

The unemployment claims only count the layoffs, but Friday’s report will show the net change in jobs.

The government actually publishes two numbers for jobless claims: a raw number, and an adjusted number that smoothes out swings based on predictable, seasonal factors, and makes it easier to compare one month to the next.

Because the record number of claims this time had nothing to do with seasonal factors, researchers at the New York Fed suggested Thursday a starting point for estimating what the Friday report might show for April’s job losses is the raw number of unemployment claims, which for the five weeks covered by the April report was 24 million.

However, those claims do not reflect every job lost in the United States, because not everyone who loses their job is eligible for unemployment insurance. There have been long delays in some states for processing claims, and the claims do not count until they are processed.

Confusing matters further, businesses that got loans under the government’s $660 billion Paycheck Protection Program may have hired back some people who had previously filed for unemployment benefits. And some states such as California, the country’s largest by population and GDP, let people file an unemployment claim if their hours were cut, even if they did not “lose” their job.


The Labor Department publishes six measures of unemployment, based on its survey of households. The main one is technically called U-3, and to qualify you need to be both out of work and looking for a job to be counted. That is the one that economists estimate will rise to 16% on Friday.

A broader measure of unemployment, called U-6, captures people who are not counted in U-3, like those working fewer hours than they would like, or who looked for work in the past but just not recently.

Reuters does not poll economists for their expectations for the U-6 rate, but last month it registered 8.7%, almost twice the official rate of 4.5%.

Even that broader measure will not tell the whole story of who is not working right now because it does not include two large groups: those on unpaid leave, and those who have lost their jobs and cannot look for or do work because of stay-at-home orders to slow the spread of COVID-19.

A pair of economists at the Chicago Fed recently crunched the numbers and found that accounting for such people would lift the real unemployment rate in the United States to between 25.1% and 34.6% – worse than the Great Depression-era numbers.

(Reporting by Ann Saphir; Editing by Heather Timmons and Daniel Wallis)