How High Will the Unemployment Rate Rise in April?


Leisure and hospitality jobs once again provide an accurate recession signal.

Leisure and Hospitality Jobs vs Recessions

Leisure and Hospitality Jobs will be the hardest hit in this recession once again so let's start there.

Image placeholder title

Leisure and Hospitality Percent Change From Year Ago Detail

Image placeholder title

In February, there were 16.8 million Leisure and Hospitality (L&H) workers.

If only half of them lose their jobs (and that is very optimistic), that's 8.4 million layoffs just in this category.

The previous worst case was -3.4% in April of 2009.

Food Services and Drinking Places

Image placeholder title

In February there were 12.3 million employed in the Food Services and Drinking Places category.

Food Services is a subset of L&H.

Assuming 80% layoffs, there will be close to 10 million losses in this subcategory alone.

Crunching the Numbers

Let's grab some numbers from my review of the March jobs report Unemployment Rate Jumps to 4.4% But Worst is Yet to Come.

Image placeholder title

If we make the assumption that the economy will only shed 10 million workers in L&H and none anywhere else, we can calculate the April unemployment rate as follows.

(7.14 million existing + 10 million new) / 162.91 million labor force = 10.5%.

That is a silly assumption given all that it ignores, yet, there are sub 10% unemployment forecasts.

Hard-Hit Employment Categories

L&H will be the hardest hit category but it will not be the only hard-hit category.

What about all the people working in retail stores and temporary help?

Risk Groups

Image placeholder title

The above chart is from the St. Louis Fed Article COVID-19: Which Workers Face the Highest Unemployment Risk?

The author did not estimate job losses but he did provide a couple of IFs.

For the unemployment rate to reach 9%, an additional 9 million workers would have to become unemployed, the equivalent of 13% of workers in high-risk occupations.

Alternatively, if 50% of workers in high-risk occupations were to be laid off, the unemployment rate would reach 24%.

Nine percent seems wildly optimistic by my model.

Estimating Job Losses

There are 151 million All Employees.

Excluding my breakouts, there were another 151-16.4-15.6-2.9-21.6-4 = 94.5 million employed in March.

If 7.5% of them lose their job, that's another 7 million.

Total Job Losses in Millions = 9.5 +2.3 +3.9 +1 +3.2 + 1 + 7 = 27.9 million.

Let's crunch that.

Estimated Unemployment Rate

(7.14 million current + 27.9 million new) / 162.91 million labor force = 21.5%.

Another way to arrive at a similar number is to assume 11.8 million L&H workers and another X% from the rest of the 46% of high risk workers plus another Y% from the bigger safe worker pool.

Unemployment Rate Spreadsheet

Image placeholder title

Both approaches yield similar numbers.

There are many ways to top the 20% mark assuming my initial estimate of 11.9 million L&H jobs is in the ballpark.

I used Seasonally Adjusted Numbers.

You can try Unadjusted Numbers but you have to use the unadjusted Labor Force and make other assumptions. Then it would be necessary to seasonally adjust the final result because the reported headline unemployment number is seasonally-adjusted.

Looking for a Range?

Try 16% to 28%. Yes, that's wide. It could take two month to reach.

Feel free to plug in your own assumptions from my starting point.

By monitoring the next three weekly unemployment reports, we can hone in on that number.

Regardless, a new record high unemployment rate is baked in the cake and that does not count those forced to part-time work.

Reference Period

The reference period week for determining the unemployment rates is the week that that contains the 12th of the month. That makes the next reference period April 12-18.

If you work as little as one hour in the reference period week, you are considered employed.

The better the unemployment rate looks in the May report (for April), the more likely hours will be reduced across the board somewhere else.

The U-6 unemployment rate which counts people working part-time who want a full time job can easily top 30%.

Heck, a straight up 30% is not out of the question if you consider dentists, dental assistants, ambulatory services, airlines, boeing, GM, and car-related manufacturing, etc, and I am low on retail and the alleged "safe" group.

How Fast a Recovery?

Eventually, the coronavirus depression will subside. But what will America look like when it does?

The huge fear now is How Do I Pay the Bills?

This is the second crisis in 12 years. These scars will last.

Note that Vehicle Sales Plunge by 5.5 Million Units, Most Since 1987.

With the stock market down and consumer savings wiped out, will car sales quickly recover?

The same question applies to eating out and airline travel. Simply put, a quick return to business as usual is not in the cards.

For a 20-point discussion of what to expect, please see Nothing is Working Now: What's Next for America?

No V-Shaped Recovery

Add it all up and you should quickly arrive at the correct viewpoint: The Covid-19 Recession Will Be Deeper Than the Great Financial Crisis.

Mike "Mish" Shedlock

Comments (41)
No. 1-12

What is fair value of the S&P 500 at 20% unemployment? Does that include reduction in EPS? What would be your target PE ratio?


“To the moon, Alice, to the moon!”

Tony Bennett
Tony Bennett

Jamie Dimon expecting a bad recession.

I think that is counter (for now) to general wisdom ( of Wall Street and business) that once virus peaks it will lead to a surge of spending as locked in consumers emerge to spend their stimulus checks. If that thinking rules, then we'll have to wait longer for more job cuts.

The Cheese Cake Factory being played out all over ... it will take till next year before economy stabilize even if virus fades away sooner rather than later.

The economic bump (not as big as Wall Street expects) when virus fades will soon meet Reality of bankruptcy / delinquency / default.

The Everything Bubble bursting won't be pretty.

Tony Bennett
Tony Bennett

Not sure how much CARES Act will stem tide of layoffs for now.

Roll out a bit of a debacle, but businesses able to get low interest loan that becomes grant if conditions met ... foremost keeping staff on board.


If a man dies of Coronavirus but he isn't tested, did he really die of Coronavirus?


I think it is not correct to use seasonally adjusted numbers when calculating the unemployed. Last week there only 5,8 million applied for benefits, not 6,6 million. The same thing the week before, only 2,9 million applied. Furthermore about 0,2 million is normally applying weekly. So I would say that you should use 8,3 million instead of 10 million when calculating unemployment figures.


A lot of work at home occupations are dependent on work at work occupations. Look for 5 to 10 percent additional for every month the foolishness lasts.


Agree with Tony Bennett that we don’t yet know how the CARES Act and future bailouts/stimuli will affect employment. But however you look at it things will be ugly in Q2.

How long will it take to recover from this depression? A v-shaped recovery seems highly unlikely; Mnuchin makes it sound like we just turn off the lights and once we turn them back on everything goes back to normal. He can’t really believe what he is saying.


"How High Will the Unemployment Rate Rise in April?"

Maybe a lot higher than it should be. Denninger pointed out this morning that the R naught peak in New York occurred the day after Cuomo announced restrictions, thus Cuomo's restrictions did not cause R naught peak and subsequent decline over the next 5 days, it was already in the pipeline.

Denninger also had a link to an interesting Youtube video, Perspectives on the Pandemic. Part Two, an interview with Knut Wittkowski, PHD. He is one of those scientific types. He said the key is herd immunity, which isn't being attained by bending the curve and according to Wittkowsi, guarantees a second wave.

Get this- Wittkowski said that if allowed to run its course, attaining herd immunity, the process takes about a month.

Note that Denninger said that the NY R naught peak occurred the day after Cuomo announced restrictions. It was on March 4 that L.A. County declared an emergency. Denninger noted R0 peaks for NY, Florida and California have occurred. Today is April 6, just over a month after L.A. declared an emergency. It would seem it falls in line with what Wittkowski said.


According to statistics 60% of people can’t come up with $1000 for an emergency. I’d be willing to bet the majority of the people in the initial layoffs are in that category. There may be a safety net in the CARE legislation but it seems to be a net with very large holes that many will slip through. I’m fiscally conservative but we really need to help those struggling with rent and putting food on the table. Even if it means deficit spending.


20% is baked in. Anything above that is really bad icing on really bad tasting cake. U-6 will also be the more important number. Additionally, you will have people filing for unemployment even when they are gig workers. That will overstate the filing rate. Coronavirus and the way they define unemployment is going to really mess up the numbers (in both directions).


Some unemployment statistics are masked with "furloughs" Honeywell, for instance, has "furloughed" much of its workforce. Nobody in collecting unemployment because employee work for a week and then stay home for a week. This is in stark contrast to retail workers who stay at home for weeks on end and collect unemployment insurance.

Global Economics