Is the Great Resignation Overblown?

We’ve been bombarded by headlines about the Great Resignation. “Workers Are Quitting Their Jobs in Record Numbers.” “We’ve Become a Nation of Quitters.” “The Great Resignation Is Accelerating.”

The Bureau of Labor Statistics, which isn’t given to exaggeration, has fed the coverage with headlines like this one, which it published on Jan. 6: “Number of Quits at All-Time High in November 2021.”

Well, guess what? The record numbers that everyone’s talking about cover only the period since December 2000. That’s the earliest month included in the Bureau of Labor Statistics’ Job Openings and Labor Turnover Survey. Waves of resignations may have been considerably higher in previous decades. In fact, as shown in the chart below, which compares the current survey with a similar Bureau of Labor Statistics survey from an earlier era, they probably were.

I’ll get to the caveats in a minute. But for now, take a look at the huge difference between then and now in the “quits rate.” The quits rate is defined today as the number of people who quit during the month divided by the number of employees who worked or were paid during a certain specified pay period that month.

The spike in quitting during World War II appears to dwarf the current spike. That makes sense. Back then, many factory workers, especially men, were enlisted in the armed services. Factory owners couldn’t raise pay to recruit replacements because there was a freeze on wages. Defense industries tried to recruit women with the famous Rosie the Riveter campaign. But the upshot was that workers could quit a job with complete certainty that another would be available — kind of like now, times 10.

A high quits rate may look like a victory for the working woman and man, but not necessarily, says Jay Zagorsky, an economist at the Questrom School of Business at Boston University. In an email exchange, he wrote, “The problem with very high quits rates is that businesses have to ensure jobs are very easy to learn and take no firm-specific skills.” He added, “When quits rates soar, businesses de-skill jobs and the end result is the opposite of what workers want.”

Now the caveats. The earlier Bureau of Labor Statistics survey, which was called the Labor Turnover Survey, was started in 1930 but was stopped in 1981 because of budget cuts. The agency brought back the survey in 2000 and expanded it to include job openings. The agency’s 2002 description of the new survey focuses on the new job openings calculation rather than the labor turnover calculation. It doesn’t carry the historical quits data on its website.

The definitions of quitting in the two surveys are different, but minimally, according to a document supplied to me on Wednesday by staff members of the Job Openings and Labor Turnover Survey (the document is not on the survey’s website). In the new survey, the document says, quitting is defined as leaving a job voluntarily. In the older Labor Turnover Survey it was defined as “terminations of employment initiated by the employee; failure to report after being hired (if previously counted as new hires); unauthorized absences if, on the last business day of the month, the person has been absent more than seven consecutive calendar days.” That sounds to me like a distinction without a difference.

I got the historical numbers from Historical Statistics of the United States, Millennial Edition Online, which is hosted by Cambridge University Press. The data from 1930 to 1981 is for the manufacturing sector only, so to make the comparison with the period since 2000 as close as possible I used the manufacturing quits rate, not the overall quits rate, in the more recent period. (There were some changes in methodology during the period from 1930 to 1981. For instance, beginning in 1943, quits rates referred to all employees; before then, they referred to production workers only.)

Laura J. Owen, an economist at DePaul University who has worked with the historical data, told me she would be cautious about comparing the data with the Job Openings and Labor Turnover Survey, saying, “I’m not sure I could make a definitive statement.”

Zagorsky was more comfortable making the comparison, saying: “The questions are pretty straightforward. You’re asking employers, how many people quit? Especially when you have administrative records.”

Even if the numbers aren’t precisely comparable, it’s hard for me to imagine that the quits rate now is higher than the rate during World War II. It may not even be as high as it was in the 1970s. Great Resignation? Maybe not so much.

Elsewhere

In “The Rime of the Ancient Mariner,” a poem published in 1798, Samuel Taylor Coleridge described the plight of a sailor dying of thirst while becalmed in the Pacific Ocean: “Water, water, every where/ Nor any drop to drink.”

In periods of financial stress, troubled banks are like the ancient mariner, argues a new working paper titled “Liquidity, Liquidity Everywhere, Not a Drop to Use — Why Flooding Banks With Central Bank Reserves May Not Expand Liquidity.” It’s by Viral V. Acharya of New York University’s Stern School of Business and Raghuram Rajan of the University of Chicago’s Booth School of Business.

Liquidity in financial terms is the ability to sell something quickly and easily, and it’s as important to banks as water is to sailors. In hard times the Federal Reserve tries to make banks more liquid by supplying them with the most liquid of assets — reserves at the central bank.

But very little of the liquidity goes to the banks that need it most, the economists write. The “perhaps most novel” reason for that, they say, is that “healthy banks may hoard liquidity and maintain unimpeachable balance sheets, in order to be perceived as safe and attract more deposit flows, rather than lending it out to stressed banks.”

The authors say their paper explains bouts of turbulence in the money markets in 2019 and 2020. They conclude that “central bank balance sheet expansion need not eliminate episodes of stress; it may even exacerbate them. This may also attenuate any positive effects of central bank balance sheet expansion on economic activity.”

Quote of the day

“Iceland, according to regulations of 1413 and 1426, had a centuries-old market price list for commodities, payable in dried fish (1 fish for one horseshoe, 3 for a pair of women’s shoes, 100 for a barrel of wine, 120 for a cask of butter, etc.).”

— Fernand Braudel, “Civilization and Capitalism, 15th-18th Century, Volume 1: The Structures of Everyday Life” (1981)

Have feedback? Send a note to [email protected].