In 2021 there were many stories about workers quitting jobs. Pundits dubbed this phenomenon the Great Resignation.

Statistically, the best measure of quits in the United States comes from the government's Job Openings and Labor Market Survey, or JOLTS. As of this writing, the latest release from the Department of Labor covered November activity. It reported that “the quits rate increased to 3.0 percent, matching a series high last seen in September.” So we are seeing the Great Resignation in the data.

As background, the Labor Department reports three major monthly national surveys. Two of these, the Household Survey and the Establishment Survey, have been used for many decades. The JOLTS survey is more recent, with no data prior to the 21st century.

The Household Survey, conducted by the Census Bureau but reported by the Labor Department, contacts a sample of 60,000 households in order to find the employment status of the adults. This survey is used to calculate the unemployment rate.

Whereas the household survey measures employment by asking individuals if they are working, the Establishment Survey measures employment by asking businesses and other organizations how many workers they have on the payroll. Because the organizations contacted by this survey cover the vast majority of jobs, economists consider this a more reliable measure of monthly job creation. The JOLTS survey also arrives at estimates by contacting employers, but it uses a somewhat smaller sample.

When the JOLTS data were first reported, economists were struck by the large number of people moving into and out of jobs each month. They were used to following the Establishment Survey, which typically shows monthly increases in job creation of 200,000 or so. In a steep recession, jobs might decline by 400,000 in a month, and during a rapid recovery jobs might increase by a comparable amount.

But the JOLTS data show that roughly 4 million employment separations take place each month, and roughly 4 million new hires take place. In comparison, the relatively small changes in monthly payroll employment in the Establishment Survey reflect the net difference between new hires and separations. Thus, if there are 4.1 million new hires and 3.9 million separations in a month, then there is a net gain of 200,000 jobs.

Of course, separations can reflect the decisions of firms or the decisions of workers. That is, a firm can fire a worker, or the worker can quit. The JOLTS survey asks firms to report whether a separation was a layoff or a quit.

Because the JOLTS data come from surveying employers, not workers, we do not obtain any information about why workers are quitting. Are they going into retirement, trying to start a business on their own, taking a position with a different employer, or taking time off temporarily before searching for a new job?

Whatever they end up doing, workers who quit have decided that the pay and working conditions no longer motivate them to remain in their jobs. In order to forecast quit rates, you would want to predict how pay and working conditions will evolve relative to expectations.

In order to quit, workers have to be relatively optimistic about their prospects for finding another job or else living off of their financial resources. Therefore, quits are likely to be higher during a boom than during a recession.

The high quit rates in recent months likely reflect adverse changes in working conditions for many employees due to the pandemic. This is especially true for blue-collar workers. While many knowledge workers were given the opportunity to work from home in their pajamas, blue-collar workers had to show up in person, where they had to comply with protocols like mask-wearing, handwashing, distancing, and getting tested.

In some sectors, such as healthcare, work loads increased. Many employees complained of burnout.

Finally, we may be seeing the effects of inflation. In the 1970s, many workers were covered by union contracts that gave them automatic cost-of-living increases. Today, many fewer workers belong to unions. In order to get a pay increase to keep up with inflation, a worker may have to find a new employer offering a higher wage.

Will we see more of the Great Resignation in 2022? Will quits stay in the range of 4.2 to 4.6 million per month? Have they already peaked? Or are they going to rise further?

My guess is that unless there is a recession, quits will remain elevated. I am expecting inflation to remain high. I think that there are a lot of adjustments that the economy needs to make in order to relieve “shortages” in some sectors and get rid of excess capacity in others. The way these adjustments will take place in the labor market is that some employers will not be able to afford to compensate their workers for increases in the cost of living, while other employers will be desperate to add capacity and will be making attractive offers. Many workers will find it advantageous to quit and take new jobs. Meanwhile, I don't see working conditions improving next year, and so I expect rates of worker dissatisfaction and burnout to remain high.

For the monthly average of quits reported in the JOLTS survey in 2022, I think that there is a 70 percent probability that it will fall between 4.2 and 4.6 million, a 25 percent probability that it will rise over 4.6 million, and a 5 percent probability that it will fall below 4.2 million.