Author
Article Highlights
- Half of people who quit are entering a period of nonemployment, with higher rates for lower-wage workers
- Unlike some European countries, U.S. unemployment insurance does not cover those who voluntarily quit
- New research models the costs and benefits of covering job quitters equally or at a lower wage replacement rate
As a general rule, saying “take this job and shove it” means waving bye-bye to UI.
Although it has economic implications, the policy decision to exclude quitters from unemployment insurance (UI) is partly—maybe mostly—a moral one. “I think people’s intuition is that if the worker was laid off, that was not the worker’s fault,” said Minneapolis Fed Monetary Advisor Jonathan Heathcote. “But if he quit, that was his choice and maybe we don’t feel sorry for him.”
In a new Minneapolis Fed staff report, Heathcote explores the incentives for firms and workers if job quitters were included in unemployment benefits (“The Great Resignation and Optimal Unemployment Insurance” with Zhifeng Cai). It’s not such a crazy notion. Many European countries impose waiting periods but otherwise allow job quitters to collect benefits. Even in the U.S., state-level “good cause” exceptions mean some job quitters still qualify if they can point to a personal or workplace issue that motivated the decision.
The economists’ model dispassionately considers the costs and benefits of supporting quitters through UI policy, focusing on the economics rather than the question of which quitters “deserve” benefits.
Many Americans are quitting without their next gig
As job-quitting spiked in the post-COVID recovery, the monthly U.S. “quits rate” has been in the spotlight as an indicator of labor market tightness. “But quits had been trending upwards for a longer time before the pandemic, ever since the Great Recession,” said Heathcote (Figure 1).
UI is only relevant, of course, to the extent that quitters are becoming unemployed (and not just jumping to a new job). The quits rate reported by the Bureau of Labor Statistics’ Job Openings and Labor Turnover Survey (JOLTS) does not make this distinction.
But by combining data from the U.S. census and JOLTS, Heathcote and Cai estimate that from 2000 to 2021, around half of all job quits were quits to nonemployment. After adjusting the JOLTS data to account for undercounting of certain workers and high-turnover businesses, the economists calculate that in any given month, roughly 1 in 50 American workers quits a job with nowhere immediate to go.1
The data on U.S. quitting reveal that lower-wage jobs are associated with much higher rates of quitting (Figure 2). The economists use their model to further decompose these quitting trends and reveal an important pattern behind the raw data (Figure 3).
The results show that while all sectors experience job-to-job quitting at similar rates, workers in lower-earning industries are much more likely to quit into nonemployment.
On the one hand: Quitting hurts the economy
Quitting is typically costly for employers. When a worker quits, “it is expensive to get a new worker in place,” said Heathcote. It takes money to recruit and train, and “firms only recoup that investment over time as the worker produces for them and adds value.”
Firms would love to know which workers are on the verge of quitting. If they could, Heathcote said, they could take action and pay those workers a bit more to keep them, saving the expense of hiring a replacement. But firms are often blindsided by a quit, which leads to economic inefficiency that hurts everyone.
“Because of these information frictions, workers tend to quit too often, no matter what the unemployment benefits are,” said Heathcote. Unemployment insurance boosts this already inefficient “quitting margin” by making it less costly for workers to quit. A key outcome in the economists’ model is that this excess quitting depresses wages for all workers across the economy.
“Firms are going to understand that workers are going to quit at a higher rate, and they’re not going to last that long” in their jobs, Heathcote said. The wage effect might be small for any given position, but it is a significant cost to society when spread across millions of workers.
On the other hand: Quitting helps the economy
This widespread wage effect is set against the reasons we might want to support workers’ ability to quit.
When workers experience a health, caregiving, or personal crisis, the best choice could be to leave work temporarily. Public policy might want to provide UI payments to help smooth their income through the shock and provide support as they hunt for a new or more suitable position. Providing quitters some support can also empower people to transition from a bad workplace. These notions are embedded in the “good cause” provisions in many states, although their generosity varies widely.2
Another major economic value of quitting is the opportunity to make a better match. “Some firms and some workers are just better suited for each other,” Heathcote said. “If they can find each other, both sides are going to be happy. But it takes time to figure that out.” Better matches make for a stronger, faster-growing economy.
Being unemployed might actually provide a better environment for people to find an improved match, especially for lower-wage workers with limited time and resources to search while working. UI benefits typically include access to state job centers and counseling.
Giving (less?) UI to quitters
Heathcote and Cai’s economic model does not distinguish between “good cause” quitters and more opportunistic ones. This might not be such a strong assumption in the model: Some state rules leave wide room for “good cause” interpretation, creating a powerful incentive to adapt one’s quitting narrative to qualify.3 The gray area is large.
The economists’ policy lever in their model is the size of the benefit—the wage replacement rate (a proportion of former salary that each unemployed worker receives). What is the optimal rate that balances the pros, cons, and incentives around quitting?
The economists begin with a scenario where the government does not distinguish job quitters from laid-off workers. In a baseline simulation where all unemployed workers are laid off (a hypothetical economy with no quitting) the optimal wage replacement rate is 49 percent. When quitters are in the mix, however, the optimal rate falls. When the economists set the proportion of quitters to reflect recent U.S. data, it is optimal to pay all claimants a wage replacement benefit of 38 percent.4
What if the government could distinguish the quitters? In this case, the model finds it is still optimal to pay benefits to quitters, but at a lower wage replacement rate than is paid to nonquitters (30 percent versus 49 percent). “Giving nothing to quitters is unlikely to be the optimal policy,” said Heathcote, given the economic benefits of subsidizing some degree of quitting.5
The economists calculate that a system which distinguishes quitters from others, allocating different benefits to each, is better for society overall. However, they note that this does not account for the administrative burden of trying to credibly sort out quitters—who would, after all, have a strong incentive to try to qualify for the higher benefit.
Given that today’s actual UI system also attempts to identify and exclude certain quitters, it might be better to bring all quitters into the fold but pay a lower benefit to all.
Policy proposal has implications for job quitters
A proposal to overhaul the 90-year-old UI program—opening the doors wide for job quitters—seems unlikely in the near term. However, a group of U.S. lawmakers have introduced the Unemployment Insurance Modernization and Recession Readiness Act of 2023, elements of which resonate with these research findings.
Among other changes (such as allowing UI benefits to automatically expand in times of recession) the proposal would provide a new “jobseeker allowance” of $250 a week “to unemployed workers who are seeking work but are not covered by unemployment insurance.” The stated intent of the provision is to support self-employed people and new workers who lack an established work history. However, the jobseeker allowance would also be a de facto, lower level of unemployment benefits available to voluntary job quitters.
The Minneapolis Fed research model does not, in its initial version, consider whether it makes sense to distinguish some quitters from other quitters. This is, however, a practical question for any update to current U.S. policy. The proposed UI reform bill would create a national baseline for “good cause” quitting across the U.S., addressing today’s wide variety of state rules. It spells out certain reasons that would allow the worker to claim UI, including illness, disability, loss of child care, and risk to health or safety.
How should we handle quitters and UI? A policy informed by economic analysis might look different—and simpler—than the patchwork policies of today. This new research offers some guidance on the pros and cons of extending benefits to quitters, and on exactly how generous those benefits should be.
Endnotes
1 While official JOLTS quits rates between 2000 and 2021 span a range of about 1.2 to 3 percent (the high point after the COVID-19 pandemic), the economists adopt the reasoning and adjustment factors from prior research that asserts these numbers undercount quits from new establishments and establishments that have failed. As a result, they calculate an average monthly quits rate of 3.7 percent in recent decades, and quits-to-nonemployment rate of 1.9 percent.
2 It is important to note that workers claiming UI benefits must also be “able and available” for work, which can be an obstacle to benefits even if quitting for personal reasons is otherwise justified under a state’s provisions.
3 The Minneapolis Fed’s home state of Minnesota provides a useful example, where benefits are available if a worker quits for anything “that would compel an average reasonable worker to quit.” Mississippi, by contrast, lacks any “work-connected good-cause provision,” according to the U.S. Labor Department.
4 According to the U.S. Labor Department, the effective replacement rate in the U.S. today averages around 40 percent. This average masks wide variety among states.
5 The model is sensitive to a parameter that estimates the “cost of work” experienced by quitters (how inclined they are to quit, even without benefits). In an alternate calibration, the economists find a much lower replacement rate for quitters, although still above zero.
Jeff Horwich is the senior economics writer for the Minneapolis Fed. He has been an economic journalist with public radio, commissioned examiner for the Consumer Financial Protection Bureau, and director of policy and communications for the Minneapolis Public Housing Authority. He received his master’s degree in applied economics from the University of Minnesota.
