Rising Labor Market Anxiety and Declining Quits
October 06, 2026
One of the defining features of today's US labor market is the relatively low level of "churn," or movements in and out of jobs. Figure 1 illustrates this "low-hire-low-fire" equilibrium. The hiring rate has been falling for the past few years, while the layoffs rate has remained low. At the same time, the quits rate and the job-to-job transition rate have declined since their postpandemic highs, marking a transition from the "Great Resignation" to the "Great Stay."
A natural, unifying explanation for these recent trends is weakening labor demand. Some firms are trying to reduce their wage bills in response to rising costs. Some firms need fewer workers today—or anticipate needing fewer workers tomorrow—as a result of advances in labor saving technologies like artificial intelligence. Some firms are simply reluctant to hire new workers in the current environment because of elevated economic and policy uncertainty. Whatever the reasons, the end result is fewer job openings, which in turn means fewer workers are willing to quit their jobs because of a lack of better opportunities.
The role of rising labor market anxiety and declining quits
In this blog post, we argue that a complementary force is at play: some workers are reluctant to quit their jobs due to the elevated perceived risks of doing so in the current environment. Survey evidence from the University of Michigan, the New York Fed, and the Conference Board reveals that Americans are increasingly worried about the prospect of losing their job, their ability to find a new job, and the availability of jobs in the future, with many indicators approaching levels last seen during the Great Recession or the COVID-19 pandemic (figure 2). As a result, many workers might be reluctant to take a new job for fear of losing that job if labor market conditions were to deteriorate in the future, especially given that firms often lay off recently hired workers first. Compounding these concerns might be the fact that job loss is particularly costly in a low-hiring environment.
Consistent with these ideas, work by Clymo et al. (2026) emphasizes how risk aversion among currently employed job seekers can depress job mobility in a world with incomplete markets where job separation risk cannot be fully insured against. It is worth noting that even if workers' fears are overblown, biased labor market perceptions can still have a real impact on job search behavior. Ultimately, depressed quits as a result of rising anxiety among workers amplifies the decline in job openings due to weakening labor demand by reducing replacement hiring, which studies have shown account for roughly half of all vacancies in the United States (Mercan and Schoefer 2020, Elsby et al. 2025).
The presence of this channel rests on the notion that workers who are pessimistic about the labor market are less likely to voluntarily quit their job. The table below provides suggestive evidence on this relationship using microdata from the New York Fed Survey of Consumer Expectations (SCE), in which we can observe labor market sentiment measures and job search outcomes for a panel of respondents. More specifically, we regress job search outcomes on three measures of labor market perceptions: (1) the probability of losing one's job within 12 months, (2) the probability of not finding a new job within three months conditional on losing one's job this month, and (3) the probability that the unemployment rate will be lower 12 months from now. We focus on repeat respondents, which allows us to include individual fixed effects and leverage within-respondent variation over time. This is important to neutralize unobservable differences across individuals and ensure that our estimates do not reflect the possibility that people who are more pessimistic on average also happen to be less likely to switch jobs, for example. To neutralize the confounding role of labor demand conditions, we include census division-by-month fixed effects, which flexibly control for time-varying local labor market conditions. Lastly, to isolate the role of labor market perceptions, we control for a variety of other economic expectations regarding inflation, interest rates, stock prices, house prices, and earnings.
In the first column, the outcome is an indicator for switching employers between consecutive waves of the survey and labor market expectations are lagged. The estimate in the first row implies that a 1 percentage point increase in job loss expectations raises the probability of switching jobs by 0.017 percentage points, relative to a baseline mean of 1.6 percent, which is consistent with standard models of job search: the prospect of losing one's job makes it less attractive compared to other jobs. Note that we can only measure respondents' job loss expectations at their current job, not their expectations about the riskiness of other jobs they may be contemplating, which is likely to have the opposite effect since workers tend to value job security (Lehner et al. 2026). The estimate in the second row implies that a 1 percentage point increase in job non-finding expectations reduces the probability of switching jobs by 0.021 percentage points. There are different ways to rationalize this finding, but one possibility is that a decline in the probability of finding a job from the pool of unemployed can make workers less willing to move to jobs that they perceive to be riskier than their current job. The estimate in the third row is close to zero and statistically insignificant, implying that expectations about the aggregate unemployment rate do not have a meaningful impact on job switching after accounting for job loss and job finding expectations.
In the remaining columns of the table, we focus on the subset of respondents who appear in the New York Fed's SCE Labor Market Survey, which contains additional questions on job search behavior. In the second column, the outcome is workers' "reservation wage," or the lowest annual salary they would be willing to accept if someone offered them a job in a line of work that they would consider. In the third column, the outcome is an indicator for having engaged in any job search activity in the last four weeks. These results reveal that job loss expectations lead to lower job mobility in part by lowering workers' reservation wage and raising their job search effort, while the opposite holds for job non-finding expectations.
The findings in the table naturally raise the question: If both job loss and job finding concerns have risen in recent years and they have opposite effects on job mobility, which force has dominated? A simple back-of-the-envelope calculation suggests that if we look at the last four years (April 2022 to April 2026), the negative effect coming from the rise in job finding concerns has dominated the positive effect coming from the rise in job loss concerns, mainly because the rise in job finding concerns has been around three times as large as the corresponding rise in job loss concerns in the New York Fed's SCE (as figure 2 shows). In terms of magnitude, the combined predicted effect is a 0.17 percentage point reduction in the job-to-job transition rate, or around half of the observed decline over this period. (This calculation assumes that the job-to-job transition rate has remained flat since the last available data point for the first quarter of 2025, mirroring the flat quits rate we see in figure 1.)
These results should be viewed as suggestive and interpreted with caution. One reason is that labor market sentiment measures might partly reflect idiosyncratic labor market opportunities not captured by our control variables for common labor market conditions faced by all workers in a particular time and place. For example, if workers who face weak local demand for their individual skill set also legitimately feel discouraged about the state of the labor market, then the effects we estimate might overstate the impact of labor market sentiment on job search behavior. Nevertheless, our findings are broadly consistent with the notion that rising anxiety about the labor market among workers has had a negative effect on quits and job-to-job mobility in recent years.
Implications of declining labor market churn
What are the implications of declining churn in the labor market? First, a less fluid labor market with fewer job openings means finding a job is more difficult. Indeed, unemployment duration has been on a steady upward trajectory in recent years, as this analysis from the San Francisco Fed notes. The decline in job openings has also hurt new entrants, as evidenced by a rising unemployment rate among recent college graduates. Second, job mobility is an important source of wage growth for workers, and it also facilitates the allocation of workers to their most productive use in the economy. As a result, lower churn puts downward pressure on wage growth and productivity.
Beyond these well-understood effects, lower churn could also have more subtle consequences. In particular, "missing" quits might be hindering firms' ability to organically downsize through attrition, perhaps eventually leading to "pent-up" layoffs that could materialize in the event of a downturn. In a low-hiring environment, such excess layoffs would lead to a sharp rise in unemployment and, unless hiring picks up, a slow recovery. Lower churn can therefore make the labor market more vulnerable to adverse shocks (Rogerson 2014).
Conclusion
To summarize, in this post we argue that rising anxiety about the labor market among workers has led to lower quits and lower job-to-job mobility, amplifying the decline in churn because of weakening labor demand. For this reason, the current US labor market is best described as "low-hire-low-fire-low-quit." Although the labor market has proven to be surprisingly resilient in recent years, a low-churn environment in which many workers are reluctant to switch jobs is unlikely to be a sign of a fully healthy labor market.