Why Has the Hiring Rate Fallen?
Key Takeaways
-
The measured hiring rate understates hiring weakness since 2024.
-
The job-finding rate for unemployed workers continues its multi-year decline.
-
Hiring of the already-employed and labor force nonparticipants fell through 2024, but have since held steady or rebounded moderately.
-
Previous episodes of labor market weakening all had unique patterns of hiring deterioration, with different underlying drivers.
For some time, observers have described the labor market as “no hire, no fire”. Unemployment is still fairly low by historical standards, but it’s gotten harder and harder to find a job if you don’t already have one. In the most recent data, unemployed workers spent nearly 26 weeks on average unemployed, up from under 20 weeks in early 2023.1 What do we know about the rate of hiring and how it’s declined since the heady days of the early post-pandemic labor market?
Today we learned that February’s hiring rate—as measured in the Job Openings and Labor Turnover Survey—fell to 3.1%. This is somewhat below the average 3.3% of the 12 months prior, and far below the 2022 average of 4.2%. On jobs day later this week, we will receive data from a different source (the household survey that furnishes the unemployment rate) that yields a different estimate of the hiring rate, accompanied by richer data that enable deeper exploration. Most importantly, the household survey allows us to distinguish between new hires who already had jobs, who were previously unemployed, and who were previously outside of the labor force altogether.2
Investigating these data for prior months, we learn that while the overall hiring rate has stabilized, this hides an increasingly dismal experience for the unemployed. As the labor market has cooled, it has become more difficult for unemployed job seekers to find employment.
How hiring rates have evolved over time
The job-to-job transitions (JtJ) and nonparticipation-to-employment (NE) rates stabilized in 2024 and 2025, but the job-finding rate for unemployed workers (UE) just keeps falling. At its peak in 2022, the share of unemployed workers who found a job in the next month was almost 30%. Today, that share stands at around 24%.3 In contrast, the job-finding rate for nonparticipants (i.e., those who do not report active search and availability for work) fell through 2024 and then rebounded to an extent in 2025. See Figure 1 below.4
It is useful now to put all these trends together and see how they each contribute to the overall hiring decline. Figure 2 shows flow rate contributions (i.e., JtJ, UE, and NE rate contributions) to the decline in overall hiring.5 When the contribution of UE (for example) becomes a larger share of the chart, it means that the job-finding rate for the unemployed is falling shorter of its original level and pushing the overall hiring rate downwards.
The black solid line shows the cumulative change in the overall hiring rate.6 Counterintuitively, it falls less than the sum of the flow rate contributions (shown by the dashed blue line). The main reason is that the unemployment level rose to an appreciable extent over the last two years. Because the unemployed tend to have a much higher job-finding rate than other workers, when the unemployment level rises, we see more unemployed workers flowing into employment. However, this occurs at the same time that the rate of job-finding falls for the unemployed.
The key insights provided by Figure 2 are, firstly, that hiring deterioration in 2023 and 2024 was driven by a reduction in job hopping and a decline in how many people on the labor market sidelines were being drawn into employment. More recently, the falling rate of job-finding for unemployed workers has taken over as the main headwind for hiring. Put another way, in 2023 and 2024, labor market cooling was mostly about the employed and nonparticipants. Today, unemployed workers are bearing the brunt. Secondly, the overall hiring rate can understate labor market weakening, particularly as it manifests in the experience of unemployed workers. This appears to have been the case in 2025 and 2026.
Every labor market downturn is unique
The same approach from Figure 2 can be applied to other periods for which we have high-quality data from the household survey. Figures 3 and 4 do so for the mild recession of the early 2000s and the Great Recession of the late 2000s. In each case, the figure shows three years of hiring rate evolution after an initial base period.
Hiring deterioration in and after the early 2000s recession, associated with the dot-com bubble, was principally driven by weakening in job-finding for the unemployed and the employed. Notably, this recession featured a very mild increase in layoffs (not shown), but it did include an accumulating weakness in hiring that can be seen in Figure 3.
By contrast, hiring during and after the Great Recession was initially slowed by a reduction in how many workers came off the sidelines each month. But in 2008 and after, plummeting hiring of the unemployed became the biggest factor in hiring weakness.
The importance of thinking carefully about flows
An underappreciated virtue of the BLS jobs data that come out each month is that all the anonymized household data are made public, allowing (for example) a better understanding of the trends described above. It will be important to continue to track these trends, both to understand the full worker experience of today’s labor market and to have a better idea of where the U.S. economy is today and might be going in the near future. The Budget Lab at Yale will periodically revisit labor market flows when they have useful lessons to convey about the state of the economy.
Footnotes
- 1
Historically, the average duration of unemployment is countercyclical, with the average duration tending to be substantially higher when the unemployment rate is higher. Since early 2023, the unemployment rate has risen about 1 percentage point.
- 2
- 3
Throughout, all variables are smoothed with a six-month trailing average; underlying data are seasonally adjusted.
- 4
Because of the government shutdown that interrupted the Current Population Survey, we interpolate seasonally adjusted gross flows for the months of October through December 2025.
- 5
Those contributions are simply the difference between an actual flow rate—like the rate at which the unemployed find jobs—and that same rate’s baseline in late 2022, with a scaling adjustment that allows apples-to-apples comparisons. Everything in the graph is shown as percentage point changes from late 2022, where rates are expressed as shares of employment. The small “OE” contribution represents so-called marginal flows into employment, including for example children who age into the 16+ sample and are employed when they do so.
- 6
Overall hiring (solid line) is the measured hiring rate, expressed as all flows into employment, divided by current employment. By contrast, the dashed line implicitly shows how the overall hiring rate would have evolved if the composition of the population (across unemployment, employment, and nonparticipation) had remained constant.