Layoff and Severance Statistics (2026)

In June 2026 there were 1.8 million layoffs and discharges, a rate of 1.1% of total employment. Over the same month there were 3.2 million quits, at a 2.0% rate.

Total separations were 5.4 million, or 3.4%. Layoffs and discharges made up about a third of that; voluntary quits made up nearly three fifths.

This is the context most layoff coverage omits. In an ordinary month, far more people leave jobs by choice than are pushed, and the layoff figure has to be read against that base rather than in isolation.

JOLTS counts layoffs and discharges as involuntary separations initiated by the employer. Quits are separations generally initiated by the employee, which is why the quits rate is read as a measure of workers' willingness or ability to leave.

Other separations, which totalled 353,000 in June 2026, cover retirement, death, disability and transfers to another location of the same firm.

The displaced worker survey uses a stricter definition again: people aged 20 and over who lost or left jobs because their plant or company closed or moved, there was insufficient work, or their position or shift was abolished.

Between January 2021 and December 2023, 6.3 million workers were displaced. That splits into 2.6 million long-tenured workers, meaning three or more years in the job, and 3.7 million short-tenured.

The total fell from 8.6 million in the 2019-21 survey period, which covered the acute phase of the pandemic.

The survey has run biennially every January since 1984 as a supplement to the Current Population Survey, which makes it the longest consistent series on what actually happens to people after a job loss.

By January 2024, 65.7% of the 2.6 million long-tenured displaced workers were reemployed, essentially unchanged from 65.2% two years earlier.

16.1% were unemployed, up from 12.4% in the prior survey. The remaining 18.2% had left the labor force, down from 22.3%.

The shift between those two categories matters more than the headline. Fewer people dropped out and more were still actively looking, which is a different and generally healthier composition even though the reemployment rate barely moved.

Reemployment was 74.5% for workers aged 25 to 54, 55.3% for those aged 55 to 64, and 34.4% for those 65 and over.

The 55 to 64 rate fell by about 9 percentage points from the prior survey, the largest movement in the release.

For anyone within a decade of retirement, that figure is the strongest argument for holding a larger emergency fund than the standard advice implies, because the expected time out of work is longer and the probability of returning at all is materially lower.

Among long-tenured workers displaced from full-time wage and salary jobs and reemployed in full-time jobs by January 2024, 62% were earning as much as or more than in the job they lost. That was little different from the prior survey.

Read the other way, 38% took a pay cut, and that group is the reason a layoff is a financial event rather than only an employment one.

The figure also excludes anyone who moved to part-time work or did not return at all, so it describes the best-case population rather than the average outcome.

46% of long-tenured displaced workers received written advance notice that their jobs would end, up from 39% in the previous survey.

Notice depends heavily on why the job went. 61% were notified when a plant or company closed or moved, up from 49%. 42% were notified when a position or shift was abolished, and only 29% when the reason was insufficient work.

The pattern follows the law: closures are the case the WARN Act most clearly covers, and a quietly abolished role frequently is not.

The Worker Adjustment and Retraining Notification Act requires employers with 100 or more employees to give at least 60 calendar days of written advance notice of a plant closing or mass layoff affecting 50 or more employees at a single site.

The 100-employee count generally excludes those who have worked less than six months in the last 12 and those averaging under 20 hours a week.

It carries exceptions for unforeseeable business circumstances, faltering companies and natural disasters, which is a significant part of why the observed notice rate is 46% rather than something closer to universal.

Severance is not required. The Department of Labor states that there is no requirement in the Fair Labor Standards Act for severance pay, and that severance is a matter of agreement between an employer and an employee or their representative.

That makes any severance you receive contractual rather than statutory, which is why it can be conditioned on signing a release and why the terms differ so widely between employers.

Where an employer-sponsored plan promises severance and does not pay it, the Employee Benefits Security Administration is the agency that may be able to assist, because the promise then sits inside benefits law rather than wage law.

There is no federal statistical series that reports how many laid-off workers receive severance, or how large it is. That absence is itself the finding.

Because severance is contractual, it is not collected the way wages, benefits access or unemployment claims are. Figures that circulate on this come from consultancies and employer surveys with their own samples and definitions.

This page does not quote them. A page built to be cited should not launder a private survey into a national statistic, and the honest statement is that the federal data covers notice and reemployment but not severance size.

In the week ending 15 August 2026, seasonally adjusted initial unemployment claims were 206,000, down 6,000 on the week. Insured unemployment was 1,799,000, an insured unemployment rate of 1.2%.

Unadjusted initial claims were 172,080, against 194,217 in the comparable week of 2025.

Continued weeks claimed across all programmes were 1,839,126. Insured unemployment counts people actively claiming, which is a narrower group than everyone out of work, so it understates the population affected.

17% of long-tenured displaced workers lost a job in manufacturing, 15% in professional and business services, 10% in retail trade and 10% in health care and social assistance.

Manufacturing's share is larger than its share of employment, which is the long-running pattern in this survey.

Professional and business services at 15% is the figure that has changed most in character, because it covers much of the white-collar work that layoff coverage now focuses on.

Reemployment was 65.2% for men and 66.2% for women, both little changed.

The unemployment share differed: 14.9% for displaced men against 17.6% for displaced women, the latter about 6 percentage points higher than the prior survey.

Meanwhile the share of displaced women who had left the labor force fell about 11 points to 16.2%, and for women aged 25 to 54 it fell about 13 points to 8.8%. More displaced women stayed in the market and were counted as looking rather than leaving.

The standard advice of three to six months of essential expenses is built for the median case. This data shows where the median is a poor guide.

A worker aged 55 to 64 faces a 55.3% reemployment rate, and 34.4% at 65 and over, so the expected duration out of work is longer at exactly the age when income is hardest to replace.

Someone in manufacturing or professional services, without written notice, and near the top of a pay band, is carrying several of these risks at once. That combination argues for the upper end of the range rather than the lower.

Federal data is authoritative and slow. The displacement survey runs every two years and the most recent covers 2021 to 2023, so it cannot describe what is happening now. Private trackers fill that gap, and they are worth reading with their limits attached.

layoffs.fyi, which has tracked tech and startup layoffs since 2020, counted 127,180 tech employees laid off across 281 companies in 2026 to date. That already exceeds the 122,606 it recorded across 278 companies in the whole of 2025.

The longer series shows how unusual 2023 was: 265,660 tech employees across 1,194 companies, roughly double 2024 and more than twice 2025. The company count has fallen much faster than the headcount, which means recent cuts are concentrated in fewer, larger employers.

Challenger, Gray & Christmas, which has published its own monthly count of announced US job cuts for decades, reported 33,429 in July 2026, the lowest monthly total in two years. June was 45,849, down 53% from May, and May was 97,006, the highest May total since 2020.

Hiring is the other half of that picture. Challenger reported announced hiring plans of 16,095 in July, and its January 2026 report recorded the lowest January hiring plans on record. Its 2025 year-end report described the highest fourth-quarter layoffs since 2008 alongside the lowest year-to-date hiring since 2010.

That combination, falling layoffs and falling hiring at the same time, is the pattern Challenger calls a low-fire, low-hire environment. For somebody who loses a job in it, the risk is not that layoffs are unusually common. It is that replacing the job takes longer.

Challenger attributes announced cuts to a stated reason, and artificial intelligence led all reasons for the fifth consecutive month in July 2026, at 10,970 cuts.

That is a claim about what employers say when they announce a reduction, which is not the same as a measurement of what caused it. Restructuring, cost-cutting and automation overlap, and a reason given in a press release is a communications decision as much as an economic one.

It is still the first time this reason has led the series for a sustained run, and it is the clearest available signal of how employers are choosing to explain reductions in 2026.

Trackers count announcements. BLS counts separations. An announced cut may take months to happen, may be partly absorbed by attrition, and may never reach the headcount announced. The two series are not interchangeable and should never be added together.

layoffs.fyi covers tech and startups, not the economy. Its 127,180 figure for 2026 sits against roughly 1.8 million layoffs and discharges in a single month across all US employers, which is the scale difference worth holding in mind.

Both are compiled from public announcements, so both are biased toward large, public, newsworthy employers. A small firm cutting five people appears in the BLS data and in neither tracker.

Monthly separations figures are from the BLS Job Openings and Labor Turnover Survey news release. Displacement, reemployment, notice and earnings figures are all from BLS Worker Displacement: 2021-2023, released 29 August 2024 as USDL-24-1777.

Severance and notice law is quoted from the Department of Labor's own pages on severance pay and on plant closings and layoffs.

Unemployment insurance figures are from the Department of Labor's weekly claims release for the week ending 15 August 2026.

Tracker figures are attributed to the organisations that compile them: layoffs.fyi for tech and startup layoffs, and Challenger, Gray & Christmas for announced US job cuts. Both are primary for their own datasets and neither is a national layoff count, which is why they are reported separately from the federal series rather than blended into it.