US Employee Turnover Statistics 2026
By Axis Intelligence Research
Co-author: David Park (Careers & Workforce) | Last updated: August 8, 2026 | License: CC BY 4.0
American workers quit 3.23 million jobs in June 2026, a quits rate of 2.0% that has now held flat for three consecutive months and sits a full percentage point below the Great Resignation peak. The turnover slowdown looks like retention. Axis Intelligence Research finds it is something else: an exit market that stopped clearing.
Quick Answer
The U.S. employee turnover rate, measured as total separations, was 3.4% in June 2026 — roughly 5.4 million separations in a single month. Across all of 2025 there were 62.8 million separations against 63.0 million hires, and Axis Intelligence Research calculates that works out to 39.6 separations per 100 payroll jobs, down from 46.8 in 2021. Voluntary quits made up 60.6% of all separations in 2025, the lowest share in five years. Median job tenure stands at 3.9 years, the shortest reading since 2002.
Key Findings
- According to Axis Intelligence Research analysis of BLS JOLTS data, the U.S. economy recorded 39.6 separations per 100 payroll jobs in 2025, down from 46.8 per 100 in 2021 — a 15.4% contraction in workforce churn over four years.
- Axis Intelligence Research’s Turnover Control Index (TCI) reads 64.7 for total nonfarm employment in June 2026, meaning roughly two-thirds of separations were worker-initiated and one-third employer-initiated.
- The Information sector — software, telecom, media, data services — posted a TCI of 35.6 in June 2026, the lowest of any major U.S. industry and down 15.0 points year over year, making it the only large sector where most exits are employer decisions.
- Voluntary quits fell to 60.6% of all U.S. separations in 2025, down from 69.9% in 2022, while layoffs and discharges rose to 33.8% — a 21.2 million annual total.
- Gallup’s 2026 World Poll data put U.S. and Canada job market optimism at 47%, down 23 percentage points since 2019, the second-lowest of any world region, even as quit rates fell to decade lows.
What Is the Employee Turnover Rate in 2026?
Two numbers answer this, and they answer different questions.
The monthly total separations rate was 3.4% in June 2026, per the Bureau of Labor Statistics Job Openings and Labor Turnover Survey. That covers every payroll exit: quits, layoffs and discharges, and other separations such as retirement, death, and intra-firm transfers. In levels, 5.4 million separations against 5.3 million hires.
The annual turnover rate is the figure HR teams actually benchmark against, and it is much larger. BLS published 62.8 million total separations for calendar 2025 at an annual average rate of 3.3% per month.
| Measure | 2021 | 2022 | 2023 | 2024 | 2025 | Source |
|---|---|---|---|---|---|---|
| Annual hires (millions) | 76.2 | 76.8 | 70.3 | 64.5 | 63.0 | bls.gov |
| Annual separations (millions) | 68.9 | 72.3 | 67.9 | 63.1 | 62.8 | bls.gov |
| Annual quits (millions) | 47.6 | 50.5 | 44.2 | 39.3 | 38.0 | bls.gov |
| Separations per 100 jobs (Axis) | 46.8 | 46.8 | 43.2 | 39.6 | 39.6 | Axis Intelligence Research |
| Quits per 100 jobs (Axis) | 32.4 | 33.6 | 28.8 | 25.2 | 24.0 | Axis Intelligence Research |
The Axis annualization: separations per 100 jobs
BLS publishes turnover as a monthly rate, which understates how much movement a year contains. Axis Intelligence Research converts it using the survey’s own definition:
Separations per 100 jobs = annual average monthly separations rate × 12
For 2025: 3.3 × 12 = 39.6 separations per 100 payroll jobs. For 2021: 3.9 × 12 = 46.8.
This is exact rather than approximate, because the BLS annual average rate is itself computed as the sum of twelve monthly separations levels divided by the sum of twelve monthly employment levels. Multiplying by twelve returns the annual event count per average job.
One reading discipline matters here. These are separation events, not distinct workers — a person who leaves two jobs in a year appears twice, and a seasonal position that turns over three times contributes three events. The measure describes how much the payroll moved, not what share of the workforce changed employers.
David Park’s read: Every HR dashboard reports turnover as an annual percentage, and every one of them quietly implies “share of our people who left.” The BLS number does not mean that, and neither does the internal number in most cases. A 39.6% churn rate on a 200-person company is 79 separation events — which might be 60 people, of whom 15 were the same three seasonal roles cycling. Before benchmarking against the national figure, check whether your denominator is average headcount and whether your numerator counts events or humans. Half the turnover comparisons circulating in comp committees are unit-mismatched.
Source: Axis Intelligence Research — CC BY 4.0
How Do You Calculate Employee Turnover Rate?
The standard formula used by BLS and mirrored in most HRIS platforms:
Turnover rate = (separations during the period ÷ average employment during the period) × 100
Worked example using a 500-person company over twelve months, with 62 total departures:
62 ÷ 500 × 100 = 12.4% annual turnover
To split it the way the national data splits:
- Voluntary turnover rate = quits ÷ average employment. If 40 of the 62 were resignations: 40 ÷ 500 × 100 = 8.0%
- Involuntary turnover rate = layoffs, discharges, and terminations ÷ average employment. If 18 of the 62: 18 ÷ 500 × 100 = 3.6%
- The remaining 4 — retirements, deaths, transfers — map to what BLS calls other separations: 0.8%
Average employment is the part that gets fudged. Using starting headcount inflates the rate in a shrinking company and deflates it in a growing one. The defensible method is the mean of monthly headcounts, which is what BLS benchmarks to through the Current Employment Statistics program.
What is a good employee turnover rate?
There is no universal threshold, and any source quoting one without an industry qualifier is guessing. The honest benchmark is your sector’s published quits rate, because that is the portion of turnover you can plausibly influence.
June 2026 quits rates, seasonally adjusted, run from 0.5% in federal government to 4.5% in accommodation and food services — a ninefold spread across a single month. A 3% annual voluntary turnover rate is alarming in finance and implausible in fast food.
Employee Turnover Rate by Industry
Quits and layoffs move independently, and reading only one of them produces a false picture of an industry’s labor market.
| Industry | Quits rate | Layoffs rate | Quits (thousands) | Layoffs (thousands) | Source |
|---|---|---|---|---|---|
| Accommodation and food services | 4.5% | 1.1% | 638 | 158 | bls.gov |
| Leisure and hospitality | 4.2% | 1.5% | 704 | 253 | bls.gov |
| Retail trade | 3.0% | 1.0% | 471 | 157 | bls.gov |
| Other services | 2.8% | 0.9% | 168 | 56 | bls.gov |
| Trade, transportation, utilities | 2.5% | 1.1% | 732 | 312 | bls.gov |
| Professional and business services | 2.2% | 2.2% | 488 | 504 | bls.gov |
| Health care and social assistance | 1.9% | 0.6% | 443 | 151 | bls.gov |
| Construction | 1.7% | 2.0% | 143 | 167 | bls.gov |
| Manufacturing | 1.5% | 0.7% | 192 | 94 | bls.gov |
| Information | 1.1% | 2.0% | 31 | 56 | bls.gov |
| Financial activities | 1.0% | 0.5% | 91 | 47 | bls.gov |
| Government | 0.8% | 0.4% | 186 | 85 | bls.gov |
All figures June 2026, seasonally adjusted. Quits from JOLTS Table 4; layoffs from JOLTS Table 5.
Three industries invert the usual pattern. In Information, Construction, and Professional and business services, layoffs equal or exceed quits. Everywhere else, workers are the ones ending the relationship.
The four-year decline explains where the churn went:
| Industry | 2022 annual quits | 2025 annual quits | Change | Source |
|---|---|---|---|---|
| Leisure and hospitality | 10,344,000 | 7,959,000 | −23.1% | bls.gov |
| Trade, transportation, utilities | 11,210,000 | 7,678,000 | −31.5% | bls.gov |
| Professional and business services | 9,029,000 | 6,288,000 | −30.4% | bls.gov |
| Retail trade | 7,337,000 | 4,759,000 | −35.1% | bls.gov |
| Manufacturing | 3,559,000 | 2,099,000 | −41.0% | bls.gov |
| Health care and social assistance | 6,414,000 | 5,639,000 | −12.1% | bls.gov |
| Information | 612,000 | 437,000 | −28.6% | bls.gov |
| Government | 2,711,000 | 2,254,000 | −16.9% | bls.gov |
Manufacturing shed 41% of its voluntary quits in three years — the steepest drawdown of any major sector, and a reminder that a falling quits rate is as often a symptom of vanished outside options as of improved conditions.
David Park’s read: Health care held up best, down only 12%, and that is not because hospitals fixed their retention problems. It is because clinical roles kept posting. Health care and social assistance still carried 1.41 million job openings in January 2026 against a 5.6% openings rate. Where the reqs stayed open, the quits stayed up. That relationship is the single most useful thing a workforce planner can take from this dataset: your voluntary attrition is largely a function of your competitors’ hiring plans, not your engagement survey scores.
Voluntary vs Involuntary Turnover: The Turnover Control Index (TCI)
The gap this dataset fills is that no public source publishes the composition of turnover by industry — the split between workers choosing to go and employers deciding for them. Axis Intelligence Research built the Turnover Control Index (TCI) to close it.
TCI formula and inputs
TCI = quits ÷ (quits + layoffs and discharges) × 100
Inputs are JOLTS seasonally adjusted monthly levels, drawn from Table 4 for quits and Table 5 for layoffs and discharges. Other separations — retirement, death, disability, transfer within a firm — are excluded because neither party chose them as a labor-market decision, and including them would mechanically raise the reading in industries with older workforces.
The scale runs 0 to 100. A reading of 50 is parity: half of exits are the worker’s decision, half the employer’s. Above 50, workers hold the exit. Below 50, employers do.
National TCI, June 2026: 3,232 ÷ (3,232 + 1,766) × 100 = 64.7
For calendar 2025 on the annual not-seasonally-adjusted series: 38,029 ÷ (38,029 + 21,200) × 100 = 64.2.
TCI by industry, June 2026
| Industry | TCI June 2025 | TCI June 2026 | Change | Quits per layoff | Source |
|---|---|---|---|---|---|
| Accommodation and food services | 84.9 | 80.2 | −4.7 | 4.04 | Axis Intelligence Research |
| Other services | 72.7 | 75.0 | +2.3 | 3.00 | Axis Intelligence Research |
| Retail trade | 72.4 | 75.0 | +2.6 | 3.00 | Axis Intelligence Research |
| Health care and social assistance | 74.2 | 74.6 | +0.4 | 2.93 | Axis Intelligence Research |
| Leisure and hospitality | 78.0 | 73.6 | −4.5 | 2.78 | Axis Intelligence Research |
| Trade, transportation, utilities | 68.1 | 70.1 | +2.0 | 2.35 | Axis Intelligence Research |
| State and local government | 68.1 | 68.8 | +0.7 | 2.20 | Axis Intelligence Research |
| Manufacturing | 62.6 | 67.1 | +4.5 | 2.04 | Axis Intelligence Research |
| Financial activities | 66.3 | 65.9 | −0.3 | 1.94 | Axis Intelligence Research |
| Total nonfarm | 63.8 | 64.7 | +0.8 | 1.83 | Axis Intelligence Research |
| Finance and insurance | 72.0 | 62.9 | −9.1 | 1.69 | Axis Intelligence Research |
| Transportation, warehousing, utilities | 64.1 | 59.8 | −4.3 | 1.49 | Axis Intelligence Research |
| Mining and logging | 59.1 | 58.3 | −0.8 | 1.40 | Axis Intelligence Research |
| Private educational services | 71.4 | 56.9 | −14.5 | 1.32 | Axis Intelligence Research |
| Professional and business services | 42.3 | 49.2 | +6.9 | 0.97 | Axis Intelligence Research |
| Construction | 44.2 | 46.1 | +1.9 | 0.86 | Axis Intelligence Research |
| Arts, entertainment, recreation | 38.0 | 41.0 | +3.0 | 0.69 | Axis Intelligence Research |
| Information | 50.6 | 35.6 | −15.0 | 0.55 | Axis Intelligence Research |
Computed from BLS JOLTS Tables 4 and 5, seasonally adjusted. Baseline reading; no prior series exists.
What the readings say
Information is the outlier of 2026. At 35.6, the sector that houses software publishers, telecom, and data services now sees nearly two exits in three initiated by the employer. A year earlier it sat at 50.6 — dead even. That 15-point swing is the largest of any major industry and the clearest labor-market signal in the June release: 31,000 quits against 56,000 layoffs in a single month.
Finance and insurance dropped 9.1 points, from 72.0 to 62.9, as quits fell from 85,000 to 61,000 while layoffs rose from 33,000 to 36,000. The sector is still worker-controlled, but less so than at any point in the past year.
Professional and business services crossed back toward parity at 49.2, up from 42.3, which reads as improvement only in context: layoffs there still outnumber quits, 504,000 to 488,000.
By region, the West posted the sharpest improvement — TCI 65.1 in June 2026 against 60.0 a year earlier — while the Northeast slipped to 58.5, the lowest of the four census regions.
TCI scope
The index measures composition, not volume. An industry can post a high TCI while total turnover collapses, which is precisely what accommodation and food services did: 80.2 is a worker-controlled reading, but annual quits there fell 23% from 2022. TCI answers “who is deciding,” not “how much is happening,” and the two questions need the quits rate alongside it. Monthly JOLTS estimates also carry sampling error and are revised; readings should be compared across quarters rather than month to month.
Methodology version: TCI 1.0, published August 2026. Components, weights, and normalization are stated above in full and will not be changed retroactively.
How Long Do Employees Stay With One Employer?
Median tenure for U.S. wage and salary workers was 3.9 years in January 2024 — down from 4.1 years in 2022 and the shortest since January 2002, per the BLS Employee Tenure release. [Older data: the tenure supplement runs biennially in January; the January 2026 collection has not yet been published, making the 2024 reading the current official figure.]
| Segment | Median tenure (years) | Source |
|---|---|---|
| Public sector | 6.2 | bls.gov |
| Mining, quarrying, oil and gas | 5.7 | bls.gov |
| Management occupations | 5.7 | bls.gov |
| Manufacturing | 4.9 | bls.gov |
| Financial activities | 4.7 | bls.gov |
| All workers | 3.9 | bls.gov |
| Private sector | 3.5 | bls.gov |
| Service occupations | 2.7 | bls.gov |
| Leisure and hospitality | 2.1 | bls.gov |
| Food preparation and serving | 2.0 | bls.gov |
Public-sector workers stay 1.77 times as long as private-sector workers. Workers aged 55 to 64 post a median of 9.6 years against 2.7 years for those aged 25 to 34 — a gap driven substantially by age composition rather than by generational loyalty, since a 28-year-old cannot have twenty years of tenure. And 22% of all workers had been in their job a year or less in January 2024, down from 24% in 2022, which is itself a churn signal: fewer recent movers means fewer people moved.
The tenure figures cross-check the JOLTS story from a different survey. The Current Population Survey asks households; JOLTS asks establishments. Both point the same direction — shorter attachments in the private sector, longer in government, with leisure and hospitality as the structural outlier at roughly a third of the public-sector median.
Why Is Turnover Falling if Workers Are Unhappy?
This is where the retention narrative breaks.
Gallup’s State of the Global Workplace 2026 report, built on 141,444 employed respondents surveyed across 2025, found global employee engagement at 20% — down from a 23% peak in 2022 and the lowest since 2020, the first back-to-back annual decline Gallup has recorded. Manager engagement fell hardest, from 31% in 2022 to 22% in 2025, with the sharpest drop between 2024 and 2025.
Set that against the exit data. Quits are at a decade low. Engagement is at a five-year low. Both cannot be evidence of retention.
The reconciling variable is opportunity. Gallup’s job-climate measure puts U.S. and Canada optimism at 47% in 2025, down from 70% in 2019 — a 23-point collapse that leaves the region second-to-last worldwide. Workers are not staying because conditions improved. They are staying because the door on the other side is heavier.
Layoff expectations reinforce it: 18% of U.S. employees told Gallup in Q1 2026 that they thought technological change would likely eliminate their job within five years, rising to 23% among employees at organizations that have deployed AI, and to 32% in finance and 31% in technology. Those are the same sectors posting the weakest TCI readings — a pattern also visible in the workforce data in our AI statistics analysis.
David Park’s read: The dangerous conclusion for a CHRO in 2026 is that low attrition validates the retention program. It does not. Suppressed quits are stored demand — a queue, not a resolution. When openings recover, the queue clears, and it clears fastest among exactly the people with the most options. The organizations that will get hurt are the ones that read a 2.0% quits rate as permission to defer comp reviews and manager training. Gallup’s own counterexample is the number worth pinning up: in best-practice organizations, 79% of managers were engaged, against 22% globally. That gap is not economic conditions. That gap is management.
The tightening is not uniform. Hiring has fallen alongside quits — 63.0 million hires in 2025 against 76.8 million in 2022, a drop of 18% — so the churn contraction is symmetrical. Employers are neither shedding nor absorbing at 2022 rates, and the June 2026 release described hires, separations, quits, and layoffs all as unchanged. That stasis is what a frozen market looks like in a statistical release.
What Does Employee Turnover Cost?
Published turnover-cost multiples vary so widely — anywhere from a fraction of salary to several times it, depending on role and methodology — that averaging them produces a number meaning nothing. Axis Intelligence Research does not publish a blended cost-per-departure figure for that reason, and treats any single “cost of turnover” percentage circulating without its underlying role mix and cost categories as unusable for benchmarking.
What can be sourced is the exposure. Gallup estimates that low engagement cost the world economy roughly $10 trillion in lost productivity in 2025, equal to 9% of global GDP, and states that this estimate deliberately excludes separately-counted turnover, safety, theft, and healthcare costs. Read as a floor rather than a total.
The defensible internal calculation runs on your own inputs: separations in the period, split voluntary and involuntary, multiplied by role-specific replacement cost built from recruiting spend, vacancy days at loaded rate, and time-to-productivity. Modern HRIS platforms expose those fields directly — a topic covered in our HR software evaluation — and a number built that way survives a CFO review in a way a borrowed multiple does not.
Turnover by Region and Establishment Size
Regional TCI readings, June 2026, computed by Axis Intelligence Research from JOLTS Tables 4 and 5:
| Region | Quits (thousands) | Layoffs (thousands) | TCI June 2026 | TCI June 2025 | Source |
|---|---|---|---|---|---|
| South | 1,347 | 643 | 67.7 | 66.6 | Axis Intelligence Research |
| West | 775 | 416 | 65.1 | 60.0 | Axis Intelligence Research |
| Midwest | 677 | 399 | 62.9 | 65.8 | Axis Intelligence Research |
| Northeast | 434 | 308 | 58.5 | 60.4 | Axis Intelligence Research |
Establishment size cuts the other way from what most retention literature assumes. In January 2026, private establishments with 5,000 or more employees posted a quits rate of 0.7%, against 2.8% at establishments with 10 to 49 employees — a fourfold difference. Large employers lose a far smaller share of their people voluntarily each month, and the effect is visible in every size band in between.
Methodology
Collection. Every figure in this dataset was retrieved on August 8, 2026 from a primary source document opened during production. Monthly turnover figures come from the BLS Job Openings and Labor Turnover Survey release for June 2026 (USDL-26-1289, published August 4, 2026) and its Tables 4 and 5. Annual 2021–2025 series come from the JOLTS January 2026 release (USDL-26-0439, published March 13, 2026), which carries the 2025 annual estimates in Tables 15 through 26. Tenure figures come from the BLS Employee Tenure release (USDL-24-1971). Engagement and job-climate figures come from Gallup’s State of the Global Workplace 2026 report page.
JOLTS survey design. JOLTS samples approximately 21,000 nonfarm establishments, stratified by ownership, region, industry sector, and size class, benchmarked monthly to Current Employment Statistics employment. Estimates are subject to sampling and nonsampling error; BLS analyses run at 90% confidence. Monthly figures are revised, and five years of data are revised annually with the January release.
Formulas disclosed in full.
- Turnover Control Index (TCI) = quits ÷ (quits + layoffs and discharges) × 100, on seasonally adjusted monthly levels. Other separations excluded.
- Separations per 100 jobs = annual average monthly separations rate × 12.
- Quits per 100 jobs = annual average monthly quits rate × 12.
- Quits share of separations = annual quits level ÷ annual total separations level × 100. This method reproduces the BLS-published 60.6% for 2025 exactly, which is the validation check for the series.
- Quits-to-layoffs ratio = quits ÷ layoffs and discharges.
All arithmetic was independently recomputed before publication.
Scope of the dataset. Coverage is U.S. nonfarm payroll employment except where a row is marked global (Gallup). JOLTS excludes agricultural workers, the self-employed, and workers placed by temporary help agencies at the client establishment; those are counted at the employer of record. Separations count events rather than distinct individuals. Two surveys with different frames are used side by side — JOLTS is establishment-based, the tenure supplement is household-based — and their figures are reported separately rather than merged, because the methodologies are not compatible for averaging. The tenure supplement is biennial, so the current official reading dates to January 2024.
Cross-source restraint. No figure in this dataset blends a survey-based estimate with an administrative count. Where two sources describe the same phenomenon differently, both are published with their methodology named.
About This Dataset
Title: Employee Turnover Statistics 2026 — U.S. Quits, Layoffs, Tenure, and the Turnover Control Index
Publisher: Axis Intelligence Research
Coverage: United States nonfarm payroll employment, plus global engagement indicators; monthly data June 2025 and June 2026, annual data 2021–2025, tenure data January 2020–January 2024
Records: 419 rows, each carrying source organization, source document, URL, retrieval date, primary-source flag, and calculation method where applicable
Format: UTF-8 CSV, long format, one observation per row
License: CC BY 4.0 — free to reuse, redistribute, and build on with attribution
Citation
APA Axis Intelligence Research. (2026). Employee turnover statistics 2026: Quit rates, layoffs, tenure, and the Turnover Control Index. https://axis-intelligence.com/employee-turnover-statistics/
MLA Axis Intelligence Research. “Employee Turnover Statistics 2026: Quit Rates, Layoffs, Tenure, and the Turnover Control Index.” Axis Intelligence, 8 Aug. 2026, axis-intelligence.com/employee-turnover-statistics/.
Chicago Axis Intelligence Research. “Employee Turnover Statistics 2026: Quit Rates, Layoffs, Tenure, and the Turnover Control Index.” Axis Intelligence. August 8, 2026. https://axis-intelligence.com/employee-turnover-statistics/.
Frequently Asked Questions
Does the JOLTS quits rate measure the same thing as my company’s voluntary turnover rate?
Close, but not identical. JOLTS quits exclude retirements and transfers to another location of the same firm, which most HRIS configurations count as voluntary departures. If your system routes retirements into the voluntary bucket, your internal rate will run above the JOLTS comparison by roughly the size of your retirement-eligible cohort. Map retirements to a third category before benchmarking.
Why do quits and layoffs need to be read together rather than separately?
Because a falling quits rate has two opposite causes and the layoffs series is what distinguishes them. Quits fall when workers are satisfied, and quits fall when there is nowhere to go. In June 2026 the Information sector recorded a quits rate of 1.1% — one of the lowest in the economy — alongside a layoffs rate of 2.0%, which is among the highest. Reading the quits figure alone would have described that sector as unusually stable.
What does a TCI below 50 mean for workforce planning?
It means the sector’s headcount is being set by employer decisions rather than worker decisions, and that replacement hiring assumptions built on backfill logic will overshoot. In Information, Construction, Professional and business services, and Arts, entertainment and recreation, more exits in June 2026 came from layoffs than from resignations. Vacancy models in those sectors should assume positions eliminated rather than positions reopened.
Is a 2.0% national quits rate evidence that retention programs are working?
The engagement data argues against it. Gallup measured global engagement at 20% in 2025, the lowest since 2020, at the same time quits sat at decade lows, and recorded U.S. and Canada job-market optimism down 23 points since 2019. Suppressed voluntary exits under deteriorating engagement describe a constrained market, not a satisfied one.
Why is median tenure falling if people are quitting less?
Tenure and quits move on different clocks. Median tenure reflects the accumulated composition of the workforce — including the very large 2021–2022 cohort of job changers, who reset their tenure clocks and are still working through the early years of new jobs. The 3.9-year January 2024 reading captures that reshuffling; the effect of the current quits freeze will not appear in the tenure series until later supplements.
How much of turnover is seasonal rather than structural?
Enough that unadjusted comparisons mislead. Accommodation and food services carried 638,000 quits in June 2026 seasonally adjusted, and the industry’s not-seasonally-adjusted figures swing substantially by month with summer and holiday staffing. All industry comparisons in this dataset use seasonally adjusted series for the monthly cuts and not-seasonally-adjusted series for the annual cuts, consistent with BLS practice.
Do large employers really lose fewer people voluntarily?
By the establishment-size data, yes, and the gap is wide. Private establishments with 5,000 or more employees posted a 0.7% quits rate in January 2026 against 2.8% at establishments with 10 to 49 employees. Compensation structure, internal mobility, and benefits vesting all plausibly contribute; JOLTS does not decompose the cause, so the figure describes the pattern without explaining it.
Which turnover figure should appear in a board deck?
Two, side by side: your annual separations rate against your sector’s annual quits rate, with voluntary and involuntary split out. A single blended turnover percentage compresses a controllable variable and an uncontrollable one into one number, and boards routinely mistake a restructuring-driven increase for a retention failure — or the reverse.
Will the quits rate recover in 2027?
The mechanism to watch is job openings, not sentiment. Openings averaged 7.1 million in 2025 against 11.2 million in 2022, and the June 2026 level was 7.4 million. Quits historically track available openings closely, so a sustained recovery in postings is the precondition. Nothing in the current series indicates that has begun.
How often is this dataset updated?
On event, not on schedule. Revisions follow material changes in the underlying BLS and Gallup releases rather than a fixed calendar, and each revision is dated on this page.
Related Axis Intelligence Research datasets: Gen Z Statistics 2026 for early-career workforce entry data, AI Statistics 2026 for employment-impact evidence, Agentic AI Statistics for role-transformation data, and Google Statistics 2026 for large-employer headcount disclosure.