The Job Openings and Labor Turnover Survey, known as JOLTS, measures how many jobs employers are trying to fill and how many workers are starting and leaving jobs. This article explains how the Bureau of Labor Statistics (BLS) produces it, what each term means, and what the latest release says.
What is JOLTS and why does it exist?
Most labor-market headlines focus on payroll employment and the unemployment rate. Those figures describe the stock of jobs and workers at a point in time. JOLTS covers the movement underneath: the flow of people into and out of jobs, plus the demand for workers that has not yet turned into hiring. Our guide to how to read the monthly jobs report covers the stock measures; JOLTS is a companion to them.
The survey is run by the BLS. According to the BLS, it draws on a sample of roughly 21,000 U.S. business and government establishments and covers all nonagricultural industries in the public and private sectors, across the 50 states and the District of Columbia. Data are collected through BLS data collection centers in Atlanta and Kansas City.
How does the survey define an “opening”?
The definitions matter because everyday usage of these words is looser than the statistical one. According to the BLS technical note, a job opening is counted only when all of the following hold:
- A specific position exists and there is work available for it.
- The job could start within 30 days.
- The employer is actively recruiting from outside the establishment.
Openings are measured on the last business day of the month. An open position that is only being considered, or one that will be filled by an internal transfer, does not count. A listing that stays online while the employer has stopped recruiting should not count either, although the survey relies on what employers report.
What do hires, quits and layoffs mean?
Hires and separations (departures from payrolls) are measured over the entire calendar month, not on a single day. The BLS defines them as follows:
| Term | BLS definition in brief |
|---|---|
| Hires | All additions to the payroll during the month, including new and rehired employees and transfers from other locations |
| Quits | Voluntary departures, excluding retirements and transfers to other locations |
| Layoffs and discharges | Employer-initiated separations, including layoffs that last or are expected to last more than 7 days |
| Other separations | Retirements, transfers to other locations, separations due to disability, and deaths |
| Total separations | Quits, layoffs and discharges, and other separations combined |
Quits are often read as a sign of worker confidence, since people tend to leave voluntarily when they expect to find something else. That is an interpretation by analysts, not something the survey measures directly. A quit may also reflect a move abroad, a family decision or a health matter.
How are the rates calculated?
Alongside counts, the BLS publishes rates that make months and industries comparable. According to the BLS, the job openings rate is openings divided by the sum of employment and openings, times 100. The hires, separations, quits and layoffs rates are each the count divided by employment, times 100.
Here is an illustrative worked example using round numbers, not actual data. Suppose employment is 150 million and there are 7 million openings:
- Openings rate: 7 ÷ (150 + 7) × 100 = about 4.5 percent.
- If 5 million people are hired that month: 5 ÷ 150 × 100 = about 3.3 percent.
The openings rate uses a different denominator because it asks what share of all positions, filled or unfilled, are vacant. The other rates ask what share of current employees were hired, quit or were let go.
What do the September 2026 figures show?
The BLS published the September 2026 release on November 3, 2026. According to that release:
| Measure | September 2026 level | Rate |
|---|---|---|
| Job openings | 7.5 million | 4.5% |
| Hires | 5.4 million | 3.4% |
| Quits | 3.4 million | 2.1% |
| Layoffs and discharges | 1.6 million | 1.0% |
| Total separations | 5.3 million | 3.3% |
The BLS described openings, hires, layoffs and total separations as little changed. Quits rose in number, and the BLS said they were up 375,000 over the year. Other separations were 313,000.
The same release revised August upward for several series: openings to 7.2 million, hires to 5.3 million and total separations to 5.2 million. August quits were revised down to 3.0 million, and layoffs and discharges were revised up to 1.8 million. Revisions like these are routine and a reason to be careful about reading a single month as a turning point.
Can hires and separations be compared directly?
Yes, and the comparison is informative. In September 2026, hires of 5.4 million exceeded total separations of 5.3 million by about 100,000 (5.4 minus 5.3). Because both figures are rounded, the exact gap could differ somewhat. Employment growth in the survey is roughly the difference between these two flows, though payroll counts from the separate Current Employment Statistics survey remain the standard measure of job gains.
Openings can also be set against hires. Openings of 7.5 million versus hires of 5.4 million shows that many positions were open at month end while a smaller number were filled during the month. The two are measured on different timing bases (one day versus a full month), so the gap is not a precise count of unfilled jobs. It is a rough indicator of how much demand for workers exists relative to how many people actually start jobs.
How accurate is JOLTS, and why does it get revised?
JOLTS is a sample survey, so every figure carries sampling error. The BLS technical note says estimates are adjusted for unit nonresponse (employers that do not answer at all) and item nonresponse (employers that skip particular questions). The BLS methodology does not state a response rate, so none is given here.
Two kinds of benchmarking keep the survey aligned with other BLS employment data. According to the BLS, each month the JOLTS estimates are benchmarked so that weighted JOLTS employment equals employment from the Current Employment Statistics survey. Once a year, the estimates are re-benchmarked to CES employment that has itself been benchmarked to the Quarterly Census of Employment and Wages, a near-complete count of employment. The BLS says estimates are revised annually with the January data release, and that five years of data are subject to revision, with seasonally adjusted figures for the most recent five years recalculated.
Seasonal adjustment is a statistical method that removes regular calendar patterns, such as retail hiring before the holidays, so that months can be compared. The BLS publishes both seasonally adjusted and unadjusted figures, and the headline numbers are usually the adjusted ones.
Why do economists and the Fed watch it?
The Federal Reserve describes its goals as maximum employment and stable prices. Labor-market data are one input into its interest rate decisions, along with inflation readings such as those described in our explainers on what the Consumer Price Index measures and how the Federal Reserve sets interest rates. Because JOLTS shows both demand for workers and turnover, analysts use it to judge whether the labor market is tight (employers struggling to find workers) or loosening. Interpretations differ, and no single JOLTS figure settles the question.
What JOLTS can and cannot tell a reader
JOLTS is useful for three things: gauging employer demand through openings, seeing how much churn there is through hires and separations, and spotting differences by industry. It is less suited to predicting the next jobs report, to describing the quality or pay of open jobs, or to showing who is filling them. It does not report wages or worker characteristics in the headline release.
Read together with payroll, unemployment and inflation data, and with revisions in mind, the survey adds a useful layer to the picture of the labor market without being a verdict on it. Trends across several months say more than any one release.
This article is general information, not personalized financial, investment, tax or legal advice.
Sources: Bureau of Labor Statistics
