Short answer:the government data shows job openings that are historically healthy but a quits rate that's been depressed for about a year — a combination that signals employers have roles open while workers don't feel confident enough to leave a job for one. Layer on top of that an application volume that's roughly quintupled per recruiter since 2022, largely from AI-assisted applying, and you get a labor market where "there are openings" and "it's hard to get hired" are both true at the same time, for different reasons.
What JOLTS actually measures — and doesn't
The Bureau of Labor Statistics' Job Openings and Labor Turnover Survey is the closest thing to an official scoreboard for the US labor market. It tracks four separate numbers every month: job openings (positions actively being recruited for), hires (people who started a new job), quits (people who voluntarily left), and layoffs/discharges (people let go). Each one tells you something different, and reading only one of them is how you end up with a misleadingly simple story.
The definition of "opening" is stricter than it might sound: BLS only counts a position if a specific role with available work exists, the job could start within 30 days, and the employer is actively recruiting outside candidates for it right now — not a role someone is merely thinking about creating, and not one already earmarked for an internal promotion. That standard is meant to exclude exactly the kind of ambiguous posting covered in the ghost-jobs piece linked above. In practice it's self-reported by employers with no independent verification, which is the gap between what the survey is designed to measure and what it can actually confirm.
It also has a real limitation worth naming: JOLTS counts what employers report about their own postings, which means it can't distinguish a genuinely open role from one that technically meets the survey's definition but isn't being actively filled — the same ambiguity behind ghost job postings. The topline numbers are still the best available signal of aggregate hiring conditions; they're just not a promise that every counted opening is one you can actually land.
Context matters for reading any single month's numbers, too. JOLTS openings hit record highs in 2022, during the tightest hiring conditions in decades, and have spent the time since drifting back down toward levels closer to 2018–2019 — still a normal, functioning labor market by historical standards, just no longer the extreme seller's market of the immediate post-pandemic period. A lot of the "the market is collapsing" commentary you'll see is really describing that normalization from an unusually hot 2022, not a labor market in actual decline. Both things can be true: hiring conditions are less frantic than 2022, and they're still within a historically ordinary range.
The June 2026 numbers
| Metric | June 2026 | Rate | What it means |
|---|---|---|---|
| Job openings | 7.4 million | 4.4% | Positions actively being recruited for |
| Hires | 5.3 million | 3.4% | People who started a new job that month |
| Total separations | 5.4 million | 3.4% | Everyone who left a job, voluntary or not |
| Quits | 3.2 million | 2.0% | Left voluntarily — the confidence signal |
| Layoffs / discharges | 1.8 million | — | Involuntary separations |
Source: U.S. Bureau of Labor Statistics, JOLTS, June 2026 release. Openings outnumber hires by more than 2 million — not unusual on its own, since not every open posting converts to a hire in the same month, but worth holding onto as context for the sections below.
The quits rate is the tell
Of the four numbers, quits is the one economists lean on most as a confidence signal, because it's revealed behavior rather than a survey opinion — nobody quits a job by accident. The quits rate has sat at or below 2% for roughly a year now, well below both the pre-pandemic norm and the roughly 3% peak reached during the 2022 "Great Resignation." A workforce that isn't quitting at its historical rate is a workforce that, in aggregate, doesn't feel confident it can walk into something better — even while openings sit at a level that looks healthy on paper.
That's the honest tension in the 2026 data: openings numbers alone would suggest a strong hiring environment. The quits rate alone would suggest a cautious, low-confidence one. Both are real, measured the same month, from the same survey. Neither cancels the other out.
This matters practically, not just academically, if you're currently employed and weighing whether to leave for something else. A depressed quits rate doesn't mean leaving is a bad idea for you specifically — it means the aggregate workforce is being more cautious than usual, which is exactly the condition under which being one of the people who does move deliberately, with a real target role and a real reason, tends to face less competition from opportunistic job-hoppers than it would in a high-quits environment. A cautious labor market is a worse environment for casual browsing and a comparatively better one for a targeted, well-prepared search.
Posting volume vs. actual hiring — where the gap widened
The number that's moved the most isn't openings or hires — it's applications. Greenhouse's March 2026 Recruiting Benchmark Report found applications per recruiter rose from 146 in 2022 to 746 in 2025 — roughly a fivefold increase in four years, even as the number of jobs closed with a hire rose only about 8% over the same period. Recruiting teams didn't get five times bigger; they got squeezed between five times the volume and roughly the same output, which is part of why response times and silence have gotten worse for applicants, not better, even as headline openings numbers looked fine.
A meaningful share of that volume increase is AI-driven. LinkedIn was reportedly processing around 11,000 applications a minute in 2025 — a 45% year-over-year jump — much of it attributed to generative-AI tools that let candidates tailor and fire off applications far faster than typing one out by hand. That's a real capability, not inherently a problem. But when it scales across millions of job seekers simultaneously, the practical effect on any single posting is the same: more applications competing for the same headcount, with no corresponding increase in how many of them get read closely.
It's worth being precise about what this does and doesn't explain. It doesn't mean every posting is drowning equally — a specialized senior role and a high-volume entry-level posting see very different application floods. And it doesn't mean your odds are fixed; most of what determines whether your application gets seen is still about matching the specific posting well, not just beating a raw applicant count. But it does mean "I applied and heard nothing" is, in 2026, at least partly a volume story and not purely a reflection of your resume.
It also explains a lot of what job seekers experience as a change in how hiring feels, even where the underlying process hasn't formally changed. A recruiting team facing five times the applications with roughly the same headcount doesn't have five times the hours to read cover letters or open every attachment — the predictable response is leaning harder on automated screening at the first pass, which is exactly why how a resume actually parses matters more at this volume than it did when a recruiter could realistically eyeball every application by hand. None of the underlying JOLTS or Greenhouse data tells you a specific company changed its process; it does tell you the incentive to lean on automated filtering got stronger everywhere, all at once.
Industries aren't experiencing the same market
National aggregates flatten real variation. The same pattern shows up in adjacent data: when Greenhouse broke down its own ghost-job rates by industry, construction, arts, food service, and legal came out highest — sectors with high turnover, seasonal cycles, or heavy staffing-agency use, where posting volume and actual headcount planning are more loosely coupled to begin with. The inverse tends to hold in more stable, headcount-planned sectors like enterprise software or established professional services, where a posting is more likely to map to a single, specific, currently-open role. Neither the JOLTS topline nor the Greenhouse benchmark data breaks quits or application volume out by industry in a way that's directly comparable here — the honest statement is that the aggregate national picture is a real average of genuinely different local conditions, not a uniform description of "the job market" you personally are in.
Remote work: stabilized, not reversing
Despite a steady drumbeat of return-to-office headlines, Gallup's tracking shows the remote/hybrid/on-site split among remote-capable US employees has been broadly stable since 2022. As of Q2 2025, hybrid work sat at 51% of remote-capable employees, down modestly from 55% two quarters earlier, with fully-remote and fully-on-site each up roughly two points over the same window — a shuffle within a stable range, not a rush back to five-day office weeks. Hybrid workers are spending about 46% of their workweek in-office on average (roughly 2.3 days), up from 42% in 2022, but that increase plateaued after 2023 and hasn't continued climbing since.
The practical read for a job search: remote and hybrid roles remain a real, durable share of the market rather than a shrinking one, even though any individual company's policy can and does change. High-profile return-to-office mandates get outsized media coverage relative to their actual weight in the aggregate numbers — a handful of large, visible employers reversing policy makes headlines in a way that a much larger number of companies quietly keeping their existing hybrid or remote setup does not. That asymmetry in coverage is worth remembering the next time a single company's RTO announcement gets framed as evidence of an industry-wide reversal; Gallup's aggregate data, drawn from a broad employee sample rather than a handful of headline-grabbing companies, is the more reliable signal of where the overall market actually sits.
What the data supports vs. what commentary claims
A lot of what circulates about "the 2026 job market" is directional commentary dressed up as data. Worth separating clearly:
- Supported by data: openings remain historically substantial; the quits rate is genuinely depressed versus pre-pandemic and 2022 levels; applications per recruiter have risen sharply; remote/hybrid work has stabilized rather than reversed.
- Not something JOLTS or Gallup can tell you:whether any specific individual posting is real (see the ghost-job caveat above), what your personal odds are in your specific field or level, or whether "the market" will look the same next quarter. National aggregates describe the whole economy, not your search.
- Genuinely uncertain, even to economists: how much of the application-volume surge is a temporary adjustment period as both job seekers and employers adapt their tools and processes to AI-assisted applying, versus a permanent new baseline. Nobody has a reliable answer to that yet, and treat anyone who states one with high confidence — in either direction — with the same skepticism this article applied to the ghost-job survey numbers earlier.
What this means for your search, practically
The honest takeaway isn't "the market is great" or "the market is terrible" — it's that volume, not scarcity of openings, is the dominant force right now. That changes where effort is best spent: less value in applying to more postings faster (that's the strategy an AI-flooded applicant pool is already doing at scale), more value in applying well to postings you're genuinely matched to.
Concretely, that means three shifts worth making if you've been applying the way the job search advice from 2019 or even 2022 suggested. First, treat a real posted date and a specific, current role as a filter before you invest time — the ghost-job signals covered separately are more relevant now than when applicant pools were smaller and recruiters had more slack to respond to marginal applications. Second, assume your resume is being read by software before a person, more often than it was a few years ago, simply because the volume math above makes manual first-pass review less feasible at scale — formatting and parsing correctness matter more as a result, not less. Third, weight referrals and direct outreach a little higher relative to cold applications than you might have a few years ago; when a recruiter is choosing among 746 applications instead of 146, a warm signal that shortcuts the pile carries more relative weight than it used to.
None of that requires guessing which of these forces is dominant in your specific case. On CareerFlint's job search, every listing is ranked against your actual resume rather than a flat keyword match, which is the practical way to spend time on the smaller set of postings worth a tailored application instead of adding volume to an already saturated pool.