You have been applying for weeks. Nothing bounces back, nothing rejects you, nothing moves.
The August 4 JOLTS release explains the shape of that silence: the ratio of job openings to unemployed workers has been stuck at exactly 1.0 for four straight months, and the next JOLTS drops September 1.
What the JOLTS June 2026 release actually said
The Bureau of Labor Statistics reported 7.4 million job openings in June 2026, 5.3 million hires, and 5.4 million total separations, with the openings-to-unemployed ratio flat at 1.0 for the fourth consecutive month. That is the tightest balance reading in years, and it is the single best explanation for why your inbox is quiet.
A quick decoder for the numbers behind the headline:
- Openings: 7.4M, little changed from May.
- Hires: 5.3M, flat.
- Quits: 3.2M, flat. Nobody is leaving voluntarily.
- Layoffs and discharges: 1.8M, flat. Nobody is being pushed out either.
- Openings-to-unemployed ratio: 1.0, fourth month in a row.
The shorthand economists use for this pattern is "low-hire, low-fire." It looks stable on a dashboard. It feels like drowning if you are the one applying.
Fourth straight month at parity, per BLS JOLTS released August 4, 2026.
Why a 1.0 ratio is the worst posture for a job seeker
A ratio of exactly 1.0 is not balance from the candidate's side. It is the posture where employers can be maximally picky and seekers cannot blame a recession, because on paper there is a job for every jobless person.
Compare the three regimes:
| Openings per unemployed | Employer behavior | Seeker experience |
|---|---|---|
| 2.0 (2022 peak) | Stretch on fit, train up | Fast yeses, counter-offers |
| 1.0 (June 2026) | Hold out for exact match | Silent inbox, long searches |
| 0.5 (recession) | Freeze hiring entirely | Rejections, but clear signal |
At 2.0, a hiring manager with two roles and one qualified applicant will make the offer and figure out the rest. At 1.0, that same manager has one role and one qualified applicant, plus 241 others clamoring for the slot. There is no pressure to compromise. The market clears against generalists.
This is the mechanism behind the split you are seeing. Headline unemployment has hovered in the 4.1% to 4.5% band since May 2025, so the total pain is not obvious, but the duration of pain has exploded. The share of unemployed workers jobless for 27 weeks or more went from 17.8% in May 2024, to 20.4% in May 2025, to 27.5% in May 2026. That is a 9.7 point jump in two years while the headline barely moved.
Long-term unemployment at 27 weeks and the duration problem
Long-term unemployment, defined by BLS as being jobless for 27 weeks or longer, covers 25.5% of all unemployed people as of the July 2026 employment situation report, or roughly 1.8 million workers. In a 1.0 market, the seekers who match get matched quickly, and the ones who do not accumulate.
Three things are happening at once:
- Filter thresholds moved up. ATS keyword floors that used to accept "close enough" now demand exact matches, because volume forces machine filtering.
- Interview rounds stretched. Employers are conducting 42% more interviews per hire than they did in 2021.
- Entry-level evaporated. Entry-level postings have dropped by 29 percentage points since January 2024, so early-career workers stack up in the long-term bucket first.
If you have crossed the 27 week mark, the honest read is that generic outreach is not going to close it. What closes it is a resume that matches one specific posting at a time, which is exactly the work Refolk takes off you: paste the posting, get your own resume back rewritten to it, with a fit score that tells you whether it is worth the send.
Why am I not getting interviews in 2026
Because the applicant-to-interview conversion rate has collapsed roughly 5x in a decade, from 15.25% in 2016 to 8.4% in 2023 to about 3% in 2024, and 2026 is tracking in the 2 to 3% band. If you send 100 cold applications today, statistical expectation is two or three interviews, and that is before accounting for ghost jobs.
The mechanism is not mysterious. AI-assisted apply tools, LinkedIn Easy Apply, and one-click extensions inflated the volume of applications without inflating the fit of applications. The average job opening now receives 242 applications, which translates to a 0.4% success rate for any individual applicant on any individual req. Recruiters responded by raising ATS thresholds and leaning harder on machine screening. The net effect: generic resumes now fail the filter stage, not the evaluator stage. A human never reads them.
Two other numbers make this concrete:
- Between 18% and 30% of online postings in 2025 were ghost jobs, meaning the 7.4M openings figure overstates real hiring intent.
- Hires are running at about 72% of openings (5.3M / 7.4M), so even the real jobs are converting slowly.
At a 1.0 ratio, a generalist resume is not underqualified. It is unreadable to the filter.
What Refolk's index shows about the candidate glut
Refolk's index of professional profiles shows a nearly 2-to-1 ratio of senior software engineers to recruiters in the U.S., and a floor of 458 people who have literally set their current job title to "Open to Work" or "Looking for Work." Both numbers explain, structurally, why nothing is bouncing back.
Here is the comparable dataset in one place:
| Metric | Value | Source |
|---|---|---|
| U.S. job openings, June 2026 | 7.4M | BLS JOLTS |
| U.S. hires, June 2026 | 5.3M | BLS JOLTS |
| Hire rate as % of openings | ~72% | Derived |
| Long-term unemployed share, May 2024 | 17.8% | Econofact |
| Long-term unemployed share, May 2026 | 27.5% | Econofact |
| Senior Software Engineers, U.S. | 180,724 | Refolk's index |
| Recruiters + Talent Acquisition, U.S. | 91,801 | Refolk's index |
| SWE-to-recruiter ratio | 1.97:1 | Refolk's index |
| Applications per opening | 242 avg | Interview Guys |
| Applicant-to-interview conversion, 2016 | 15.25% | HiringThing |
| Applicant-to-interview conversion, 2024 | 3% | HiringThing |
Do the math on the tech numbers. If every recruiter in the U.S. reviewed only senior software engineers and did nothing else, each would carry two senior candidates plus every junior, mid, and staff+ applicant on top. That is impossible without machine filtering, which means the filter is what you are applying to, not a human.
Refolk's index. A floor estimate of workers who have stopped hiding the search.
The 458 figure is small in absolute terms, but the signal matters. These are workers who have given up on the polite fiction that they are still employed. Top self-declared "employers" in the index include "#opentowork" and "Looking for Job." When a market plateau produces enough duration fatigue that hundreds of people rewrite their identity around the search, you are looking at a structural dynamic, not a temporary lull.
What to change about resumes and targeting at 1.0
Three moves matter at a 1.0 ratio: cut volume, raise fit, and buy referrals with the hours you save. The old playbook of 50 applications a day is actively hurting your conversion, because the marginal application is going into a filter you have not tuned.
Cut volume, raise fit
At 3% interview conversion and 242 applications per opening, sending more generic resumes does not scale. Sending fewer, tuned resumes does. The concrete move:
- Pick 10 to 15 postings a week where you actually match 70%+ of the requirements.
- Rewrite the resume against each posting's language, not a master version.
- Send a matching cover letter that references two specific lines from the JD.
- Track which postings actually get read.
The rewriting is the expensive part in hours, and it is why most people default to blasting. Refolk does that rewrite per posting and drafts the matching cover letter in the same pass, so the constraint stops being your evenings.
Buy referrals with the saved hours
Referred candidates are hired at 30% versus 7% from other sources per Jobvite 2024, a 4.3-to-1 ratio. In a 1.0 market, one warm intro is worth roughly the entire cold funnel of a typical job seeker. Given 242 applications per posting, cold apply is a lottery ticket, and the referral path is the actual product. Spend the hours you saved on tuned resumes reaching two people at every target company, not on application 51.
Check for ghost jobs before you invest
If 18% to 30% of postings are not real, you need a filter before the ATS filter. Cheap signals:
- Reposted with the same date every 30 days? Ghost.
- No recruiter on LinkedIn tagged to the req? Likely ghost.
- Been up 60+ days at a company that just did layoffs? Ghost.
- Requires a login-walled portal with a 45-minute application? Assume ghost until proven otherwise.
What changes on September 1
The July 2026 JOLTS release drops September 1, and the number to watch is whether the ratio breaks off 1.0 in either direction. Everything else is noise.
Three scenarios and what each means for your search:
- Ratio ticks up to 1.1 or higher. Employers get slightly more pressure, filters relax marginally. Push volume of tuned applications.
- Ratio stays at 1.0 for a fifth month. The plateau is structural, not seasonal. Double down on referrals and internal moves.
- Ratio drops below 1.0. Layoffs are re-accelerating. Prioritize any offer in hand, and expect the long-term unemployment share to climb past 28%.
Until then, the operating assumption is that the market is matched, employers have zero slack, and the resume you send has to look like it was written for the posting, because it was.
FAQ
What is the job openings to unemployed ratio and why does 1.0 matter?
The ratio is BLS's monthly count of open jobs divided by the number of unemployed workers. Above 1.0 means more openings than seekers, which pressures employers to compromise on fit. Below 1.0 means more seekers than openings, which produces obvious layoff cycles. Exactly 1.0, where the market has sat for four months, is the posture where employers can hold out for exact matches without paying a wage premium, and it is why generic applications now vanish into filters.
Why is long-term unemployment climbing if headline unemployment is flat?
Because duration and headcount move independently. The unemployment rate has stayed in a 4.1% to 4.5% band since May 2025, but the share of unemployed workers jobless for 27 weeks or more went from 17.8% in May 2024 to 27.5% in May 2026. At a 1.0 ratio, workers who match get matched fast, and workers who do not accumulate in the long-term bucket. The average is stable, the distribution is not.
How many applications should I send per week in this market?
Fewer than you think, tuned harder than you think. The applicant-to-interview conversion rate is roughly 3% in 2024 to 2026 data, down from 15.25% in 2016, so volume without fit produces a rounding error. A realistic weekly target is 10 to 15 postings where you match 70%+ of the requirements, each with a resume rewritten to the posting and a cover letter that quotes two lines from the JD, plus two referral asks per target company.
When does the next JOLTS release come out?
The July 2026 JOLTS release is scheduled for September 1, 2026. The number to watch is whether the openings-to-unemployed ratio moves off 1.0. A move up loosens filters; a move down accelerates layoffs; a fifth month at 1.0 confirms that the plateau is structural and referrals become the dominant strategy variable rather than an optimization.