Job openings in June 2026 were about 7.36 million, slightly above the 2019 average. But there are more unemployed people competing for them, so openings per unemployed person fell to 1.04 from 1.21 in June 2019. Hires are down about 10 per cent against 2019, quits down 11 per cent, and the median unemployment spell has stretched from 9.2 weeks to 10.5. Layoffs are the only series below its 2019 level in a way that helps you.
The headline everyone quotes
Job openings ran about 7.36 million in June 2026 against a 2019 average of 7.15 million. On that number alone the market is back to normal, and that is the figure most reporting reaches for.
It is also the single most flattering number available, because it counts vacancies without counting the people chasing them.
The number that corrects it
In June 2019 there were about 7.19 million openings against 5.94 million unemployed people — 1.21 openings each. In June 2026 there were 7.36 million openings against 7.09 million unemployed — 1.04 each. At the March 2022 peak the figure was 2.04.
So on the measure that actually describes competition, conditions are worse than 2019 despite a higher openings count. More vacancies, and more than a million additional people looking at them.
Employers are advertising more than they are hiring
Hires ran about 5.35 million in June 2026 against a 2019 average of 5.83 million — down roughly 10 per cent. Expressed against openings, employers completed about 0.73 hires per opening compared with 0.82 in 2019.
That gap is what a job seeker actually experiences: processes that start and do not finish, roles that stay posted, interviews that lead to a pause rather than an offer. The openings figure cannot show it and the hires figure can.
And almost nobody is moving
Quits are down about 11 per cent on 2019 and layoffs down about 4 per cent. Low firing combined with low quitting is a frozen market rather than a collapsing one.
It is also self-sustaining, which is the part worth understanding. Most openings are created by somebody leaving. When people stop leaving, vacancies stop appearing, which gives the people who remain even less reason to move. The market does not correct itself; it has to be thawed by something external.
What it costs in time
The median unemployment spell was 10.5 weeks in July 2026 against a 2019 average of 9.2. Longer, and nowhere near the 25.2-week peak of June 2010 — which is the right scale for judging how bad this actually is.
A fortnight of additional median search time sounds mild and is not, because the distribution has a long tail. The median moving from 9.2 to 10.5 usually means the tail got considerably heavier, and it is the tail that ruins finances.
Why 2021 and 2022 are the wrong baseline
Openings peaked at 12.3 million in March 2022 and quits at about 4.5 million a month in April 2022. Both are the highest readings in the series by a distance.
Anyone who changed jobs in that window experienced multiple offers, fast processes and employers competing openly. Comparing today against that produces a story of collapse. Comparing against 2019 produces a story of a normal market with unusually little movement inside it, and the second is the more useful description.
What “tough” means specifically right now
Not mass redundancy — layoffs are below their pre-pandemic norm and that is genuinely good news. It means more candidates per opening, slower processes, fewer speculative postings converting into hires, and fewer people vacating roles.
Those are the conditions of a market where the people who have jobs keep them and the people who want one wait. Which is difficult, and is a different difficulty from the one most coverage describes.
How to read the next scary headline
Find the baseline. If the comparison is to 2021 or 2022, it is describing a return to normal as a catastrophe. If the comparison is to 2019 and the number is genuinely worse, that is a different and more serious claim worth taking seriously.
Then check whether the figure quoted is openings or hires. An article about “millions of jobs available” built on the openings count is describing the one series that currently flatters the picture.
Common questions
Are there fewer jobs than in 2019?
Openings are slightly higher. But openings per unemployed person fell to 1.04 from 1.21, so competition for each one is greater.
Are employers actually hiring?
Less than they advertise. Hires were about 5.35 million in June 2026 against a 2019 average near 5.83 million, down roughly 10 per cent.
How long is a search taking?
The median unemployment spell was 10.5 weeks in July 2026, against 9.2 in 2019 and a peak of 25.2 weeks in June 2010.
Are layoffs rising?
No. At about 1.77 million a month they sit below the 2019 average, which is the one series clearly in a job seeker's favor.
So is the market bad or not?
Neither a collapse nor normal. Similar advertising, fewer completed hires, more competition and longer searches.
Is the job market actually bad right now?
Harder than 2019 despite a similar openings count. Openings per unemployed person fell to 1.04 from 1.21, hires are down about 10 per cent and searches run longer.
Why does a frozen market persist?
Most openings are created by somebody leaving. When people stop leaving, vacancies stop appearing, which gives everyone else less reason to move.
How long is a search taking?
The median unemployment spell was 10.5 weeks in July 2026 against 9.2 in 2019, and far below the 25.2-week peak of June 2010.