Last month, when Kamala Harris still had some chance of winning the election, we were not surprised to learn that according to the extremely political Bureau of Labor Statistics, in August the number of job openings unexpectedly soared from an upward revised 7.7 million to 8.040 million, which was not only a 3-sigma beat to expectations, but was also above the highest Wall Street forecast. Fast forward to today, when Kamala’s chance of winning are effectively zero – as even the suddenly apolitical Jeff Bezos now admits – and shockingly moments ago the BLS reported that in September, the number of job openings  plunged from over 8 million to just 7.4 million, the lowest since early 2021…

… a staggering 6 sigma miss to expectations, and a number which came below the lowest estimate; in fact as shown below, in just one month we have gone from a staggering beat to one of the biggest misses on record!

There’s more: not only was the drop in job openings a whopping 418K, it would have been far worse had the BLS not revised the August print sharply lower, from 8.04 million to 7.861 million. And for those who keep track of the infinite downward revisions at the Biden Department of Goalseeked Government Data, where every strong data point is quietly revised down to flat or negative in subsequent months, this is what the JOLTS revisions look like now.

According to the BLS, the number of job openings decreased in health care and social assistance (-178,000); state and local government, excluding education (-79,000); and federal government (-28,000) but increased in finance and insurance (+85,000)

Amusingly, after we mocked last month’s stunning surge in construction job openings just as a record chasm had opened between the manipulated number of construction jobs and openings…

… which meant the biggest monthly surge in construction job openings on record at a time when the housing market has effectively frozen thanks to sky high interest rates, a simply glorious paradox of manipulated bullshit data…

… the BLS realized that it had to make an adjustment after getting called out, and Construction Job openings tumbled by 40K to 288K and once again rapidly approaching the post-covid lows. Oh, and yes, the number of “construction jobs” is about to fall off a cliff just as soon as Orange Man Bad enters the White House.

Setting the glaring data manipulation aside, in the context of the broader jobs report, in September the number of job openings was 690K more than the number of unemployed workers, slightly down from the previous month and not too far from inverting once again, similar to what happened during the covid crash.

And while the job openings collapse was a shocking return to the contraction observed for much of 2024, the one silver lining was in the number of hires which rebounded by 123K to 5.558 million, just above the post-covid lows.

That, however, was not matched by the other closely followed category, the number of quits, which in September tumbled to 3.071 million, not only a new post-covid low and down by 107K from August, but a number first recorded back in October 2016, which means that true state of the job market is about as “strong” as it was 8 years ago, when US GDP was about $10 trillion lower. Quits decreased in professional and business services (-94,000) but increased in state and local government, excluding education (+22,000) and in real estate and rental and leasing (+18,000).

Finally, no matter what the “data” shows, let’s not forget that it is all just estimated, and it is safe to say that the real number of job openings remains still far lower since half of it – or some 70% to be specific – is guesswork. As the BLS itself admits, while the response rate to most of its various labor (and other) surveys has collapsed in recent years, nothing is as bad as the JOLTS report where the actual response rate remains near a record low 33%

In other words, more than two thirds, or 70% of the final number of job openings, is made up!