White collar workers show why US jobs openings data is riddled with red flags
If you look at job openings, you’d think professional and business services are doing much better than the rest of the job market. They aren’t.
The US job openings and labor turnover survey showed an unexpectedly large jump in postings for August, up to over 8 million.
Openings in government, construction, and trade, transportation, and utilities sectors drove this increase, but there was another surprising sector that moved up meaningfully as well: professional and business services.
Stepping back, job openings in professional and business services are virtually flat year-on-year (down 30,000, or -2%), while total job openings are down a whopping 14%.
That must mean demand for labor is stronger in professional and business services than the economy as a whole, right?
This sector amounts to roughly 15% of total employment, but has accounted for just 5% of net job growth over the past year. Payroll growth in this sector is well below-average.
Ah. Well. Perhaps this is a case of a sector-specific labor shortage, and employers simply being unable to find qualified people to fill those positions. But if that were happening, we’d expect better pay growth in this industry to entice workers to stay put rather than head for greener pastures. And that’s not playing out either, judging by the Employment Cost Index’s wage data.
So this is an instance of the internals of the job openings data being incongruent with most other metrics we have on the state of the labor market. And if it’s job openings against the world, I’ll take the world. Couple that with the overall very low response rate for this survey (in the low-30s% since mid-2022) and it’s yet another example of the pitfalls that await those who put job openings front-and-center in their jobs market analyses.
I have not been a fan of the Federal Reserve’s use of job openings – or the ratio of job openings to unemployed Americans – as a good catch-all metric for labor market conditions over the past few years. A few more reasons:
The ratio of job openings to unemployment has clear cyclical elements – going up when the economy is good and down when it is less good – but it has also trended higher over time. This is telling me there is something about the nature of job openings that evolved over time (i.e., it is easier to do so).
A lot of net monthly job growth comes from people who weren’t even in the labor force and looking for a job a month ago. This means the available pool of labor is always larger than what the headline number of unemployed would imply.
The ratio of vacancies to unemployed tends to track the private sector quits rate over time, and quitting is a real action. There’s the phrase about the classic bacon-and-eggs breakfast: the chicken was involved, the pig was committed. Given the difference in power dynamics, a worker quitting a job sends a much stronger signal about labor market conditions than a company posting a job opening.
Job openings are a nearly costless call option for employers to see if The Perfect Candidate is out there. You have the ability to find a great hire, but no obligation to react to resumes that come in. This feeling has been reinforced by my work experience, where I’ve seen job postings linger for no apparent reason, long after the role had been filled.
[And an aside to the analysts who have suggested “just de-trend JOLTS to normalize for the upward drift over time” – now may be a bit of a rubber meets the road time for that thesis, since there’s a nascent disconnect between job openings (moving sideways-ish) and the private sector quits rate (down to 2015 levels).]
