Why is hiring weaker than quits?

Untangling a JOLTS puzzle

TL;DR: This piece explores an interesting labor market discrepancy: quits and layoffs suggest a stronger US labor market than hires. Two possible explanations are labor hoarding and induced quitting.

1. The Divergence Between Hires and Quits

For the last few years, I’ve been regularly sharing a pair of charts that benchmark hiring and quits to the unemployment rate.¹

During the super-hot “Great Resignation” phase of 2021 through early 2022, I was mostly focused on the fact that turnover (hires, quits) was running running way, way higher than you would have expected given the unemployment rate. That gap got progressively smaller as the labor market cooled, and in early 2023 hires “normalized”.

Quits were still slightly elevated at that point:

And as the labor market cooled further over the past 6-7 months, hires fell below what the unemployment rate would suggest, whereas quits fell to normal levels. Another way of saying this is that quits are behaving as if the unemployment rate is around 3.8%, but hires are behaving as if the unemployment rate is a little north 4.5%.

That prompted me to look at layoffs with the same framework, and lo and behold, they’re running lower than you’d expect.

If you add quits and layoffs, you end up with something that looks nearly identical to hires in terms of deviations from the 2010s relationship: a little lower than you would expect given the recent unemployment rate, and more consistent with an unemployment rate slightly north of 4.5%.

So to sum up, hires and the sum of layoffs and quits are behaving as if the labor market is a little cooler than implied by the current unemployment rate. But quits are behaving as if it’s normal, and layoffs as if it’s hotter.

For what it’s worth, this observation - that while quits are normal, quits plus separations are lower than normal - may be a partial explanation for the sentiment expressed in a recent article by Chip Cutter for the Wall Street Journal:

2. Possible Explanations

I don’t have a rock-solid explanation for why this is going on, but I do think there are two sets of answers: behavioral and taxonomic. Two in particular worth considering are labor hoarding and induced quitting.

The behavioral explanation that seems consistent with this constellation of data is labor hoarding - traumatized by labor shortages they experienced during the Great Resignation, employers want to hold onto their existing workers during a temporary(?) soft patch(??) in the economy. So they’re keeping layoffs very low. However, since they aren’t laying folks off, and don’t want to grow their workforces too fast either, employers are also keeping hiring low. Hiring and layoffs both landing lower than normal means quits can be normal - i.e. jobseekers still see a reasonable range of opportunities out there given the current state of the labor market.

A taxonomic explanation is that “layoffs” and “quits”, as reported by businesses, aren’t necessarily consistent concepts over time. As Matt Darling noted a few years ago, employers have a bunch of levers they can use to “encourage” quitting. Since I mentioned Chip Cutter’s latest article, let it be noted that he’s written a bunch of fascinating articles about this kind of “induced quitting” (#1, #2, there are more).

I don’t have a lot of confidence on which of these phenomena is a more important driver, and I don’t know how you would empirically distinguish between them. There are also explanations I have surely forgotten - if you can think of any, please let me know!

1. Nothing econometrically fancy: just a quadratic relationship between the unemployment rate and the hires/quits rates, based on data from the 2010s expansion.