On Job Openings and Job Availability

Paul Krugman is a tireless writer. That’s the good part. The bad part (you knew this was coming) is that…well, he can also be a tiresome writer.

Consider this: A Rising Natural Rate. Here, he is commenting on some analysis by the SF Fed trying to estimate some measure of the “natural” rate of unemployment. Fine, nothing wrong with this. But then he slips this in:

Right now, there are very few job openings relative to the number of unemployed:

So there’s no question that right now, the demand side is what is constraining unemployment.

Ya got that? Paulo says that thar’s no question bout it. Thar’s deficient demand out in them thar hills. An y’all see that l’il ol’ dyergram up thar? Well…that thar jus’ goes ta prove it. Lessen’ yer blind, that is. Lessen’ yer sum evil laysay fare type.

Well, I hate to break it to those who demand and consume this brand of religion, but there might just be some question about it. Shhhh…what I am about to say is super secret…economists aren’t really sure what’s going on. I mean, think about it. If we knew what was going on, there would be no need for economic research. You know…research…that activity that brings so much joy to you know who (The Joy of Research).

But I don’t want to be too hard on Paulo. Evidently, he has an agenda to push and he pushes it from a particular philosophical perspective. I can respect that. What I don’t like is the constant allusion to certainty–the lack of humility in what we know–the notion that the data “speaks for itself.” These are the tactics used by politicians, not academics. This is what I find so tiresome in his otherwise fine writing.

But maybe I should cut him some slack. Evidently, it must be some sort of Nash best-reply to fluff up one’s feathers this way and show no sign of weakness. There is always some right-wing nut job out there waiting to pounce, to tear apart, and to misrepresent anything that might be construed as capitulation on his part. He knows this. I know this. Now we all know this. So let’s set it aside and take a closer look at that data.

The chart above appears to be drawn from the Job Openings and Labor Turnover Survey (JOLTS). This is a great data set, but it has its limitations.

The question I’d like to ask is whether it really is the case that there are more unemployed workers than available jobs. According to JOLTS, the answer is yes. But this does not mean it is so in the economy.

It could be the case that many, perhaps even most, job openings are not advertised (hence not picked up by JOLTS). There are, evidently, a lot of farm jobs available that Americans refuse to work at (see: Despite Economy, Americans Don’t Want Farm Work). Many unadvertised jobs are poor-paying jobs. Everybody knows they’re out there. If you need a quick (and legitimate) buck, you send your application to McDonald’s. There are arguably millions of these low-skill low-pay jobs around. Jobs are not scarce. (What is scarce are good jobs that are well-matched with the characteristics of all those available to work.)

The JOLTS data itself provides some evidence that many job openings are not measured. In particular, take a look at this:

So there’s no question that right now, the supply side is what is constraining unemployment.