One Slack Measure to Rule Them All?

Some quick thoughts about the labor market and wage growth

If you have to look at one number to understand how much slack is left in the labor market, where do you look? As the WSJ points out, quite a few people like the difference between the U6 and U3 measures of unemployment. The difference shows (roughly) the share of the work force that is “marginally attached” or working a job part time even though they’d like more work.

Another popular measure, and one that I and others at Equitable Growth like is the employment-to-population ratio for workers between the ages of 25 to 54. Or the prime-age EPOP.

I was thinking about this a bit and played around with some data on FRED and thought I’d share the results.

Quick reminder on wage data: we have average hourly earnings for all workers and for non-supervisory and production workers. Let’s just call them AHE and NSP for short. They come out once a month with the jobs report and have a relatively narrower definition of our other wage metric.

Which is the Employment Cost Index. It comes out quarterly, but has the benefit of capturing more of what employers are actually compensating workers with.

Here’s what wage growth looks like for those three measures over since before the Great Recession

They seem to be in agreement for most of the recovery until we get to 2014. ECI seems to accelerate, NSP seem to be heading higher as well before declining at the end of 2014 and the beginning of 2015. AHE, however, really doesn't seem to be moving.

Slack definitely seem to be on the decline. But if you want to cite one stat, which one do you turn to? Prime-age EPOP? U6-U3?

First, what’s definitely not a good measure right now: the unemployment rate.

(Note: I didn’t include non-sup and production wage growth in the following charts. They came out too cluttered and figured the other two stats were more important to keep given the debates about wage growth in recent months.)

The decline in U3 doesn't seem to be matching up with the movements in the wage growth data. Maybe the up-tick in the ECI is a sign of declining slack as the unemployment rate gets closer to 6 percent. AHE is just sitting still though.

How about the difference between U3 and the U6 measure of unemployment?

Similar thoughts here. ECI seems to be picking up as U6-U3 gets below 6 percentage points. But no smoking gun here.

How about prime-age EPOP?

I know, I know. Duel y-axes = the devil. But it works here.

Looks to me that the up-tick in ECI growth seems to be happening around the same time as the up-tick in the growth of the prime-age EPOP. AHE, shockingly, doesn’t really seem to be doing much.

Obviously a quick and way too simple exercise to really change opinions. But I thought it was interesting how well prime-age EPOP did and that it tracked ECI so well.