May Payrolls (and more) Confirm Slowdown (and more)

What’s left open for argument and concern is now a matter of how much of a downside there might end up being. Politicians and other officials have confidently reassured the public this is a natural and welcome progression or transition from red-hot inflation-y to, in the President’s word, stability.

Markets (inverted curves) are stridently betting there’s more to it than that; and by more, they’ve priced less.

Having set up basically an arbitrary statistical range, we have to judge the data by those bounds rather than by indicated headcount. While 400,000 months are double what the labor market used to average before 2020, that’s not an apples-to-apples comparison.

Building off this year’s benchmark changes, we start with how between October 2021 and February 2022 the monthly gains were consistently near the top of the range. Since gasoline prices surged in March, they now average consistently near the bottom. With three in a row in that position, chances of this being random noise or statistical aberration fall with each successive monthly datapoint.

This “something changed” is replicated all over both the CES and CPS figures (ADP’s, too).

I pointed out last month how it had been the Household Survey which already suggested that very point. Given the HH is the noisy stepcousin of CES means you can’t put too much emphasis on a single month of data, though the fact that it was the first decline since the recession in 2020 already strongly proposed there was more to it than randomness.

By being substantially negative, that itself meant there was a more than significant possibility April truly had been a “bad” month for workers even if we couldn’t say for sure just how bad.

As expected, the latest HH employment estimate for May rebounded, though not enough to offset last month’s drop.

To show you what I mean, look (below) at how many times (how few) the HH Survey has been negative on any two-month basis over the last eleven years dating back to 2011 and the start of the labor market “recovery” after the Great “Recession.” Apart from 2013’s government shutdown and 2017’s double-hurricane wreck, there are only a handful of two-month minuses and every one of those corresponds with economic downturn.

In other words, the economy finished up last year and began this year on far more uncertain ground than it might otherwise have appeared, or had been made to appear (especially when viewed from any CPI perspective). The jobs market wasn’t taking off into the red-hot stratosphere of Phillips Curve inflation, merely still trying to dig out from the prior huge hole, more and more struggling to do so as that momentum of reopening began to fade.

The difference between those two scenarios accounts for what we see developing now (and why markets have been priced in anticipation of it). Had it truly been inflation/max employment/full recovery playing out instead of mere reopening, we wouldn’t see wage gains decelerate as they have nor could we account for the timing.

It’s not a crash in wages by any means, but that’s not something you ever see anyway. Second derivatives are what matter here, and whether by 3-month change or 6-month, wage gains in hourly and weekly terms are absolutely slowing down and have been since late last year (that is Euro$ #5 not labor shortage).

Even the level of acceleration in wages had always been a product of the supply shock, too, like overall consumer prices a much faster recovery in demand for workers than the labor market’s ability to supply them.