May 2022 payroll estimates weren’t quite at the level of downgrades that President Phillips had warned of, though it’s increasingly only a matter of time. In fact, although the overall monthly change in the establishment survey is slightly better than expected, it and more so, other employment data still show a clear slowdown in the labor market.
What remains open to discussion and concern is now a question of how much harm there might be. Politicians and other officials confidently assured the public that this was a natural and welcome progression or transition from searing inflation to, in the president’s words, stability.
The markets (inverted curves) are betting vehemently that there is more to it; and in addition, they are less expensive.
The establishment survey had been thought to add about 350,000 this month, up from about 450,000 in recent months and on its way to the 150,000 Biden had predicted. The BLS today reported a rise of 390,000 from April to May seasonally adjusted, but also downward revisions to each of the previous two (March and April). Everything eventually evened out, which is exactly what CES is “famous for.”
After essentially setting up an arbitrary statistical range, we have to judge the data by those limits rather than the stated strength. While 400,000 months is double the labor market average before 2020, this is not an apples-to-apples comparison.
Building on this year’s benchmark changes, we start with how, between October 2021 and February 2022, monthly gains have consistently been near the top of the range. Since gas prices jumped in March, their average is now constantly close to the bottom. With three in a row in this position, the odds of it being random noise or a statistical aberration decrease with each successive monthly data point.
Rather than staring at +390,000 and breathing a politically sized sigh of relief thinking all is well with the economy appearing to hold on despite all the headwinds, what the data actually shows is that Something amended.
This “something has changed” is reproduced everywhere in the figures of the CES and the CPS (the ADPs too).
I pointed out last month how it was the household survey that already suggested this specific point. Given that the HH is CES’ loud cousin-in-law, you can’t overemphasize a single month of data, although the fact that it was the first drop since the 2020 recession suggested already strongly that there was more than chance.
By being materially negative, this in itself meant that there was a more than significant possibility that April had indeed been a “bad” month for workers, although we could not say for sure by how much.
As expected, the latest HH employment estimate for May has rebounded, although not enough to offset last month’s drop. In other words, it’s been two consecutive months where the net change remains negative, further reducing the odds of it being an anomaly, simultaneously propelling the possibility that the data really does pick up something bad, at any least like the establishment survey. a still undetermined slowdown in the labor market.
To show you what I mean, look (below) how often (how few) the HH survey has been negative on a two-month basis over the past eleven years since 2011 and the start of the “recovery”. of the labor market. after the Great “Recession”. Other than the government shutdown in 2013 and the sinking of the 2017 double hurricane, there are only a handful of two-month downsides and all one of them is the economic downturn.
The HH survey also agrees with CES regarding the second half of last year and part of this year. Both point to an acceleration in the labor market, although in this series it is believed to have occurred between July 2021 and March 2022.
Given that euro $#5 started in May 2021 (according to my unofficial statement), this acceleration went against what “should” have been, and normally would have been, a much faster decline of employment (considering employment as a lagging indicator) thus strongly suggesting that a particular rise in the labor force had been uneconomic; the last big push of the definitive reopening (after the delta-COVID interference).
This non-real recovery is almost certainly why the employment picture had seemingly challenged the #5 euro, if only more recently.
In other words, the economy ended last year and started this year on much more uncertain ground than it could have appeared or had been made to appear (especially from the point of view of CPI). The labor market was not taking off in the scorching stratosphere of Phillips Curve inflation, just trying to climb out of the previous huge hole, increasingly struggling to do so. the momentum of the reopening has begun to fade.
The difference between these two scenarios explains what we see developing now (and why markets have priced in anticipation of this). If it was really inflation / maximum employment / full recovery that was happening instead of just reopening, we wouldn’t see wage gains to slow down as they did, and we couldn’t explain the timing either.
If the economy was currently playing along the Phillips curve, we would be above full employment and wages would be accelerating wildly – you know the thing: the shortage of labor and the scarcity of workers stimulate competition between companies which must then pass on higher labor costs to consumers.
It’s not a drop in wages by any means, but it’s not something you see anyway. Second derivatives are what matter here, and whether by 3-month or 6-month change, wage gains in hourly and weekly terms are absolutely slowing and have been since the end of last year (i.e. say the euro # 5 and not the labor shortage).
Even the level of wage acceleration has always been the product of the supply shock, as have aggregate consumer prices, a much faster recovery in the demand for workers than the ability of the labor market to supply them. It was never a shortage, just the same artificial imbalances (rightward shift in demand given the inelastic labor supply, often self-imposed, raising prices but not the level activity report; see below).
Put all of that together and yes the labor market slowdown, although that one stems from the emerging downside of that supply shock not an inflationary overshoot beyond the recovery, now bailed out by a government proactive that raised rates (and the absurdity of QT’s theatrical manipulation of ever-useless bank reserves). As I wrote a few days ago:
There is a big difference between it’s-alright-so-pump-the-brakes-a-little-bit and the plane-is-never-really-far-from-the-ground-and-now-its-engines-are-off. The political theater surrounding rate hikes and now this administration-wide Phillips curve trend is to make believe the first.
Therefore, the slowdown that is already everywhere in all labor market data looks very different from the “stability” heralded by President Phillips. Despite the relief from this month’s headlines, even that is more than meets the eye. And by more, like markets, I mean less.
Editor’s note: The summary bullet points for this article were chosen by the Seeking Alpha editors.