Total nonfarm employment in the US returned to pre-pandemic levels in July, hotly-anticipated government data out Friday showed.
It was a milestone. The White House will doubtlessly seize the opportunity to counter a deluge of bad inflation press by touting the administration’s efforts to engineer a complete restoration of the labor market.
Although this won’t be explicit, the implication of any such spin will be clear enough: Yes, inflation is unacceptably high, but the tradeoff was worth it. I’m not weighing in on the veracity of that claim. I’m merely telling you how Democrats will spin the numbers.
The US economy added a remarkable 528,000 jobs last month, the government said, more than double the 250,000 additions consensus expected (figure below). No economist out of nearly six-dozen surveyed predicted a headline NFP print above 325,000.
Revisions added a combined 28,000 jobs to the prior two months.
With July’s gain, the economy has added 3.3 million jobs in 2022. Say what you will, but that’s a lot of jobs in a very compressed time frame.
The unemployment rate likewise recovered pre-pandemic levels, dropping to 3.5%, below consensus, and tied with February 2020 for the lowest since 1969 (simple figure below).
The participation rate ticked lower (again) to 62.1%.
Wage growth was scorching-hot. Average hourly earnings rose 0.5% MoM, far more than expected, and 5.2% YoY, easily ahead of the 4.9% economists projected. Those prints were consistent with the message from last week’s Q2 employment cost index data: The US is at risk of a wage-price spiral.
Taken together, Friday’s figures suggested labor demand remains voracious and labor supply constrained.
The read-through for markets was straightforward. Fed officials keen to cite a strong labor market in the service of dispensing with a recession narrative that’s gathered adherents over the past two months were handed an extremely robust jobs report accompanied by wage growth figures which are in no way consistent with the Committee’s inflation target.
July’s NFP report argued loudly for the continuation of rate hikes, and should increase the odds of a third consecutive 75bps hike at the Fed’s September meeting. If July’s CPI report comes in hot, expect markets to price a 75bps increment for next month’s gathering as a virtual lock.
Some market participants recently pointed to a disparity between the household and establishment data in the jobs report to suggest negative readings on the former may presage downward revisions to the headline NFP prints. July’s report showed a 179,000 gain on the household survey.
That pushed the three-month moving average back into positive territory (figure above).
And in any event, there’s scant evidence for the narrative that says negative household readings are generally associated with an eventual “reckoning” (so to speak). Earlier this week, Bloomberg’s Cameron Crise ran the numbers.
“I sorted for all months since 1996 in which the initial nonfarm payroll gain was at least 200,000 and the household data was negative, and checked to see what happened with the final NFP revisions,” he wrote, adding that as “it turns out, not only did none of these episodes occur during an NBER recession, but the average payroll revision was +19,000.” He called the results “kind of surprising,” on the way to concluding that, “I guess it illustrates why household employment figures are relegated to secondary status in the monthly report — they don’t really tell us that much about the underlying state of job growth.”
Employment is a lagging indicator, and jobless claims have moved up. The four-week moving average for initial claims is the highest since November, and job openings fell the most on record in June outside of the initial pandemic plunge. At the same time, a bevy of corporate heavyweights have announced hiring freezes and layoffs as part of an ongoing reassessment of staffing needs in the context of an expected deceleration in economic activity both at home and abroad.
All of that said, July’s jobs report was unequivocal: The US labor market hasn’t rolled over yet. Maybe, in the fullness of time, we’ll discover that in fact, the US was in a proper recession during the summer of 2022. As any regular reader will quickly attest, I’m certainly open to such a thesis. I’d wager I’ve seen more recessions than many of my readers. I’m the furthest thing from a macro Pollyanna.
But right now, as it stands, it’s very difficult to make the case for a recession in the NBER sense, and even if you could make the case, the Fed wouldn’t listen, which is really all that matters.
Of course, the Fed will get another jobs report between now and the next policy decision. Anything can happen. BMO’s Ian Lyngen captured it well. “If the FOMC meeting was tomorrow, [July’s jobs report] would make a 75bps hike the path of least resistance,” he said. “Alas, there is still a lot of data between now and the September 21st meeting.”
Glad you led off with the milestone aspect. It matters to me in a gut sense, another signpost indicating there’s a way out of this global mess, eventually. Should be an interesting day as the market digests all this, but I’ll bet if one counted up the number of job-report days in history where the number was 500k+ and the market closed red, it would be 0.
You are spot on- if a Fed meeting was held tomorrow we would be looking at a big FOMC hike. The numbers suggest that employers are hoarding labor in past times many of the hiring freeze announcements would be layoffs- so it looks like you can throw out the playbook on this economic cycle. It is going to be very interesting for this rest of this year.
I was looking to see if the huge discrepancy between the establishment versus household surveys persisted. It did.
I also found this little nugget:
The number of persons employed part time for economic reasons increased by 303,000 to
3.9 million in July. This rise reflected an increase in the number of persons whose
hours were cut due to slack work or business conditions.
But as our Dear Leader pointed out, that doesn’t matter if the Fed chooses to overlook those stats.
Jobs gain, mostly by women, with no increase in participation rate, and increase in PT work.
Hints at more people taking second jobs, companies hiring but not at full hours, more “low quality” jobs.
You did read all the color in the article about the “huge discrepancy,” right? If you run the numbers, there’s scant evidence to suggest that the combination of robust headline NFP and negative household survey prints invariably presages downward NFP revisions.
As Bloomberg’s Crise went on to point out, even if you “broaden the scope of the study to sort for all months in which the household data was negative and the initial payroll figure was positive” you still get the same conclusion.
Crise called the results “surprising,” and some might even find them frustratingly counterintuitive, but this is another one of those cases where the numbers either support the thesis or they don’t. In this case, they don’t.
You can’t just plot the household survey with the NFP headline using a six-month window (as I noticed one notoriously silly website did), circle the point of divergence and call that “evidence.” You actually have to go back, find the original (i.e., pre-revision) NFP prints that corresponded with negative household survey prints, then compare the pre-revision NFP prints to the twice-revised, final figures.
When you do, here’s what you discover: https://heisenbergreport.com/wp-content/uploads/2022/08/NFPHouseholdRevisionChartJul2022.png
Thanks. I DID read your missive after I went to the BLS site to look. So it appears just to be a curious anomaly of some sort. I’ve been marginally curious about the household survey results after being a participant 20 or so years ago. It was a PITA because every week they kept asking the same questions in multiple ways. At one point I shouted “Please let me summarize this for you – my job has not changed nor the hours I work very week. Thank you.”
That said, all of these survey-based stats with heavy seasonal adjustments are best taken as indicators rather than hard stats.
To be sure, I don’t “trust” any of these numbers at all. I put the scare quotes around “trust” to indicate that, as most of you know, I despise conspiracy theories, so I’m certainly not using “trust” in that sense of the word. I just mean that I don’t think any of the data captures what it purports to capture accurately and could therefore be totally unreliable at any given time.
That’s just a circuitous way of saying that you could be totally right and so could all the people suggesting the labor market isn’t nearly as strong as it looks. I have no idea. And neither does anyone else, including the people who compile the data.
The only thing I do know, from extensive personal experience, is that finding discrepancies and otherwise poking holes in whatever the “official” narrative is, long ago completed an unfortunate transformation from critical endeavor aimed at uncovering information we can all use to make better decisions to parlor game, where most of the participants are playing for amusement with nothing material staked on the analysis being accurate.
That, more than anything else, defines my daily dilemma: How to convey my deeply ingrained, old school skepticism without lapsing into “new school”-type bear propaganda. It’s a very delicate balance, mostly because you can be right 1,000 times as an optimist, but the one time you’re wrong, it undermines your credibility. Bears, on the other hand, enjoy a situation where they can be wrong 1,000 times, but are celebrated as geniuses the one time they’re right.
I never win that psychological tug-of-war with myself. It’s ongoing. The best I can do is call it “balanced coverage.”
“I never win that psychological tug-of-war with myself. It’s ongoing. The best I can do is call it “balanced coverage.”
Sir, you do that damn well. Nicely done!
Any remarks on that labor participation rate? It’s looking more like one of those “since the 70’s” records except that was a historical time with women not yet entering the labor force in full.
Can someone explain to me why 528k additional jobs is so bad? So the way I see it is that, with more openings filled these additional employees are partly able to address some of the supply constraint issues. Consequently, with fewer job openings, employers will no longer need to offer higher rates to fill those jobs and therefore this should dampen down wage inflation. Whilst a lower unemployment rate of 3.5% should bid up wages, could a lower participation rate be counteracting some of this as well? Surely, the important metric in all of this is the MoM wage inflation number of 0.5%, which annualises to 6.2%, but this is very much a lagging indicator. This figure is well below the headline inflation number that would allow workers to maintain their current living standard and is a catch up consequence to the initial inflation shock from higher commodity prices feeding through the economy. I would be interested to know how much weight the Fed attaches to these numbers in making their rate decisions.
Based on conversations around town- from auto mechanics shops to restaurants- employers are having to give up on finding employees who are willing to work 6 days/week and be drug tested.
A significant number of workers only want to work 4 days/week and not be subjected to drug testing. Several shop managers/owners that I have spoken with have said that instead of trying to hire one person to fill a 6 day position- they are hiring 2 people. They also said they think they can sell the extra business needed to justify the additional labor costs.
It seems that by 2023, our economy will look different than it did in January, 2020 in terms of labor, remote work, spending/ savings patterns, interest rates, globalization, etc. The stock market is still trying to figure this out- me too.
Texas Governor Abbott is shipping newly arrived migrants from the southern border to D.C. and NYC. I’m not one to make light of immigration struggles, but I do think the mayor’s of both cities could get creative with employers and embrace his political move. Low skill (how I hate using that term) workers are needed and it would turn a political stunt on its head without just crying about it.