August jobs report: possibly the best report so far all year

 

 - by New Deal democrat

My Big Theme for the past few months has been that the AI Boom (or possibly bubble) is counterbalancing a stagnant or even shallowly recessionary rest of the economy. Last month I wrote that July’s poor jobs report, which showed a -23,000 decline, had “a MAJOR caveat. Take out the -49,600 loss in local government education jobs, and we eked out a +27,000 gain for the month.” This month had the same caveat in reverse. Take out the 41,900 gain in local government jobs, and this month’s gain, while still good, was +120,000.

Below is my in depth synopsis.

HEADLINES:

• +162,000 jobs gained. Private sector jobs increased 127,000, while government jobs added 40,000, all of which were in local education. The three month average rose to 71,000.
• The pattern of downward revisions to previous months was reversed this month, as June was revised higher by +11,000, and July was revised higher by +44,000 (from a decline to a gain of +21,000) for a total increase of +55,000.
• The alternate, and more volatile measure in the household report, rose sharply, by +569,000 jobs. On a YoY basis, this series which had been negative for six months in a row, is now higher by 1.414 million. 
• The U3 unemployment rate remained steady at 4.1%, its lowest level in two years. 
• The U6 underemployment rate declined -0.2% to 7.7%, its lowest in over 12 months.
• Further out on the spectrum, those who are not in the labor force but want a job now declined -173,000 to 5.747 million, the lowest number in the past 12 months..

Leading employment indicators of a slowdown or recession

These are leading sectors for the economy overall, and help us gauge how much the post-pandemic employment boom is shading towards a downturn vs. rebounding. These were almost entirely positive for the second month in a row.

• The average manufacturing workweek, one of the 10 components of the Index of Leading Indicators, was unchanged at 41.7 hours, equal to the highest number in 5 years, just surpassing its 2021 peak.

• Manufacturing jobs rose +16,000, the 5th increase in the last 12 months.
• Truck driving reversed its decline for the second month in a row, increasing by +4,800.
• Construction jobs rose +22,000.
• Residential construction jobs, which are even more leading, rebounded their 3 year low last month, up +7,300.
• Goods producing jobs as a whole rose +41,000. 

• Temporary jobs, which had declined by over -650,000 since late 2022, rose by another +6,800, continuing to improve from their post-pandemic low set last October.
• The number of people unemployed for 5 weeks or less rose +40,000 to 2.000 million, still very low compared with the last 3 years.

Wages of non-managerial workers 

• Average Hourly Earnings for Production and Nonsupervisory Personnel increased $.11, or +0.3%, to $32.53, for a YoY gain of +3.3%. Except for last month’s +3.2%, and several months affected by pandemic shutdowns, this is the lowest since December 2019. This is equal to the 3.3% YoY inflation rate as of July.

Aggregate hours and wages: 

• The index of aggregate hours worked for non-managerial workers rosse +0.1%, and is up 0.9% YoY, about average for the past 12 months.

• The index of aggregate payrolls for non-managerial workers rose a sharp +0.5%, and is up 4.0% YoY, tied for the lowest comparison for the past 5 years, but up 0.7% above the YoY inflation rate through July.

Other significant data:

• Professional and business employment rose for the fifth month in a row, by +10,000. These tend to be well-paying jobs. This remains above its low from last October, and has turned higher YoY as well.
• The employment population ratio reversed its recent declines, rising +0.2% to 59.1%, vs. 61.1% in February 2020, and its lowest since October 2021.
• The Labor Force Participation Rate rose +0.2% to 61.6% , vs. 63.4% in February 2020, and the lowest since February 2021. IMPORTANT: both the EPOP and LFPR are greatly affected by the retiring Boomer population. In the prime age 25-54 demographic, they are virtually unchanged.

SUMMARY

Yesterday “finance bro” George Peakes commented on Bluesky that “People really want Trump to be an economic disaster and sorry, it's just not playing out that way;” and more succinctly agreed with a response that “He is an economic disaster (over the long term, given plausible models and parameters) but not the short term (recessions are not caused by vice).” This month’s jobs report is potent evidence of that point.

Most importantly, not only was the headline positive, and not only was the unemployment rate tied for a two year low (in accord with what I’ve been writing almost every week as a trend telegraphed by low jobless claims), but as indicated above, almost *all* of the leading indicators in the report, chiefly dealing with the goods-producing sector, continued to be positive, of a piece with the positive trends we have seen in the ISM and regional Fed manufacturing indexes for almost a year. And several “problem children,” i.e., professional and service jobs and temporary help, continued their rebounds. There is more tenuous evidence of a rebound in trucking as well.

To the extent there was a significant negative, it was that nonsupervisory wages continue to grow at a relatively low rate, and if inflation in August picked up again with gas price increases, meaning that real, inflation adjusted wages could be negative YoY for the 4th month in a row.

But given the breadth of the gains, and the positive leading signals going forward, this was probably the best report so far this year. More evidence of an inflationary expansion.

The economically weighted ISM indexes for August continue to show a stagflationary expansion

 

 - by New Deal democrat

The economically weighted ISM manufacturing + services indexes continue to be the best timely snapshot of the US economy. On Tuesday the manufacturing sector was updated; this morning services were.  The weighting, based on their impact on the economy, is 25% manufacturing and 75% services. Further, to cut down on monthly noise, I particularly look at the three month averages.

The summary version is that both the headline and the more leading new orders components continue very positive, but the prices paid component indicates that if anything inflationary pressures are increasing, while employment is showing outright contraction.

To the numbers: the headline services index rose 0.8 to 55.4 [recall that any number over 50 indicates expansion]. The three month average was 55.2. Since the three month average for manufacturing was 54.5, the economically weighted average was 55.0 [note: in all of the graphs below, the manufacturing number is blue, and services gray]:

New orders rose 3.7 to a very strong 60.9, its most positive reading in over three years. The three month average was 557. The three month average for manufacturing was 55.5, so the economically weighted of this most forward looking component was 57.2:

So far, so good. But employment in the services index was contractionary for the second month in a row, rising 0.4 to 47.8. The three month average was 48.8. Since the manufacturing employment subindex averaged a slightly positive 51.2, the economically weighted average was below 50 for the second month in a row as well, increasing 0.2 to 49.4:

Importantly, although it has been better than summer of last year, the ISM weighted average has only shown expansion in two month this year: February and June. The authoritative QCEW metric, which suggested that nonfarm payrolls overcounted employment in the first three months of this year, thus also suggests that we may see more weakness in both the monthly and benchmark revisions of that metric. And we’ll see how August compares tomorrow.

Finally, widespread price increases continue to be a problem, with the prices paid index for services rising 2.3 to 72.6, with the three month average at 70.2. The three month average for manufacturing did ease a little this month at 71.7, but the economically weighted average increased 1.2 to 72.4:

This isn’t quite as bad as during the post-pandemic inflation, but not by much. Indeed, This is the worst reading since mid-year 2022 (note that unlike the other three graphs, this one shows the last five years for better comparison).

To recapitulate, the economically weighted ISM averages indicate that as of August, the economy remained in reasonably strong expansion, and new orders suggest it might get even stronger. But inflationary pressures are getting even stronger, and employment is not increasing at all. In other words, as I’ve written before, a stagflationary expansion.

(Almost) nobody is getting laid off - still

 

 - by New Deal democrat

Let’s take our regular weekly look at jobless claims, a very good short leading indicator. To cut to the chase, nobody is getting laid off - still.

Initial claims rose 2,000 to 206,000, while the four week moving average rose 1,500 to 207,250. Both of these remain very low numbers on a historical basis. With the usual one week delay, continuing claims rose 8,000 to 1.779 million, typical for the post-pandemic era:

On the YoY% basis more important for forecasting, initial claims were down -12.7%, the four week average down -10.1%, and continuing claims down -8.2%:

These are *very* positive numbers, consistent with a good economic expansion.

Finally, let’s take our last look at what this suggests about the unemployment rate over the next few months:

All of the pressure is to the downside, i.e., the unemployment rate declining towards 4.0% or even lower. We’ll find out if that was the case in August tomorrow.

July manufacturing core capital goods order add to evidence of manufacturing rebound; while transportation (for June) flags

 

 - by New Deal democrat

Yesterday our month started out with reports that manufacturing in August was less positive than in the several months prior, while construction (ex-AI data centers) was absolutely recessionary. This morning was followed up by the final July report on manufacturers’ durable goods orders, a short leading indicator. 

While total durable goods orders (blue) rose, core capital goods orders (red), which are much less volatile, backed off slightly from record levels:

The YoY comparison shows, particular with regard to core capital goods orders, that expansion in the sector is still accelerating, with core orders up 12.6%, the biggest increase in four years, and only slightly behind June’s level:

Meanwhile the Freight Transportation Services Index for June (so lagging by several months) showed a -0.3% decline to the lowest level since December 2022:

It is interesting to compare this with sales of heavy weight trucks, which I find very useful because they generally decline sharply, and with much less noise, well ahead of recessions, as well as increasing a number of months into expansions:

For oncoming recessions, sales of heavy weight trucks have given the better and more leading signal. Interestingly, though, on a YoY basis, the freight index has tended to turn negative a number of months before sales of heavy weight trucks do:

For the last two months, truck sales have turned slightly positive YoY, in accord with their general recovery this year. But the freight services index is down -1.7%. That’s not recessionary (for that I would expect to see readings worse than -2.0% YoY on a three month average basis), but it does raise the question of whether sales of heavy weight trucks might turn back down in the next several months.

In general, this is confirmation that manufacturing has been doing well this year, while transportation, which includes construction materials as well as manufacturing inputs and outputs, has been much more mixed.

July JOLTS accords with low-hhire, low-fire economy; so does August ADP employment estimate

 

 - by New Deal democrat

One piece of data released yesterday that I didn’t report on was the JOLTS labor market report for July. This parses turnover in the market by hires, layoffs, and quits, among other things. It is a minor indicator, but let’s take a look.

The first graph below shows the “soft statistic” of job openings (blue), actual hires (red), and quits (gold), all normed to 100 as of just before the pandemic:

can see that for the past several years the “hard data” of hires and quits have run below their level of 2019. Only openings, which even at the worst level were just barely low their 2019 level, have been increasing again this year. In July actual hires were at their 3rd lowest for the entire post-pandemic period, in accord with the actual jobs losses suggested by last month’s employment report.

Similarly, here is the graph for layoffs:

Layoffs have been running at close to their lowest post-pandemic levels over the past nine months.

In other words, a very low hire and low fire job sector.

Also, a good case has been made that the quits rate (blue, right scale) leads YoY average hourly wages (red):

For the past year, the quits rate has been close to completely flat. That argues that *nominally* average hourly wages YoY should be stabilizing at a 3.4%-3.5% rate. 

Much more currently, this morning’s ADP employment number for August (red) showed a slight 44,000 gain in private jobs. I also show the more noisy weekly ADP number (blue) in comparison with the official payrolls number (gold, right scale):

What is interesting is that the ADP numbers more closely track the official QCEW numbers (to which payrolls will be benchmarked) which were finalized for 2025 last week, i.e., slower hiring in the first three quarters of last year, followed by improvement at the end of the year and mild improvement in the first several months of this year.

August manufacturing “less good”; July construction recessionary

 

 - by New Deal democrat

As per usual, we start out the month with the ISM manufacturing report (for August) and the construction report (for July). Since these are two of the sectors that lead the economy, they give us a good first look at the remainder of the year.

And to cut to the chase, the first, while positive, was less so than in the past few months. The latter, for the second month in a row, was outright recessionary. 

Let’s start with the ISM manufacturing report. The headline number declined -1.0 to 54.6 (any number above 50 signifying expansion). For forecasting purposes, I average the last three months, which comes out to 54.5:

The more leading new orders subindex declined -3.0 to 53.7, and the three month average was 55.5:

Both the headline and new orders numbers indicate continued expansion, albeit at a lesser pace, while the three month averages remain very good. This means the expansion in this sector can be expected to continue at least a few more months.

The “less good” news didn’t stop there. Employment decelerated -1.6 to 51.2. The three month average also remained (slightly) in expansion at 51.2:

Finally, there was another negative month in the prices paid subindex, which was unchanged at 71.1, indicating very widespread price increases upstream. The three month average was 71.7, significantly below its worst levels from the last few months, but still very close to the post-pandemic inflationary peak:

In other words, the manufacturing rebound this year continues, but at a cooler pace. What *hasn’t* slowed down is the upward price pressures. Can you say “inflationary expansion”? 

Keep in mind that for forecasting purposes, I weigh manufacturing at 25%, and the other 75% from the ISM services report, which will be updated Thursday.

But if manufacturing indicated a somewhat slower inflationary expansion, for the second month in a row the construction spending report was if anything recessionary. Total construction (blue in the graph below) declined -0.5% in July, and was down -3.8% YoY. The more leading residential construction sector (red) declined -1.1% for the month, and is down -7.3% YoY. Since the price of construction materials (not shown) rose 1.4% in July and was up 10.5% YoY, the declines in real terms were even steeper:

The dismal news continued in manufacturing construction spending as well, down -1.0% for the month and down -21.2% YoY and over 30% from its September 2024 peak:

Quite simply, all of the above are recessionary.

Even spending on AI data center related construction didn’t help that much.  Here is an update on spending on power construction (blue, left scale) and water supply (orange, right scale), the two sectors most closely aligned with the AI data center Boom:

The former rose 0.5% for the month, and is up a strong 5.3% YoY at least nominally, while the latter, after declining in June, was unchanged in July, and is up only 0.2% YoY at this point. Keep in mind once again that these are nominal figures and don’t take into account the 10.5% increase in construction materials YoY. In other words, even the AI data center construction center may be flagging.

So, to recap: manufacturing continues to expand, but at a slightly attenuated pace, but with continued strong inflationary pressures; and construction outside of AI is recessionary. This makes the services report which will be released on Thursday all the more important.

The post-pandemic affordability crisis and the election of 2024

 

 - by New Deal democrat

There are times when there is little new economic data of import, and not too much else to say. At other times, the import of particularly striking data on even long term historical trends could justify entire book chapters almost overnight. Right now is one of the latter. Bear with me a little bit while I lay the groundwork.

This particular moment got started when Jamelle Bouie took Michelle Cottle to the woodshed over the issue of “wokeism” in a podcast I saw reported late last week. Here’s the essence of the exchange:

Cottle: My guess is that you then completely reject the idea that we wound up with this wretched administration in part because the backlash to the excesses of woke, especially alienating young men who were like based told they were the problem, just generally speaking.

Bouie: I completely reject that for like simple reasons of like linear time. That like this administration was elected in 2024 at peak woke when like Nancy Pelosi was wearing Kente cloth, Democrats won a trifecta, right? …. I think that if you look at the 2024 election and you ask yourself, why did Trump win? The very obvious answer is he won because inflation was high and people didn't like it. And people backfilled a lot of cultural reasons. [P]eople [we]re mad about inflation, and that directly tracks with President the President's declining approval.”

G. Elliott Morris amplified that in a substack article, which I excerpt below:

The reality of the 2024 election is that it was going to be hard for a Democrat to win, regardless of who they were or how they campaigned. The broader economic and political conditions were so favorable to Republicans that you would have expected Trump to win about 90% of the time, regardless of campaign or candidate effects.

Political scientists have been pointing out for decades that you can predict presidential elections reasonably well using just two pieces of information: how voters feel about the incumbent president, and how voters feel about the economy. There are many variants of this model — such as the “Bread and Peace” model (Douglas Hibbs), the “Time for Change” forecast (Abramowitz), Ray Fair’s “Fair” model, and Wlezien/Erikson’s work with “Leading Economic Indicators” — but all use a similar set of economic and political “fundamentals” to predict the result of the election.  

….

In 2024, Kamala Harris received about 49.3% of the two-party vote. The model — fit on data from 1956 through 2020, with 2024 held out — predicted she’d get about 48%, with an 80% prediction interval of 46.6% to 49.9%. Harris’s vote share lands on the upper end of this range, but still squarely inside of it.”

Here is the model that Morris is talking about:

He concludes: “[G]enerally speaking, ‘inflation was high, and Harris was going to lose anyway’ is a much better explanation for 2024 than anything else in isolation.”

That is a very unpopular opinion in many progressive quarters, especially when you look at the breakdown in the 2024 vote by race. It is very obvious that T—-p was popular with, and remains generally popular with, Whites, and in particular White men. But elections are decided at the margin. Rock solid bigots are going to vote bigot, no matter what. More broadly speaking, people tend to have unmovable opinions about what are generally called “social issues,” vs. more malleable opinions about the broad economy. And so generally it is the latter group of people who wind up deciding national elections. And incidentally, I tracked both the “bread and peace” and “leading economic indicators” models during 2016, and both - unlike the political pundity - forecast a very close election, with Clinton getting slightly more votes. She did, but lost in the Electoral College.

Even during 2024, that the economy was not actually doing all that well in terms of delivering goods to the people, broadly measurered, was something I argued many times.

What are the two most important purchases that average consumers ever make? Houses and cars. And younger people in particular were priced out of the market during most or all of Biden’s term.

Here is what happened to the average price of existing homes, as measured by the FHFA repeat sales index (blue) vs. average hourly nonsupervisory wages (red) normed to 100 as of just before the pandemic:

The average price of a house increased 38.2% by June 2022, while average wages were only up 14.5%. And that imbalance has never resovled. Even as of this past June, house prices were up 57.8% since February 2020, while wages were up 34.7%.

Here’s the same comparison applied to the average of new and used car prices:

At their peak difference, in February 2022, vehicle prices were up 28.4% compared with just before the pandemic, vs. 12.2% for wages. The difference did not finally resolve until May 2024, when both were up 24.8%.

And that’s not all, because the interest rate to finance mortgages (purple) and vehicle loans (orange) increased sharply as well [note: the latter are dots because the data is only reported once a quarter]:

Just before the pandemic, mortgage rates averaged 3.47%. At their peak in October 2023, they were 7.62%. Vehicle loans averaged 5.29% just before the pandemic, and peaked in February 2024 at 8.65%.

Finally, here is what the graph of monthly real median household income, compiled by Motio Research, looked like through December 2024:

Real median household income languished below its pre-pandemic 2019 peak all the way until the end of 2023. By Election Day 2024, it was only about 1% higher.

It’s true that job creation was red hot and even white hot throughout Biden’s term. And it is also true that real wages rose stoutly from their June 2022 trough linked to gas prices during the initial Ukraine invasion by Russia. But it is not true that average consumers, and in particular younger ones, prospered during Biden’s term - particularly as to the items they needed to be most affordable - at least not until 2024, by which time it was too late.

There is much more to say of a much broader issue that is led into by the above. For now, let me just give you a quick foretaste. The below is the updated graph through last Friday of corporate profits reported to Wall Street through last Friday:

At an index value of 110.26, they are almost 50% higher than they were even one year ago. And they are almost double their index value of 54.45 from only three years ago, in Q2 2023.