I said in last week's video that "the economy could help out the Fed by slowing." Meaning, if the jobs data were to weaken, and if CPI next week were to soften, pressure to raise the fed funds rate would abate markedly, and of course risk markets would love it.
Well, Friday's jobs number -- coming in much better than expected -- indeed did not help, and markets -- albeit not terribly -- sold off accordingly.. Although I suspect the "not terribly" had to do with the fact that under the hood the jobs data really wasn't all that stellar.
Here's from the macro wrap below:
"We are not going to overstate it, because the composition deserves scrutiny. Restaurants and bars added 59,000 jobs after two months of declines. Local government schools added 42,000, reversing the prior month. Government payrolls swung by 85,000. Those are corrections of earlier weakness rather than new hiring. Information-sector employment fell again, concentrated in data processing and web hosting — the categories where artificial intelligence appears to be displacing work rather than creating it. And the average monthly job gain over the past year is still only 31,000."
As for next week's inflation data, while, for obvious reasons, I'd love to be able to handicap it, well... here's more from this weekend's wrap:
"The service sector's measure of input prices reached a four-year high in August. The manufacturing equivalent stayed elevated. The Federal Reserve's own survey of its twelve districts found prices rising in eight of them. Oil rose roughly 9% on the week. Businesses reported paying more for fuel, diesel, gasoline — and, notably, for the specialized computer chips that power artificial intelligence, which were newly added to the list of goods in short supply."
Yet, at the same time:
"...labor costs per unit of output rose at an annual rate of just 1.2% last quarter, while productivity rose 1.4%. Wage growth slowed to 3.1% over the past year."
I.e., prices are rising, but it's not due to a tight (or strong) labor market... And maintaining labor market strength is one part of the Fed's dual mandate... The other being inflation containment.
In last week's video I offered up my view of the federal debt constraints that I think are a much bigger deal than the market, and most macro commentators, are giving it credit for... Here's the link if you happened to have missed it.
Also, on that topic (the politics of it), from below:
"We have argued for months that this Federal Reserve faces political pressure that will bias it toward patience, whatever its rhetoric. That has been an inference on our part. This week it became visible: a Governor on the record, the executive branch on the record, and a market that moved thirteen percentage points on one man's sentence."
Now, all that said, we could nevertheless see the Fed hike rates this year -- the futures market certainly sees it happening... As for this and next months' meetings, I think that would be a very heavy lift with mid-terms looming... As for December, well, then I'm back to the trillions of debt due to roll over in the coming months... But, again, could happen if inflation doesn't cool off a bit between now and then.
Stay tuned...
Here's this week's PWA Index scoring summary, followed by your weekend macro wrap*:
The Week Everything Moved and Nothing Changed*
Week of August 31, 2026
Last week we wrote about the Chairman's speech and about the peculiar shape of the interest rate move that followed it. This week gave us a great deal more information, and almost all of it argued with something we thought we knew.
The job market surprised
For most of the summer, the story in the labor data has been slow deterioration. Hiring had slowed to a crawl. Two of the three summer months showed net job losses. And eight days ago, the government published its annual reconciliation of the payroll survey against actual tax records, which took roughly 79,000 jobs out of the count for the year through March — with retail employment alone revised down 155,000.
Friday's report went the other way, emphatically. The economy added 162,000 jobs in August, against expectations of about 53,000. June and July were both revised upward, by 55,000 combined — enough to turn July from a reported loss into a gain. The unemployment rate held at 4.1%. And the labor force grew by 683,000 people without pushing that rate up, which is the healthiest possible way for a jobs report to be strong.
We are not going to overstate it, because the composition deserves scrutiny. Restaurants and bars added 59,000 jobs after two months of declines. Local government schools added 42,000, reversing the prior month. Government payrolls swung by 85,000. Those are corrections of earlier weakness rather than new hiring. Information-sector employment fell again, concentrated in data processing and web hosting — the categories where artificial intelligence appears to be displacing work rather than creating it. And the average monthly job gain over the past year is still only 31,000.
But one number in that report cannot be explained by composition: the average workweek got longer. More people working, and each of them working more hours, is a genuine increase in the amount of labor the economy is using. That is difficult to argue with, and we are not going to try.
We moved our employment reading up a step this week. Not two steps.
Prices are the problem now, and wages are not the cause
Here is the combination that we think matters most, and it has not received much attention.
The service sector's measure of input prices reached a four-year high in August. The manufacturing equivalent stayed elevated. The Federal Reserve's own survey of its twelve districts found prices rising in eight of them. Oil rose roughly 9% on the week. Businesses reported paying more for fuel, diesel, gasoline — and, notably, for the specialized computer chips that power artificial intelligence, which were newly added to the list of goods in short supply.
And at the same time: labor costs per unit of output rose at an annual rate of just 1.2% last quarter, while productivity rose 1.4%. Wage growth slowed to 3.1% over the past year.
Read those together. Prices are accelerating and wages are not causing it.
This matters for two reasons. First, it removes the Federal Reserve's most familiar justification for raising rates — you cannot argue you are breaking a wage-price spiral when wages are decelerating. Second, and more relevant to how we invest: an economy where prices rise faster than labor costs is one where the arithmetic of the national debt improves, because that debt is measured against an economy valued in current dollars. We have written about this before. It is working. But it works only if the government's own borrowing costs stay contained, and this week they did not.
Short-term government bills — the instrument the Treasury relies on most heavily — repriced meaningfully higher across every maturity we watch. The one-year bill now yields more than the Federal Reserve's own target rate by a visible margin. That is the market charging the government more to fund itself, in the week the Chairman's colleagues began publicly disagreeing about what happens next.
A crack inside the Federal Reserve
Which brings us to the most consequential development of the week, and it was not a data release.
On Thursday, a sitting Federal Reserve Governor said publicly that he would be inclined to support leaving rates unchanged at the September meeting if the coming inflation report shows prices cooling. Markets moved immediately — the odds of a rate increase fell from roughly 62% to about 50% on that sentence alone, stocks had their best day of the week, and the dollar fell.
Also on Thursday, the Vice President said the Federal Reserve should be cutting rates to make housing more affordable — days after the Chairman had hinted at the opposite. And a third Fed official declined to endorse the Chairman's framing, describing the rise in long-term rates as a sign of economic strength rather than a problem requiring a response.
Then Friday's jobs report pushed the odds back up to 58%.
We have argued for months that this Federal Reserve faces political pressure that will bias it toward patience, whatever its rhetoric. That has been an inference on our part. This week it became visible: a Governor on the record, the executive branch on the record, and a market that moved thirteen percentage points on one man's sentence.
Surprises/Changing Signals
Two corrections, both worth stating plainly.
On energy, we got a signal and it evaporated. A week ago the physical data suggested oil was beginning to flow more freely out of the Middle East, and one credible estimate had exports back to roughly two-thirds of pre-conflict levels. We flagged it as the first real evidence against our positioning in energy, and we said we would verify it before acting on it.
Seven days later, that signal is gone. Hostilities resumed after roughly a month of calm. Shipping through the critical waterway fell back sharply. The world's largest tanker operator now says it expects no return to normal before year-end. Oil rose 9%, and inventories drew down far more than expected — a physical tightening, not just a headline reaction.
We did not act on the earlier signal, and that discipline is what protected the position. But the more useful lesson is about the measurement itself: a signal that reverses within a week is being measured too frequently for the decision it is meant to inform. We are changing how we monitor it — moving to a four-week average with a threshold written down in advance, rather than reacting to the latest weekly figure. That is a genuine improvement in process, and it came from being wrong.
On the consumer, our own indicator changed its mind. For three weeks we have noted that one of our internal measures — the relationship between defensive consumer staples and discretionary spending stocks — was failing to register a consumer slowdown that several other sources described. We flagged it as a limitation of the measure.
This week it started registering. It reached the highest reading of this cycle, driven by consumer discretionary stocks hitting a new relative low, in the same week the employment data was at its strongest. We moved that reading to its most negative level.
We want to be honest that this is contested. Restaurants added jobs. Auto sales beat expectations at their best pace of the year. Weekly chain-store sales accelerated for a second straight week. That is a consumer who is still transacting. It is entirely possible this is stock market rotation rather than household distress. Retail sales on September 16 will tell us.
What our internal gauge says
Our internal conditions gauge — the composite of 67 economic and market measures we maintain and score every week — moved to +5.97, down from +7.46.
That is a decline of a single field out of sixty-seven, in a week that delivered a 109,000-job surprise, a 9% oil rally, renewed conflict in the Gulf, service-sector prices at a four-year high, the ten-year interest rate at its highest since November 2023, and a public split inside the Federal Reserve.
We think that gap deserves an explanation rather than a paragraph pretending it does not exist. The gauge measures whether conditions cross from one category into another, not how loud the week was. This is now the third consecutive week where the news has been dramatic and the crossings have been few — manufacturing activity slipped but stayed firmly healthy, interest rates rose but stayed within their range, inflation expectations rose but stayed near target. The information was real. It simply landed in places already accounted for.
The reading is still comfortably positive, and still better than roughly 95% of the last four years. The level is what tells you something. The weekly change is not a summary of the week.
What we are watching
Thursday and Friday, September 10 and 11, bring producer and consumer price data for August. This is the most important item on the calendar. It arbitrates the disagreement inside the Federal Reserve, it tests the Governor's stated condition, and it resolves a genuine conflict between two national surveys that reported price pressure moving in opposite directions in the same month.
Wednesday, September 9, the Treasury's expanded bond-buying operations finally begin. As we wrote last week, whether long-term investors take that offer during a period of expected rate increases will tell us whether the problem in the long-term bond market is the supply of government debt or confidence in the inflation outlook. Those point to different futures.
Wednesday, September 16, August retail sales and the Federal Reserve's decision land on the same morning. The retail figure settles the disagreement inside our own consumer readings.
Friday, September 18, the Bank of Japan decides. Japanese officials spent this week opening the door to a larger increase than usual, and the yen had its second-largest two-day gain since 2024 as a result. We hold a position that benefits from a stronger yen, and we added to it modestly the day before that move. We are holding the remainder until after the meeting rather than chasing.
This commentary is provided for informational and educational purposes only and does not constitute an offer to sell, a solicitation of an offer to buy, or a recommendation regarding any security, sector, or investment strategy. The views expressed reflect the opinions of Private Wealth Advisors as of the date written and are subject to change without notice. Information has been obtained from sources believed to be reliable, but its accuracy and completeness are not guaranteed. Past performance is not indicative of future results, and all investing involves risk, including the possible loss of principal. Nothing contained herein should be construed as personalized investment, tax, or legal advice.
*The section titled "The Week Everything Moved and Nothing Changed" was drafted with the assistance of artificial intelligence tools under the direction and editorial review of Marty Mazorra, Chief Investment Officer of Private Wealth Advisors. All analysis, conclusions, and final content are his own.
No comments:
Post a Comment