The Aug PPI declined -0.1%, softer than expected (0.3%) although PPI excluding food, energy & trade was in-line with expectations at 0.3%.
What does this mean?
Core goods prices are still rising at an above trend pace, while broader price pressures eased.
Details in the 🧵
Let's look at Goods Less Food And Energy, what I was referring to as "Core Goods."
Prices for core goods were up 0.32% in Aug, which is well above trend despite stepping down slightly from July.
Looking specifically at Finished Core Consumer Goods and Private Capital Equipment, it's clearer that there is still upward pressure on goods inflation.
Honing in on finished core consumer goods, we can see still elevated Durable Goods inflation and a jump in Nondurable Goods.
Here's the same two measures, but each on a monthly basis rather than a monthly contribution basis.
Again, it's clear both are trending the wrong direction through all the noise.
So, why was headline PPI down in Aug despite this ramp up in Core Consumer Goods prices?
Well, they account for less than 6% of the total PPI index.
Services on the other hand account for over 2/3 of the PPI index, and prices declined -0.24% in Aug for Services.
That big negative orange bar above is Final Demand Trade Services, which is a proxy for trade margins.
After a few big months of rising trade margins, they reversed sharply in August.
Why did margins get squeezed in Aug?
It could be that consumers finally balked at margin padding of recent months or maybe end of summer discounting is eating into margins...
Looking ahead, here are the key categories that flow into the PCE inflation report later this month.
While most were relatively benign, Airline Passenger Fees surged 1% in Aug, reversing some recent weakness, and Portfolio Management Fees also remained elevated at 2%
Altogether, this morning's report is likely to keep the Fed worried about tariff pass-through effects, given still hot core consumer goods inflation despite the headline miss.
On the other hand, broader price pressures do not look overly concerning.
On net, it is unlikely this week's data releases will sway the Fed from restarting the easing cycle next week.
• • •
Missing some Tweet in this thread? You can try to
force a refresh
Here we go again with the BLS Birth-Death model hot takes…
A lot of analysis circulating today claiming the birth-death model inflated January's 130k job gain.
Some are even subtracting the B-D contribution from the headline to argue the "real" number was negative.
The underlying logic is flawed – but there's a kernel of truth worth examining. 🧵
First, the fundamental error: you cannot subtract the not seasonally adjusted (NSA) birth-death adjustment from the seasonally adjusted (SA) headline payroll number.
They're in different statistical frameworks. The B-D is applied to *not seasonally adjusted* data. The headline is *seasonally adjusted* - mixing them is arithmetically meaningless.
That said, there's a fair observation buried in the bad math.
The Jan ‘26 total nonfarm B-D adjustment of -61k was less negative than the typical January reading, hence the chart at the top, which has ranged from roughly -100k to -250k over the prior decade. So yes, the B-D model was a relative tailwind compared to history.
The BLS just released the *final* CES benchmark revisions with this morning's jobs report.
The 2025 labor market was dramatically weaker than originally reported... and for 12 straight months, job growth was never statistically different from zero.
A deep dive with charts in the 🧵
The final benchmark revision for March 2025: -898k (SA), or -0.6%.
The NSA revision: -861k, or -0.5%.
The -861k revision is more than 8x the prior decade's average (-103k).
The preliminary estimate from September was -911k.
The final came in slightly less severe, but the story is unchanged.
The March benchmark is just the anchor. What matters is what it does to the full year of data.
Previously published 2025 total nonfarm job growth: +584k (avg +49k/mo)
Revised: +181k (avg +15k/mo).
That's not a rounding error. That's a fundamentally different labor market.
Secondary mortgage spreads are down a massive 20bps+ on the week, with most of that move happening this morning.
What's happening? The mortgage market is undergoing a structural repricing following the directive for Fannie & Freddie to deploy $200B into MBS.
Let's dig in a bit in the 🧵
The $200B figure is massive relative to GSE capital. For context, Fannie and Freddie have a combined accumulated net worth of ~$173B as of late 2025.
By directing a $200B build-up, the administration is effectively deploying the entirety of the capital cushion built up over the last six years to support the secondary market.
Look again at the Current Coupon MBS OAS.
We are seeing a "forced" mean-reversion of spreads toward 2018–2019 levels.
By acting as a price-insensitive buyer, the GSEs will be bypassing the Fed and targeting mortgage spreads directly.
If the spread stays compressed, mortgage rates can fall even if 10y UST yields remain elevated.
Pretty notable upward revisions to 2Q25 real GDP growth this morning: up 50bps to 3.8% from 3.3% prior.
All of the upward revision was due to stronger than previously reported Consumer spending on of Services, Business Investment and a modest bump to Government consumption.
More details in the 🧵
Looking at the revised series, here's contributions to real GDP growth over time, showing all of the slowdown occurred in Q1, with exports and inventories driving most of the swing.
Stripping out those two volatile categories (i.e. inventories & net exports) gives you a good view on the core underlying drivers of the economy: Final Sales to Private Domestic Purchasers.
Again, now we see there was a modest dip in Q1, with Q2 back to the prior trend pace.
After spiking to a new post-pandemic high due to fraudulent claims in Texas, initial jobless claims have plunged back below their 3y average (222k) to 218k.
Meanwhile, continuing claims also ticked down (1,926k vs 1,928k the week before) but remained near post-pandemic highs.
Some details in the 🧵
After one of the highest weekly increases in jobless claims in a while, we now have had outsized back-to-back weekly declines...
What happened?
As mentioned at the top, a massive surge in fraudulent claims in TX is fading after spiking a few weeks ago.
Cumulatively, there have been nearly 40k excess claims above normal for this time of year in TX over the past 3 weeks.