Core inflation was above expectations and had its highest print since January.
But I don't think that really conveys how much more elevated it is in 2025, as shelter disinflation continues to cover up higher services and goods.
This is worrying. Let's dig in: /1
Here's core broken down by goods, shelter, and services. Goods prices are increasing from a negative trend, as expected from tariffs.
But non-housing services have picked up too, masked by the expected measured housing disinflation. /2
If we dig into non-housing services, 'transportation services' is a driver. Much of that is airline prices rebounding after price drops earlier this year.
But some is from 'motor vehicle maintenance and repair', showing the line between tariffs and services is complicated. /3
Energy slowed in July, bringing down headline vs core, but it's been increasing fast in 2025. Here's the price index for electricity. It took two years, 2023-2024, for it to increase 5%. It's increased ~5% so far this year.
Bad time to take a lot of green energy offline imho. /4
But inflation is broadening. Here's a graphic I had honestly retired but am bringing back into circulation, the percentage of 82 CPI items having at least 3% price growth (annualized) in a month. The evolution of this was a big deal in 2023.
It's back to over 50% recently. /5
Meanwhile, tariffs are doing tariff things, with preregistered categories seeing prices increase, especially from trend. (Apparel is a notable exception.)
We do have to balance these price increase against the <checks notes> 37,000 jobs lost in manufacturing since April. /6
Now I preregistered those categories above, but here are the 12 highest price increases for goods against trendlines.
I assume working families being hit with higher prices alongside more insecure health care from Medicaid and ACA cuts aren't going to sweat some data-mining. /7
Overall worse than I expected, with the background of government disinvestment and deportations weakening potential growth.
Trump was handed a stable economy. I expected them to screw up either the labor market, prices, or growth. I didn't expect them to fumble all three. 8/8
• • •
Missing some Tweet in this thread? You can try to
force a refresh
Job revisions have a well-known strong cyclical component; as the economy slows it is more difficult to estimate business births and deaths.
Here's average revisions as % employment for 18 months before and after a recession, from 1979 to 2025. It's negative in the slowdown. /1
Or consider the last decade. Essentially zero pre-covid.
Sharply negative under lockdowns. But very positive when the economy was gaining millions of jobs back in 2021, as the models couldn't keep up with recovery. But then negative as that recovery leveled out in 2023-2024. /2
Here you can see that with a longer-timeframe. This is very common.
The question right now, as it was in 2023, is whether the revisions indicate slowing into a lower stable steady-state (as it did then), or an actual downward freefall. 3/4
This is a bad jobs report. 73,000 jobs would have been worrisome to begin with, but deeply negative revisions to the previous two-months wiped out much of the recent gains. 2025 looks a lot worse the further we get into it.
There's a lot to cover, let's dig in. /1
First: revisions. Negative revisions naturally occur at the end of recoveries/economic turning points from complexities of estimating business births, deaths, and seasonal adjusts.
While it was negative in early 2024 alongside strong jobs growth, it is collapsing now in 2025. /2
Middle column here is just May and June incorporating the revision, where private education and health services account for 170% of all private sector job growth.
Manufacturing jobs are being lost at a comparable clip to Federal jobs, -11/-12 in July and -13/-18 previously. /3
The GDP numbers for the 2nd quarter, like in the first, are being driven by wild swings in exports and inventories.
While GDP rebounds for a still weak 1.22% in the first half of 2025, under the hood final sales to private domestic purchasers looks quite weak. Let's dig in. /1
You can see this in the components of GDP, where investment (driven by inventories) and net exports are doing wild, probably historically unique, things.
But look at the last two consumptions prints. People are spending less, and it's a notable drag. /2
With 6 months we can compare 1st half of 2025 to the 2nd half of 2024 and note that, across the board, GDP and labor market, 2025 is doing worse.
In 2024 everything looked great except for the unemployment rate; in 2025 everything looks worse except for the unemployment rate. /3
Solid jobs numbers today. Headline unemployment unchanged, and 139,000 new jobs.
But a few things I'd flag for underlying cooling: (1) Last two months revised down -95K. 2-month monthly revisions have averaged -63K in 2025. So reasonable chance this ends up under 100K. /1
The rounding gods were kind this month, with unemployment ticking up to 4.24 percent. You can see the slow increase over the past few months.
At that slow but steady pace you are at ~4.7 percent unemployment at the end of year, consistent with some forecasts on tariff impact. /2
More, the composition gods were very kind this month, with unemployment falling because people aren't leaving their jobs, which helped offset the new entrants who can't find work.
The last 2 recessions were crazy; but weakness leaving unemployment drives smaller recessions. /3
As we risk stagflation and chaos to bring back manufacturing jobs, I took a look at how manufacturing workers reacted to the hot labor market of 2022-2023.
A result that surprised me: it turns out they were the industry with the highest increases in their quits rate. 1/4
Link here, which builds off the recent manufacturing jobs polling and then NEC director Gary Cohn's 2018 (incorrect) dive into JOLTS data.
In terms of absolute increase in quits rate from an environment with some income security and plentiful job openings, manufacturing was only matched by low-wage work in leisure and hospitality. 3/4
Quits and hires rate are now a notable step below 2019 levels.
Quits does better predicting labor market conditions and there's an increasing trend in job openings across the 21C. So, at this critical moment, worth weighing quits more when we are watching labor market. /1
If the concern is in (wage-inertial?) services, hires and quits levels are also below there, and have been falling consistently.
There has been a lot of shifting and upgrading in the labor market during the recovery, and that process has played itself out. Time to reassess. /2
If you are still focused on job openings because you think 'openings over vacancies' is a proxy for output deviations that gives a nonlinear Phillips Curve that fits this recovery, well, sorry bub, that's over. It fell off months ago.