The GDP Slowdown Nobody's Panicking About (And Why AWS Just Stole the Show)

So this week was genuinely a lot to process, and I mean that in the best possible way — if you're the kind of person who actually enjoys watching a 1.5% GDP print land while Amazon's cloud business is absolutely cooking and South Korean chip stocks are doing things that have never happened before in recorded history. Which — okay, that's a niche audience. But that's why we're here.

Let me walk you through what I think is actually the story of this week, because the headlines are pulling in three different directions at once, and the real picture only makes sense when you put them side by side.

The GDP Number That Should've Scared Everyone More

Thursday's Bureau of Economic Analysis release showed Q2 2026 GDP growth came in at 1.5%. Annualized. That's the number. And by any honest read, it's soft. Economists were expecting something closer to 2%, and we missed it.

But here's what I actually think about this — and I'll admit my first reaction was "oh no." Then I read the breakdown. The miss didn't come from consumers pulling back or businesses going fetal. It came from a decline in federal government spending and a drawdown in inventories. Those are real, but they're also the kind of drag that doesn't necessarily signal the economy is cracking at the foundation.

Consumer spending? Still moving. Business investment in equipment? Holding up. The underlying engine of the economy — the part driven by actual human beings buying stuff and companies building things — wasn't the problem.

That said, I'm not going to sugarcoat 1.5%. That's slow. Slower than 2025's average. And the broader context matters here: the Fed has been running tight policy for a while now, yields are elevated, and credit conditions aren't exactly inviting people to borrow and spend freely. We wrote about exactly that dynamic a few weeks back in The Great Credit Squeeze and honestly, it's still playing out exactly as described.

Going a step further — the 1.5% number by itself isn't the problem. The problem is what's sitting right next to it.

Core Inflation at 3.3%: The Number That's Actually Running the Show

June core inflation came in at 3.3%. That's the Personal Consumption Expenditures measure the Fed actually watches, not the headline CPI most people see. And 3.3% is... not 2%. Not even close to 2%. The Fed's target has been 2% since basically forever, and we've now been above it for over five years straight.

Five years. Let that sink in for a second. A kid who started kindergarten when inflation first broke above target is now reading chapter books.

So you've got GDP at 1.5% — slow — and inflation at 3.3% — still hot. That combination is exactly what three Fed dissenters argued about this week when they publicly pushed back against waiting to raise rates further. Their argument is essentially: every month you wait with inflation above target, you're making it harder to credibly bring it back down. Which is a reasonable concern.

My honest take: they're not wrong, but the political and economic pressure on the Fed right now is enormous. You can't raise rates aggressively into a 1.5% growth environment without taking real risk of tipping into something worse. The Fed is genuinely stuck, and I've said that before in a few different ways. The dissenters are being intellectually honest. Whether their view wins out inside the FOMC chamber is a different question entirely.

Now Here's Where It Gets Interesting

While economists were digesting the GDP miss on Thursday, Amazon was quietly dropping an earnings report that made a lot of investors forget about macroeconomics for approximately 45 minutes.

AWS — Amazon Web Services — outperformed. Not by a little. Wall Street analysts are out with renewed "buy" ratings and raised price targets this morning, with the general sentiment being that Amazon's AI infrastructure bet is playing out exactly the way the bulls have been saying it would. Enterprise spending on cloud and AI workloads isn't slowing down even as the broader economy decelerates. That's a significant data point.

Here's what that means for you if you're trying to figure out where to put money right now: the AI capital expenditure cycle appears to be somewhat divorced from the normal economic cycle. Companies are treating AI infrastructure spending the way they treated internet infrastructure in the late 1990s — as a competitive necessity rather than a discretionary line item. That's either the most exciting thing happening in markets or a setup for a brutal reckoning sometime later. Possibly both.

Apple's the Plot Twist Nobody Fully Knows How to Read

And then there's Apple.

Apple's Q2 earnings landed, and Wall Street is... divided. Which is a generous way of saying nobody quite agrees on what to make of them. The demand side looks okay — people are still buying iPhones and Macs and whatever the Vision Pro has become by now. But the supply side is a mess. Memory prices are climbing, chip supply constraints are real, and Apple's ability to manage its cost structure in this environment is genuinely uncertain.

Look, I could be wrong here, but I think Apple's situation is actually a useful window into the broader AI trade tension. The companies that are direct infrastructure providers — chip designers, cloud computing platforms, data center operators — are doing extremely well. The companies that depend on affordable chips and smooth supply chains to sell consumer hardware are getting squeezed. Apple sits firmly in the second category, which is uncomfortable for a stock that trades at a premium multiple.

The Korean Chip Story That Most People Slept On

Okay, this one genuinely surprised me — and I say that as someone who checks markets before my coffee most mornings.

The Kospi, South Korea's benchmark index, just had its best single day ever. And the driver was Samsung and SK Hynix — the two memory chip titans — posting historically large gains. We're talking about moves that, at the index level, broke records that had stood for decades.

Why does this matter for US investors? Because Samsung and SK Hynix make the DRAM and NAND memory that goes into basically every AI server ever built. When those stocks rip like this, the market is telling you something: demand for AI-adjacent hardware is accelerating, not decelerating. The infrastructure build-out has more legs. And that's consistent with what AWS just reported on the revenue side.

But wait — there's more to this. This is simultaneously the story that makes Apple's situation harder to solve. If memory prices are rising because AI demand is vacuuming up supply, then consumer electronics margins take a hit. iPhone production costs go up. The same boom that's lifting Nvidia's suppliers is quietly squeezing Apple's cost model.

That's the bifurcation that's been developing in tech all year: AI-adjacent wins, AI-adjacent winners are simultaneously strangling the parts of tech that relied on cheap components.

The Broader Market: More Stocks Joining the Party

Here's something that actually gave me a more optimistic read on equity markets this week than I expected. The number of S&P 500 stocks trading above their 200-day moving average just hit its highest level in two years. Which sounds technical and dry, but what it actually means is: this rally has been broadening out.

For most of 2025 and early 2026, you basically had Nvidia, Microsoft, and a couple of other AI names carrying the entire market on their backs. The breadth was narrow and that always makes me nervous — a market held up by seven companies isn't the same as a healthy bull market. But when you see participation widen, when value stocks and industrials and consumer staples start climbing alongside the AI names, that's a more durable foundation.

RBC analysts called this week's volatility a "valuation opportunity" in U.S. stocks. I don't always agree with Wall Street strategists, but on this specific point, I think they're onto something. Pullbacks driven by macro noise — GDP prints, inventory adjustments, some Fed meeting uncertainty — don't typically end bull markets. They reset entry points. The July selling has felt more like a stress test than a structural break.

That said — and this is the part that genuinely worries me — you still have core inflation at 3.3% and a Fed that can't easily cut even if growth keeps sliding. There's a version of the next six months where GDP drifts down to 1% and inflation stays stubbornly at 3%, and that's when things get dicey. We're not there yet. But the door to that scenario isn't closed.

Putting It All Together

So here's the picture as I see it heading into the back half of 2026.

The macro backdrop is genuinely mixed — slower growth than most people wanted, inflation that refuses to cooperate, and a Fed with three dissenters who want to hike while the other members are trying not to tank what's left of the expansion. That's not a comfortable place for policy to be.

But the earnings picture inside that backdrop is more interesting than the GDP headline suggests. Companies with direct exposure to AI infrastructure spending — cloud platforms, memory chip makers, data center suppliers — are putting up numbers that suggest the AI capital cycle is nowhere near done. Companies with indirect exposure — consumer hardware, for instance — are dealing with the cost pressures that come with that same cycle hoovering up component supply.

The trade you've actually been rewarded for this year isn't "buy tech." It's more specific than that. It's "buy the digital picks-and-shovels": the infrastructure layer, the cloud platforms, the companies selling the compute that everyone else desperately needs. AWS just confirmed that thesis for another quarter. Korean chip stocks having their best day ever confirmed it in a completely different part of the world.

And the 1.5% GDP print? Uncomfortable. But not the reason to abandon equities. The reason to be cautious about equities is still the same one it's been all year: inflation that doesn't want to die, and a Fed that's slowly running out of options for threading the needle between "too tight" and "not tight enough."

That's the tension everyone's going to be watching for the rest of the year. And I think it's actually more interesting than people are giving it credit for.