Dr. Nathan Hayes: Four hundred and seventeen billion dollars.
Sarah Lin: That's — yeah, hi, morning — what are we doing with that number.
Dr. Nathan Hayes: That is what five large AI-related companies lost in market cap on June 25th. The same day Micron reported forty-one-point-four-six billion in quarterly revenue — a three-hundred-and-forty-six-percent year-over-year increase — and its stock surged up to fifteen-point-seven percent. Top S&P 500 performer on the session.
Sarah Lin: And this is the thing that sort of — it's disorienting? Because Micron winning and those AI names losing four hundred billion, those aren't separate stories. Micron is, um — the simplest way I can say it — they're the landlord of a neighborhood everyone needs to live in right now. Rents went up. They won. Apple had to pay the rent hike and announced price increases on MacBooks and iPads the same day. Their stock dropped.
Dr. Nathan Hayes: And the indexes reflect exactly that split. Dow up slightly. S&P flat — essentially unchanged at seventy-three fifty-seven. Nasdaq down nearly half a percent. No clean read.
Sarah Lin: The market cheering and panicking about the exact same supply constraint.
Dr. Nathan Hayes: That's precisely the question worth sitting with.
Dr. Nathan Hayes: Now here's where I want to separate signal from noise. The headline is 'Micron blowout quarter.' What's actually new — and I mean genuinely new, not a bigger version of something we already knew — is that HBM capacity is sold out. Not constrained. Not tight. Sold. Out. Which means Micron isn't competing on price anymore. They're rationing. That's a different kind of market power.
Sarah Lin: Wait, there's a difference there that I want to — what does rationing actually do to how buyers behave?
Dr. Nathan Hayes: It means pricing psychology shifts entirely. And that's what the $50 billion guidance for next quarter is built on — long-term supply agreements, hundreds of billions locked in. Net profit jumped to $28.24 billion. Data-center revenue beat estimates by nearly 70%. These aren't incremental beats. But — and here's my concern — it's one quarter extrapolating from a very specific moment of constraint.
Sarah Lin: Needham raised their price target to $1,550 though. And Citi's whole 'durability versus melt-up' framing — that suggests some analysts do see this holding.
Dr. Nathan Hayes: Right, and that framing is exactly why the SK Hynix number matters so much. SK Hynix also makes HBM. Also had positive demand signals. Dropped 10% in Asian trading the same day Micron surged. SoftBank fell 11 to 13%. The optimism — it didn't travel. The market is rewarding Micron's specific position, not validating the memory trade broadly. That's a meaningful distinction.
Sarah Lin: Hm. So Qualcomm moves up with Micron, SK Hynix drops — that's not sector rotation, that's the market sort of... picking individuals out of a crowd and saying we trust you specifically.
Sarah Lin: And that's — okay, that's the take I want to push on. Because everyone's calling this a market exhale. A relief rally. And I think that framing is doing a lot of work it hasn't earned.
Dr. Nathan Hayes: Agreed. And the cleanest evidence against it is Nvidia. Slid on June 25th. Same session. If this were sector-wide relief, Nvidia doesn't go down.
Sarah Lin: Right. And here's the texture of it — someone opens their brokerage app Tuesday morning, Apple down roughly six percent. Forty inherited shares, something they thought was safe. That's a $2,400 loss before coffee. And their work sends an email the same morning: company MacBook Pros are now four hundred dollars more. The same supply constraint made both of those happen simultaneously.
Dr. Nathan Hayes: Now — and I want to be precise here — Apple's price hike is specifically downstream of HBM constraint. That's not a general AI tax. Conflating those two things misleads people about the actual mechanism.
Sarah Lin: Mm. But does the person feeling it care about the mechanism?
Dr. Nathan Hayes: No — but we should, because there's also fresh May inflation data showing prices rising again, Federal Reserve rate anxiety still live in the background. So you have HBM constraint, macro pressure, and $417 billion in AI-complex cap gone the same day Micron soared. That is not an exhale. That is coexisting euphoria and doubt.
Sarah Lin: And Needham's $1,550 target, Citi's whole framework — even the bullish reads are saying Micron is priced for perfection. Which means... one stumble. That's it.
Dr. Nathan Hayes: And priced for perfection is — right, that's the real exposure. Because Micron's guidance projecting through 2028 assumes constraint holds. If one hyperscaler decides the memory cost doesn't pencil against the ROI they need to justify their own valuation, that whole architecture starts to — I mean, it doesn't collapse, but it shifts. Fast.
Sarah Lin: Hm. And the Federal Reserve piece is still just... sitting there underneath all of it.
Dr. Nathan Hayes: It is. Rate anxiety doesn't disappear because Micron had a good quarter.
Sarah Lin: Sarah Lin: So the question is whether the Nasdaq being down nearly half a percent while the Dow ticked up is, um... a one-day anomaly, or whether it's the market starting to say something about who actually absorbs the cost when memory doesn't get cheaper.
Dr. Nathan Hayes: That's it exactly. And we have one quarter of data from which to answer it.
Dr. Nathan Hayes: If HBM constraint eases — and at some point supply responds to pricing, that's not conjecture, that's how markets work — does Micron's rationing power hold, or does the whole rent-extraction model unwind before the hyperscalers ever show the returns they promised?
