Listen to the audio version of this article (generated by AI).
Memory stocks get nailed… Apple forced to raise prices… the memory trade today and who’s winning it… your MU earnings checklist for tomorrow… last chance for tomorrow’s “Time Machine”
As I write on Tuesday morning, memory stocks are getting slammed.
Micron (MU), SanDisk (SNDK), and South Korean AI-memory giant SK Hynix are all down double-digits as investors unwind one of the year’s hottest (and most crowded) trades ahead of Micron’s earnings tomorrow.
At first glance, investors appear to be sending a very different message than the one Apple CEO Tim Cook delivered last week.
Cook warned that memory prices have become so extreme that Apple can no longer absorb the costs.
So, who’s right?
The market? Or Tim Cook?
Well, let’s back up to better understand what Apple is telling us about one of the most important supply shortages in the AI economy.
Tim Cook calls it a “hundred-year flood”
For months, Apple CEO Tim Cook has been trying to hold the line on memory costs — absorbing supplier price increases rather than passing them to customers.
That’s over.
Last week, Cook told The Wall Street Journal that price increases are “unavoidable” due to the ongoing memory shortage:
We’re doing our best to mitigate the huge increases that are being passed to us, and we’ve been trying to shield our customers from the increases, but the situation has become unsustainable.
At the heart of this is one problem: memory – the chips that power virtually every computing device you own.
Here’s the WSJ with a good way to think about it:
Memory, also called DRAM, and storage, also called NAND, are like elements of a mid-20th-century office: The memory is a desk that holds all the papers a worker needs to perform a task, while storage is the filing cabinet that holds everything else.
Your iPhone uses both. So does your laptop, your car, and your doctor’s medical equipment.
Source: Bagus Hernawan
So, what’s the problem?
Well, AI is eating all of it.
Skyrocketing memory prices with no relief in sight
AI servers are memory-hungry in a way ordinary computing never was – a single AI server requires roughly eight to 10 times the memory of a traditional one.
This has resulted in enormous demand – and surging prices…
Since last year, when Google (GOOG), Microsoft (MSFT), Meta (META), and Amazon (AMZN) began announcing massive increases in their AI capital spending, prices for memory and storage chips have both quadrupled.
The three companies that dominate DRAM production – Micron, SK HynixandSamsung – have responded by redirecting their fabrication capacity toward the specialized high-bandwidth memory AI customers demand. That leaves less supply for everyone else.
Cook put it plainly:
There’s less supply at a time when consumers want devices and the memory guys are passing along huge price increases.
We definitely need memory pricing and supply to return to reasonable levels for consumer products. That’s the bottom line.
Apple is one of the largest buyers of memory on the planet, spending in the low tens of billions of dollars per year. Historically, it has used that heft to wring the lowest prices out of suppliers, playing them against each other.
But with the supply/demand equilibrium radically shifted today, that’s changed. And even Apple is feeling the effects.
Back to Cook:
This is a hundred-year flood.
I’ve never seen anything like it in any area in over 40 years.
When the world’s most powerful hardware procurement machine can’t negotiate around a commodity price shock, it tells you something definitive about the demand dynamic underneath.
Which brings us to the investment implications…
The stocks that are winning thanks to the memory shortage
If memory stocks are in such demand, why are they tanking today?
Part of the answer is simple: these stocks have already had enormous runs so far this year.
Micron entered today up nearly 300% in 2026. SanDisk had surged more than 700%. When investors are sitting on gains like that, it doesn’t take much to trigger profit-taking.
Adding to the pressure, Micron reports earnings after the close tomorrow. Traders often lock in gains ahead of major reports, especially after a stock has already enjoyed a huge run.
But if you’re worried that today’s selloff signals a top, let’s shift our focus back to the supply/demand balance and pricing.
Here’s our technology expert Luke Lango. From his Daily Notes in last week:
The pricing power of Micron (MU), Lam Research (LRCX), KLA (KLAC), ASML (ASML), Applied Materials (AMAT), SanDisk (SNDK), Seagate (STX) and Western Digital (WDC) is not a temporary cyclical phenomenon.
It is a structural reflection of demand running faster than supply can respond — the multi-year undersupply thesis confirming itself in real time.
Luke’s read is that the thesis still has runway: all three major memory producers have pre-sold their entire 2026 HBM (High-Bandwidth Memory) production under long-term contracts, with order visibility stretching deep into 2027.
And Morgan Stanley forecasts DRAM wafer capacity growing 30% by 2027 – yet even with that expansion, wafers for consumer tech will still fall up to 15% short of demand.
Still, be aware that memory is a historically cyclical industry, and when oversupply eventually arrives, the ensuing fallout could be brutal if history repeats.
But the numbers suggest we’re not there yet – even if it feels like it today.
To illustrate, Luke points to research shops Wedbush and Stifel, which both just raised their price targets on Micron, citing “both firms citing durable AI demand extending the memory upcycle well into 2027 and beyond.”
Here’s Luke’s bottom line:
The cumulative pattern across all of this sell-side activity is the estimate revision cycle we have been describing all week: actual AI infrastructure demand consistently exceeds consensus models, analysts revise upward, and stocks re-rate accordingly.
For Luke’s current positioning in across the memory ecosystem – including which names he’d buy at current levels versus which have gotten ahead of themselves – .
With this in mind, one thing to watch tomorrow
There’s a framework that we often refer to in the Digest – legendary investor Louis Navellier’s Iron Law of the Stock 蜜桃传媒.
In short, stock prices can diverge from earnings trends for a while, but over the long run, if a company keeps growing its earnings, its share price follows.
That framework is especially relevant for the memory trade – which brings us to tomorrow…
Micron reports third-quarter earnings after the close. Here’s what Louis, editor of – whose subscribers are up 162% in MU – expects heading in:
Third-quarter earnings are forecast to surge 930.9% year-over-year. Revenue is expected to jump 270.6% year-over-year.
The analyst community has revised third-quarter earnings estimates a whopping 73.8% higher in the past three months.
As to this morning’s selloff, Louis put a lot of the blame on SPCX, which continues to pull back. But then he pivoted to the memory trade and Micron, saying:
The results will be ridiculous, and it’ll be the strongest stock for sales and earnings, and we’ll see how it reacts…
So, hang in there, everybody. Let’s just celebrate Micron’s earnings on Wednesday. We’ll see what the aftermath of that is, what the guidance is going to be.
Earnings – both Micron’s and the broad sector’s – are what we need to watch to monitor where we are in the memory cycle.
As long as they’re growing and topping estimates, there’s more life in the trade.
When earnings begin disappointing, and management begins guiding lower, that’s when Louis’ Iron Law will cut against us, and caution becomes critical.
We don’t expect that to happen tomorrow with MU. Still, this is the issue to watch.
Specifically, watch for three details:
- whether guidance on HBM demand remains strong…
- whether analysts continue raising forward estimates in the days that follow, and
- whether management’s commentary on order visibility into 2027 holds firm.
Broadly speaking, if these signals stay intact, the memory trade stays intact – despite the huge run so far.
To learn more about joining Louis in Growth Investor to help navigate the memory trade, .
One more thing – and it’s time-sensitive
Everything we’ve covered points to the same underlying challenge.
The memory trade is real. The structural thesis is sound. But the stocks that made early investors wealthy, like MU and SNDK, have already made huge moves.
The readers who benefited most weren’t the ones who chased the headlines. They were the ones who saw the bottleneck before Tim Cook was forced to call it a hundred-year flood on the front page of The Wall Street Journal.
The next trade will work the same way – it always does.
This is why tomorrow’s event from Marc Chaikin and his colleague Joe Austin is worth your time.
If you’ve been reading the Digest over the last week, you already know the setup. Marc spent 60 years on Wall Street building analytical tools – including the Power Gauge, a 20-factor stock rating system that helped him navigate the COVID crash, the 2022 bear market and last year’s tariff selloff. Joe spent four decades as a portfolio manager overseeing more than $10 billion in assets.
Together, they’ve built what they’re calling the most powerful tool of Marc’s career – an AI-powered platform that scans decades of market history to find stocks whose fundamental and technical fingerprints match the early profiles of proven multi-baggers, before Wall Street catches on. They’re calling it the .
Tomorrow – Wednesday, June 24 at 10:00 a.m. – they’re unveiling it publicly for the first time, .
This is the last call here in the Digest. If you want to be in the room when they show which stocks the Time Machine is flagging right now, .
Have a good evening,
Jeff Remsburg
(Disclaimer: I own AAPL, MU, ASML, GOOGL, MSFT, AMZN)
