The same argument, at your own pace
Charts, the contract terms, the buyback arithmetic, and the quarter-by-quarter build from today's price to the target. Every number carries its source in the footnote.
Equity research · NASDAQ: MU
Written August 2026 · MU at $926
Micron into 2027.
The market is pricing a bust. The contracts say earnings keep climbing instead, with the downside protected. The share price at least 50% higher within a year, and possibly within six months.
The thesis
The bear case
Three kinds of memory
What moves earnings
The price floor
The multiple
Micron and NVIDIA
If the case is wrong
Close
Matt Hargrove
Financial and valuation analyst
Working with me ↓ Matt.Hargrove5@gmail.com 214-592-5326 LinkedIn ↗
Working with me ↓ Matt.Hargrove5@gmail.com 214-592-5326 LinkedIn ↗
Where this differs
A cycle peak, or a floor that was never there before?
Memory has always been a boom and bust business, and the multiple on the shares says the market expects the bust on schedule. My case is that three things changed at once: demand that does not flinch at price, supply that is physically fixed for two more years, and sixteen contracts that put a floor under half of revenue through mid‑2031.
The slides lay out the bull case, the bear case and the assumptions joining them. The research shows every calculation underneath.
Start with the bear case ↗What would make this wrong
-
01Supply catches up sooner. New capacity arrives faster than the three years it has always taken, or AI demand for memory slows.
-
02The floor is softer than it reads. Contract terms, the share of revenue covered, or customers not performing on what they signed.
-
03The cash never reaches shareholders. Spending on new plants, or buyback decisions, differ from what the model assumes.