DRAM's Crunch Has No Quick Fix: Why Micron, Samsung and SK Hynix Keep Pricing Power Into 2027
The Wall Street Journal headline frames the memory shortage as a problem to be solved. It isn’t. The more accurate reading of the supply picture is that the crunch is the predictable output of a fixed production base being reallocated toward AI, and there is no near-term lever — industrial or political — that changes that math before 2027. For the three companies that own the supply, that is not a crisis. It is the most durable pricing-power setup the industry has seen in a generation.
The supply side is fixed, not slow
Three firms — Samsung, SK Hynix and Micron — control the DRAM market, and a comparable handful make NAND. New memory fabs take years and tens of billions of dollars to bring online, so the supply curve over the next several quarters is effectively vertical. When demand surges, the only thing that moves is price.
What changed is not total demand but its composition. The hyperscalers building AI infrastructure want high-bandwidth memory, and HBM consumes far more wafer area per bit than commodity DRAM. Every wafer redirected to HBM is a wafer that leaves the consumer pool. The result is a reallocation: the same factories, pointed at a higher-value customer, leaving phones, laptops, consoles and cars to bid for what’s left. Morgan Stanley estimates consumer memory wafers will fall roughly 15% short of demand by 2027. That gap is not a manufacturing failure. It is a deliberate allocation choice by suppliers who have discovered they can sell every bit they make to the highest bidder.
This is why capacity expansion doesn’t fix the near term. The U.S. has committed tens of billions in grants and tax credits — including Micron’s domestic build-out — but a fab announced today produces revenue years from now. The supercycle plays out entirely inside the window before that capacity arrives.
The one fast valve is locked shut
In a normal cycle, the relief mechanism would be marginal supply from outside the top three. China has that supply in CXMT (DRAM) and YMTC (NAND), and Washington is reportedly weighing whether to ease sanctions on both to take pressure off consumer prices. The article’s central insight is why that valve stays closed: the two Chinese makers are several generations behind the leaders, so the volume relief would be modest, while the cost would be strategic. Loosening export controls risks exactly the spillover of tooling, process know-how and talent the controls were built to prevent — accelerating the one competitor the U.S. semiconductor strategy is designed to slow.
That is the trap. The fast lever (Chinese supply) is blocked by national security, and the slow lever (domestic fabs) can’t act in time. Policymakers are left without a tool that works on the timeline that matters. For an administration sensitive to consumer prices, that’s a political problem with no clean answer — which is precisely what makes it an investment thesis with an unusually clear shape.
The asymmetry is the trade
Strip away the consumer-harm framing and the article describes a clean bifurcation. On one side, the suppliers: pricing power they don’t have to share, demand that absorbs every incremental bit, and a competitive set that can’t be expanded by policy. DRAM and NAND prices have roughly quadrupled since last year, and the pass-through mechanics favor the makers — Micron and its peers set the price, and everyone downstream pays it.
On the other side, the memory consumers, who are now the casualties. Nintendo warned on console margins and has shed more than 30% year to date. Xiaomi is down around 20% in 2026, Canon down roughly 10%, both on memory-cost exposure. Apple is raising prices on its products, with CEO Tim Cook describing the situation as a “hundred-year flood” unlike anything in his four decades in the supply chain. When the company with the most supply-chain leverage on earth says it cannot engineer around the shortage, that is the market telling you the constraint is real and the suppliers hold the cards.
What it means
The investable conclusion is duration. The bear case on memory has always been cyclicality — that today’s shortage seeds tomorrow’s glut as capacity floods in. This cycle is structurally different on the timeline because the glut requires either new fabs (years away) or Chinese supply (politically blocked), and neither arrives inside the window where AI demand is still climbing. One strategist quoted in the coverage floated the crunch persisting in some form as far as 2030.
For the names already in the supply seat — Micron most directly for U.S. investors, with Samsung and SK Hynix as the global complement — the question is no longer whether pricing holds. It’s how long, and the supply math argues it holds longer than a normal cycle would suggest. The single data point that would change this thesis isn’t a demand wobble; it’s a policy reversal on CXMT and YMTC. Watch the sanctions debate, not the spot price. That is where the cycle actually turns.