The memory chip giant soared 9x, but 3 past crashes reveal what could go wrong.
Wall Street loves a comeback story, and right now, Micron Technology (MU) is writing one of the most dramatic ones in recent memory. The chipmaker’s stock has climbed nearly 9x over the past 12 months, pushing its market value above $800 billion. For a company that once watched its shares fall into the single digits, that kind of turnaround feels almost impossible. But the most important question is the one investors are quietly asking while the gains keep rolling in: have we been here before?
The answer, unfortunately, is yes. Three times in recent history, Micron flew high before hitting the ground hard. Understanding each of those crashes matters more right now than the headlines do.
The 3 times Micron’s stock crashed after a major rally
1. The 2022 to 2023 crash was the most brutal in financial terms. Post-pandemic demand evaporated almost overnight, and supplier inventories ballooned to 31 weeks by early 2023. Micron posted its largest-ever quarterly net loss of $2.31 billion, cut 10% of its workforce, and slashed its capital budget. The stock fell roughly 50% from its 2022 peak.
2. The 2018 to 2019 inventory collapse followed a period of heavy over-ordering by cloud operators. As inventories grew through 2018, NAND prices dropped roughly 60% and DRAM fell about 40%. Micron peaked near $64 in May 2018 before falling to around $28 by year-end, a decline of approximately 57%.
3. The 2014 to 2016 DRAM downturn unfolded as manufacturers built capacity for a PC market that no longer existed. Consumers had already migrated to mobile. Micron’s stock fell roughly 70%, sliding from about $37 in 2014 to under $10 by early 2016.
Why some analysts think this time is actually different
Investors who remain bullish on Micron point to three structural factors they argue separate this AI-driven cycle from everything that came before it.
1. Demand is no longer growing linearly. Nvidia’s H100 chip used 80 gigabytes of high-bandwidth memory (HBM). Its upcoming Rubin Ultra targets 512 gigabytes per GPU module. As AI shifts from training large models to running them for everyday users in real time, the need for ultra-low-latency memory has become non-negotiable across the industry.
2. The contract structure has fundamentally changed. In March, Micron signed what the industry described as its first-ever five-year HBM supply agreement, locking in both volume and pricing. That level of visibility is something the memory business has never seen before, and it directly addresses the abrupt order cancellations that historically sent prices into freefall.
3. Manufacturing complexity is slowing rivals down. HBM requires significantly more wafer capacity per bit than standard DRAM, and its production complexity limits how quickly competitors can scale. Micron’s global HBM revenue share climbed from 9% in Q4 2024 to 21% in Q4 2025, even as the overall HBM market roughly doubled.
The risk that has not gone away
Even with those tailwinds, the oversupply threat is not off the table. Micron has guided fiscal 2026 capital spending above $25 billion. SK Hynix is expected to spend roughly $27 billion. Samsung is also pushing aggressively into the same space. History shows that when all three major memory producers scale capacity at the same time, oversupply tends to follow within two to three years.
D.A. Davidson maintained its Buy rating on May 11 with a $1,000 price target. The firm argued that the market is failing to grasp the new economics of memory in the AI era. Micron has pre-sold its entire HBM production through 2026 under binding contracts, and hyperscalers including Microsoft, Alphabet, and Meta are projected to spend a combined $700 billion on AI infrastructure this year alone.
But if those same hyperscalers face pressure to show better returns on their enormous AI investments, spending could slow. And when AI spending slows, memory demand tends to follow.
Disclaimer: This article is for informational purposes only and does not constitute financial advice. The author and publication are not registered investment advisors and do not provide personalized investment recommendations.