Memory chips have always been the most boom-and-bust corner of the semiconductor industry, and Micron's own chart proves it, swinging from under $10 to well over $1,000 across cycles. The question now is whether the AI-driven memory shortage is different enough to justify today's price.
The table below shows our year-end price scenarios from 2027 through 2035, measured against MU's recent price of $848.34. Memory stocks swing harder than most sectors, so treat the bear and bull columns as genuine tail scenarios rather than close calls.
| Year-End | Bear Case | Base Case | Bull Case | Base vs. Today | Cycle Stage Assumed |
|---|---|---|---|---|---|
| 2027 | $420 | $1,050 | $1,500 | +24% | HBM supply still sold out through 2027 |
| 2028 | $300 | $1,150 | $1,750 | +36% | First real signs of oversupply risk |
| 2029 | $280 | $1,000 | $1,850 | +18% | A standard memory down-cycle begins |
| 2030 | $350 | $1,100 | $2,000 | +30% | Recovery into the next up-cycle |
| 2031 | $400 | $1,250 | $2,150 | +47% | HBM4E and next-gen AI memory ramp |
| 2032 | $380 | $1,150 | $2,250 | +36% | A second cyclical trough |
| 2033 | $450 | $1,350 | $2,400 | +59% | Structural AI memory demand floor rises |
| 2034 | $470 | $1,450 | $2,550 | +71% | Share buybacks reduce float meaningfully |
| 2035 | $500 | $1,550 | $2,700 | +83% | Micron trades as a mature AI infrastructure name |
What Just Happened: A Historic Fiscal 2026
Micron's fiscal year ended August 2026, and the numbers are almost hard to believe next to its own history. Full-year revenue reached $133.19 billion, up from $37.38 billion the prior year, a 3.5-times jump driven almost entirely by memory used in AI servers. GAAP net income was $84.97 billion, and operating cash flow came to $89.68 billion.
The fourth quarter alone, reported September 30, delivered revenue of $54.23 billion, up 31% sequentially, with a non-GAAP gross margin of 87% and adjusted earnings per share of $33.42. DRAM revenue hit a record $39.8 billion for the quarter, up 343% year over year, and now makes up 73% of total sales.
For fiscal Q1 2027, Micron guided to roughly $61.5 billion in revenue and $38.15 in adjusted EPS, both ahead of the $57 billion and $35.40 analysts had expected. CEO Sanjay Mehrotra described AI as evolving into what he called "Super Intelligence," with memory becoming central to how competitive AI platforms actually perform.
Why HBM Changed Micron's Business Model
High-bandwidth memory, or HBM, is the product line behind this shift. HBM stacks multiple DRAM dies together so GPUs from Nvidia and AMD can feed data in and out fast enough to keep pace with AI training and inference workloads, and Micron is the only major US-based maker of it.
Management says the company has already locked in agreements covering the vast majority of its calendar 2027 HBM bit supply, at meaningfully higher prices than the prior year. That pricing power is narrowing the margin gap between HBM and ordinary DRAM, and industry-wide HBM demand is expected to keep growing faster than conventional DRAM through calendar 2028.
To meet that demand, Micron is deploying serious capital. The company broke ground on an HBM advanced packaging facility in Singapore, is expanding DRAM capacity in Japan and Taiwan, and is building a roughly $100 billion megafab in New York. Full-year capital expenditure guidance was recently raised to $60 billion to $64 billion, up from a prior ceiling of $56 billion.
Is Micron a Good Investment, or Is This the Top of a Cycle?
The bull case and the bear case for Micron both start from the same fact: memory is a commodity business where price is set by supply and demand, not brand loyalty. Which side wins depends on how these three pillars hold up.
- Pricing power looks genuinely different this cycle. Past memory booms were driven mostly by PC and phone demand, which is elastic and easy to overbuild into. HBM is sold directly against specific AI accelerator roadmaps from Nvidia and AMD, with multi-year agreements that give Micron visibility it rarely had before.
- Margins have expanded to levels memory stocks have never sustained before. An 87% non-GAAP gross margin would be remarkable for a software company, let alone a chipmaker that historically ran gross margins in the 30s and 40s during downturns. The open question is how much of that holds once supply catches up.
- Chinese competition is the wildcard. ChangXin Memory Technologies, a Chinese DRAM maker, is reportedly preparing an IPO worth roughly $8.55 billion, and its emergence as a credible volume producer is exactly the kind of supply shock that has ended past memory up-cycles early.
The balance sheet gives Micron room to weather whatever comes next. The company ended fiscal 2026 with $73.48 billion in cash, marketable investments and restricted cash, and adjusted free cash flow of $62.31 billion for the year. Its board raised the buyback authorization to $35.16 billion in October 2026, and a small dividend remains in place with a negligible yield.
MU Stock Price Prediction 2027: Riding Out the Peak of the Boom
For 2027, we assume Micron's locked-in HBM pricing keeps gross margins elevated near 80% to 85% through most of the year, with revenue growing to roughly $170 billion as data center demand remains the dominant driver. That would put full-year non-GAAP earnings per share somewhere near $90 to $95.
Applying a price-to-earnings ratio of about 11, reflecting the market's historical habit of discounting cyclical peak earnings heavily, produces a base target near $1,050, about 24% above today's price. The $1,500 bull case assumes the AI memory shortage proves more durable than the market expects and multiples expand instead of compress. The $420 bear case assumes early signs of oversupply, whether from Chinese competitors or a slowdown in AI capital spending, trigger the kind of sharp de-rating memory stocks have experienced in every prior cycle.
MU Stock Price Prediction 2028: Watching for the First Cracks
By 2028, industry HBM bit shipment growth is expected to still outpace conventional DRAM, per Micron's own guidance, but this is also the point in past cycles where new capacity from competitors typically starts catching up with demand. Our base case assumes revenue growth slows meaningfully, with earnings per share peaking somewhere in the $95 to $100 range before pricing pressure begins.
At a similar 11 to 12 times multiple, the base target for 2028 is $1,150, about 36% above today's price. The bull case of $1,750 assumes capacity constraints persist longer than expected. The $300 bear case models a genuine price collapse, the kind Micron investors have seen before, triggered by oversupply arriving faster than demand can absorb it.
MU Stock Price Prediction 2029: The Down-Cycle Scenario
Our base case actually assumes revenue and earnings decline in 2029, not because AI demand disappears, but because memory pricing cycles rarely get eliminated, only delayed and stretched. If HBM oversupply does arrive, even temporarily, gross margins could fall from the high 80s back toward the 50s or 60s, the kind of swing that has happened to Micron multiple times in the past decade.
That produces a base target of $1,000 for year-end 2029, a modest 18% above today's price and actually below our 2028 target, reflecting a genuine cyclical dip rather than a straight upward line. This is the year most likely to test the patience of anyone who bought Micron purely on the strength of the AI narrative.
The $1,850 bull case assumes no meaningful down-cycle materializes this time, which would be unprecedented in Micron's history but not impossible given how structurally different AI demand is from past PC and mobile cycles. The $280 bear case assumes a sharp correction similar to the 2022 to 2023 memory downturn, just arriving a few years later than usual.
MU Stock Price Prediction 2030: A New Up-Cycle Begins
By 2030, we expect the market to have worked through whatever correction occurred in 2028 or 2029, with a new product cycle, likely built around HBM4E or its successor, driving a fresh recovery. Our base case assumes revenue stabilizes near $150 billion to $160 billion with margins recovering into the 70s.
That supports a base target of $1,100, about 30% above today's price. The bull case of $2,000 assumes Micron has, by this point, established durable pricing discipline across the memory industry, something analysts have speculated about for years without it fully materializing. The $350 bear case assumes the down-cycle that started around 2028 or 2029 simply runs longer than historical patterns suggest.
2031 to 2035: Does the Supercycle Ever Fully Mature?
Looking a decade out, the central question for Micron is whether AI memory demand eventually becomes structural enough that the old boom-bust pattern softens, the way cloud computing eventually smoothed out some of the old enterprise IT spending cycles.
Our base path for the back half of the forecast assumes these developments.
- 2031 and 2032 ($1,250 and $1,150): A new product ramp drives a recovery, followed by another, shallower cyclical dip as the industry tests whether capacity discipline can hold.
- 2033 and 2034 ($1,350 and $1,450): AI memory demand establishes a higher structural floor, with buybacks funded by the buyback authorization beginning to meaningfully shrink the share count.
- 2035 ($1,550): Micron trades more like a mature, strategically important AI infrastructure supplier, with cycles still present but shallower than historically.
The $2,700 bull target for 2035 assumes memory effectively becomes a long-term growth market rather than a cyclical one, a genuine shift from a century of industry history. The $500 bear target assumes old patterns reassert themselves fully, leaving Micron as a profitable but unpredictable stock that never quite earns the steady multiple investors hope for.
Bull Case vs. Bear Case: What Could Go Right and What Could Break
These are the clearest paths toward our bull scenarios.
- Bull Catalyst 1, sold-out capacity through 2027: With the vast majority of next year's HBM bit supply already contracted at higher prices, Micron has unusual near-term visibility for a historically unpredictable business.
- Bull Catalyst 2, AI demand simply outrunning new supply: If hyperscaler and AI lab capital spending keeps growing faster than new fabs can come online, pricing power could persist well past when past cycles would have turned.
- Bull Catalyst 3, genuine pricing discipline: A more consolidated industry, with only Micron, Samsung and SK Hynix as serious HBM producers, could behave differently than the fragmented commodity DRAM market of the past.
These are the risks most likely to drag the stock toward our bear scenarios.
- Bear Risk 1, Chinese competition arriving faster than expected: CXMT's planned IPO and expansion plans are exactly the kind of new, well-funded supply that has ended memory up-cycles before, and Chinese capacity has historically ramped faster than Western analysts expect.
- Bear Risk 2, capital expenditure outrunning demand: Micron's own capex guidance has already been raised once this year, to $60 billion to $64 billion. If every major memory maker overbuilds at once, as has happened repeatedly, oversupply can arrive within a single product cycle.
- Bear Risk 3, extreme valuation sensitivity: With a price-to-sales ratio that has approached 15 during the rally, even a modest disappointment in guidance can trigger double-digit single-day drops, as Micron has already shown multiple times in 2026.
The Verdict: Who Should Buy Micron, and Who Should Wait
Micron suits investors who specifically want AI infrastructure exposure through the memory layer rather than through GPUs, and who can tolerate genuinely violent price swings. This is not a hold-and-forget stock, and it has delivered single-day moves above 10% multiple times in 2026 alone, in both directions.
It is poorly suited to anyone who needs stability, has a short time horizon, or would be tempted to sell during a sharp pullback rather than hold through it. Retirees, income investors and anyone new to cyclical stocks should either avoid Micron or size it very small.
For those who do want exposure, we suggest 2% to 4% of liquid assets as a sensible ceiling, with a bias toward buying in tranches after pullbacks rather than chasing strength, given how sharply this stock has moved on both good and bad news throughout 2026.
Frequently Asked Questions
These are the questions we hear most often about Micron's valuation, the memory cycle and its long-term AI positioning.
Sources and Reference Materials
The data in this analysis comes from Micron's SEC filings and earnings releases, cross-checked against reputable financial coverage and market data current as of late 2026.
- SEC EDGAR: Micron Technology Filings (10-Q, 10-K, 8-K)
- Micron Investor Relations: Fourth Quarter and Full Year Fiscal 2026 Results
- CNBC: Micron Beats on Earnings, Issues Strong Guidance as Data Center Revenue Jumps
- Investing.com: Micron Q4 2026 Earnings Call Transcript
- Seeking Alpha: Micron Q4 2026, These Numbers Don't Look Real
- Webull News: Why Micron (MU) Stock Is Down Today (Barchart syndication)
This article is for informational purposes only and is not financial advice. Price targets are CoinDecimal model scenarios based on public data as of October 16, 2026, and actual results may differ materially.
