Is Micron Stock Too Cheap to Ignore?

Micron expects HBM total addressable market to grow from $25 billion in 2025 to $100 billion by 2028, but says it can currently meet only half to two-thirds of customer demand in the medium term. The article argues the stock remains inexpensive at 8.3x forward earnings, though it warns memory chips are cyclical and profits could fall once supply normalizes. Overall, the piece is constructive on Micron's long-term demand outlook but cautious on timing and cyclicality.

Analysis

The market is treating MU like a classic cyclical and, importantly, that discount is rational only if supply normalizes faster than demand can compound. The underappreciated second-order effect is that HBM scarcity does not just lift Micron’s own pricing power; it also becomes a gating factor for AI server deployment, which can slow shipment growth for accelerators and network gear even when end-demand remains strong. That creates a more durable earnings tailwind for the memory suppliers than for the broader AI hardware complex if capacity remains constrained through 2026.

The key risk is not demand exhaustion, but capex overbuild. If the current margin spike attracts too much industry investment, the inflection point will show up first in contract pricing and lead times, then in utilization, with a 6-12 month lag before the P&L rolls over. Investors anchoring on the current multiple should be careful: 8x forward earnings can still be expensive if peak earnings are being capitalized, especially in a business where earnings can mean-revert sharply once inventory normalizes.

The contrarian read is that the market may be underpricing the duration of the shortage because HBM is not a generic memory cycle anymore; qualification, packaging, and yield learning curves create a much slower supply response than legacy DRAM/NAND. That argues for a longer runway of elevated returns than most cyclical investors expect, but only if AI capex remains concentrated in a handful of hyperscalers that continue to absorb constrained supply. If spending broadens into weaker buyers or customer inventories rise, the thesis breaks quickly.

From a tape perspective, MU is likely to outperform on any earnings or guide raise, but the cleaner trade may be relative value rather than outright long exposure. NVDA and INTC both benefit from the same AI spend, yet neither has the same direct scarcity leverage to HBM pricing, so MU should retain a valuation premium as long as supply stays tight. The main catalyst to fade is any sign of new capacity ramping earlier than expected or customer commentary shifting from shortage to inventory digestion.

Key Decisions for Investors

• Long MU into earnings/guidance updates over the next 1-2 quarters; use 5-10% trailing stop because the stock is vulnerable to any sign of pricing inflection before capacity catches up.
• Pair trade: long MU / short INTC for 3-6 months. MU has the cleaner exposure to AI memory scarcity; INTC is more exposed to an execution-heavy turnaround with less direct pricing leverage. Target relative outperformance if HBM tightness persists.
• Buy MU call spreads 6-9 months out instead of stock if seeking convexity on a continued shortage. This captures upside from multiple expansion while limiting drawdown if the cycle turns earlier than expected.
• Take profits on MU into any sharp post-earnings gap-up unless management explicitly raises medium-term supply constraints again; the main risk is the market shifting from scarcity pricing to peak-cycle pricing.
• Monitor NVDA and major hyperscaler capex commentary as a leading indicator. If AI accelerator shipments decelerate while MU lead times remain tight, reduce exposure because demand destruction can emerge faster than the equity market expects.