A "Buy" on Paper, a Coin-Flip on Price. It Scores 62/100 After the Biggest Quarter in Memory History
Data as of the Oct 2, 2026 close. MU priced at roughly $1,075 (Oct 1–2 range ~$1,060–$1,095; sources disagree by a few percent).
Section A: The Market Backdrop (run once)
| Item | Reading (date) | What it means |
|---|---|---|
| VIX | 15.4 (Oct 2). 52-wk range not pulled | "Stability" regime (15–20): normal conditions, fair valuations. |
| S&P 500 / Nasdaq Comp. | 7,722.8 / ~27,190 (Oct 2). S&P ~+15% y/y. Nasdaq 100 hit a record intraday | Tech/AI leading; breadth % not pulled, so treat as narrow-ish leadership (estimate). |
| GDP | Q2 2026 real GDP +1.5% annualized (2nd est., Aug 26); Q1 +2.1% | Weak (1–2%). Differentiation matters; AI capex is the exception. |
| Inflation | Aug CPI +3.4% y/y; core PCE 3.3% (July). Fed sees 2% only in 2029 | Too hot. Oil/Iran-war premium is the culprit. |
| Fed | Hiked 25bp to 3.75–4.00% on Sep 16 (first hike since 2023). Dots: one more hike in 2026 | Hiking = higher discount rates, tougher on long-duration stocks. |
| Yields / curve | 10-yr ~5.2% (Oct 2); 2-yr 4.74% (Sep 16). 2s10s roughly +50bp (my estimate, mixed dates) | Borderline flat/normal. A 50bp move in rates swings high-multiple stocks hard; MU's low P/E cushions it. |
| Labor | Sept payrolls +29k (est. 90k); Aug revised to +133k; unemployment 4.2% | Cooling. Cut October hike odds, but not enough to change the Fed's inflation focus. |
| Currency / commodities | Dollar fell Oct 2 (level not pulled); gold ~$4,170. Oil whipsawing; Brent ~$73 in June after the Iran-war spike (current level not verified) | Energy shock is the main macro swing factor. |
| Catalysts, next 60 days | October FOMC (hike vs hold); MU ex-dividend Oct 14; Q3-earnings season for AI customers' capex |
Macro Verdict: HEADWIND, multiplier 0.90. Growth is slowing, inflation is rising, rates are rising. The one offset is a calm VIX.
Section B: MU
1. Business & Moats (score 78)
FQ4: DRAM $39.8B (73% of sales); data-center SSDs near $10B (up 10x+); core data center $18.0B (33% of sales, ~90% gross margin). Micron is the only U.S. maker of high-bandwidth memory (HBM), the stacked chips AI accelerators need. Moats: (1) scale in a three-player DRAM market, (2) HBM technology and packaging know-how, (3) 26 long-term supply agreements covering ~35% of volume through 2030 with $32B in customer cash commitments, (4) very high capital barriers to entry. Weakness: memory is still a commodity cycle, and a few AI customers drive the order book. China sales shrink to single digits of revenue in FY27.
| Strengths | Weaknesses |
|---|---|
| Record margins; net cash; HBM leadership; >75% of FY27 shipments already committed | Earnings ~25 margin points above any prior peak; capex heavy; cyclical history |
| Opportunities | Threats |
| AI inference, server LPDDR (SOCAMM revenue doubled sequentially); pricing floors from contracts | 2027–28 industry capacity wave (Goldman flags HBM price pressure in 2028); AI-spending digestion; export/geopolitics |
Top 5 risks: (1) Competitive/Macro: supply catches up with demand in 2028–29. (2) Execution: $50B+ FY27 capex launched at peak margins. (3) Valuation: price assumes margins stay historically abnormal. (4) Macro: 5%+ Treasury yields and Fed hikes. (5) Customer/Geopolitical concentration in AI buyers, Iran conflict, export rules.
2. Financial Health (score 96, grade A+)
| Metric | Value (source: Micron FQ4/FY26 release, Sep 30, 2026) | Grade |
|---|---|---|
| Gross margin (non-GAAP, FQ4) | 87.0%, up from 84.9% in FQ3 | A+ |
| Operating margin (FQ4) | 82.3% ($44.64B / $54.23B) | A+ |
| Net margin (non-GAAP, FQ4) | 70.8% | A+ |
| P/E (TTM, non-GAAP) | ~14.2x ($1,075 / FY26 EPS $75.52) | A (cheap by scale, but peak EPS) |
| Forward P/E | ~7x on the FQ1 guide run-rate ($38.15 x 4 = $152.60) | A (if earnings last) |
| Price/FCF; FCF yield | ~20x; ~5.0% on FY26 adj. FCF $62.3B vs ~$1.24T market cap | B+ (FY26 FCF includes ramp quarters) |
| Debt; cash; net cash | $5.2B; $73.5B liquidity; $68.3B net cash (~$60/share) | A+ |
| Interest coverage / Debt-to-equity | Enormous / very low (exact equity not pulled) | A+ |
| ROE, ROA, ROIC, current & quick ratio | Not pulled from the 10-K; clearly far above WACC at an 80%+ operating margin (ROIC estimate: well over 50%) | A (est.) |
| Dividend | $0.15/quarter (~0.06% yield); raised 30% in March; ex-date Oct 14 | C (token income) |
3. Earnings Quality & Growth Drivers (score 88)
| Quarter | Revenue ($B) | YoY | Non-GAAP net income | Non-GAAP EPS | vs. consensus |
|---|---|---|---|---|---|
| FQ3'25 | 9.30 | n/v | $2.18B | $1.91 | n/v |
| FQ4'25 | 11.32 | n/v | ~$3.2B (GAAP) | $3.03 | n/v |
| FQ1'26 | 13.64 | +57% | n/v | n/v | Beat (per Zacks) |
| FQ2'26 | 23.86 | +196% | $14.02B | $12.20 | Beat |
| FQ3'26 | 41.46 | +346% | $28.86B | $25.11 | Beat |
| FQ4'26 | 54.23 | +379% | $38.40B | $33.42 | Beat ($31.61 expected) |
n/v = not verified live. I tabulated six verified quarters, not ten, rather than fill gaps from memory.
Quality: GAAP ($32.87) and non-GAAP ($33.42) EPS nearly match; cash conversion is ~61% of revenue in FQ4 (adjusted FCF $33.2B). That is high-quality. But why did it grow? Mostly price, not volume. Management guided single-digit bit growth for FQ1 against double-digit cost growth, and said price increases were already moderating. That is the key tell: growth now comes from tight supply, not from volume.
My driver view: FY26 revenue $133.2B; FQ1'27 guide $61.5B (~$246B annualized), with sequential growth each quarter, so ~$270B for FY27 is plausible (my estimate). Operating margin ~83% near-term (86.25% gross margin minus ~$2.1B opex). I assume margins fade as new clean rooms arrive in late 2028 and beyond.
4. Peers (score 70, low confidence)
I did not pull a live peer table (SK hynix, Samsung, SanDisk), so I won't invent numbers. Qualitatively: MU leads on U.S. HBM positioning and now matches Korean rivals on margin; its premium is justified only if contracts and HBM pricing hold through 2028.
5. Valuation & What's Priced In (score 62)
Market cap ~$1.24T (1.15B diluted shares, company guidance basis); EV ~$1.17T after $68.3B net cash. P/E has compressed as the stock rose: roughly 20x+ a year ago (estimate) to ~14x now, because earnings grew faster than the price. The market is saying "peak earnings, don't pay for them."
Reverse-engineering the price: at an 11% discount rate and 3% terminal growth, $1,075 requires free cash flow to settle around $95B a year after the boom. That is about 60% of my FY27 estimate (~$150B) and far above the pre-AI normal. Plausible? Yes, if contracts and HBM hold. If FY27–28 cash flow misses by 20%, value drops roughly $75/share (~8%), more if the multiple compresses too.
6. DCF (score 50)
FY27E: revenue ~$270B, ~83% op. margin, 15.5% tax (company guide) gives NOPAT ~$189B. Add ~$20B D&A (estimate), subtract $50B+ capex (company: $25B in 1H, higher in 2H) and working capital, giving FCF ~$150B. WACC ~11% (10-yr at 5.2% plus a high-beta equity premium; almost no debt).
| Bear | Base | Bull | |
|---|---|---|---|
| FCF path FY27→FY31 ($B) | 130, 110, 60, 40, 35 | 150, 160, 110, 80, 70 | 160, 200, 220, 220, 210 |
| Terminal FCF / growth | $35B / 2% | $70B / 3% | $150B / 3.5% |
| Discount rate | 13% | 11% | 9.5% |
| Value per share | ~$460 | ~$905 | ~$2,155 |
Equity bridge: EV + $68.3B net cash, divided by ~1.15B shares. Triangulation: price $1,075; analyst average target ~$1,470–1,520; Morningstar fair value $700 (cut from $850; it sees a 2028 peak and 2029 slump). Probability-weighted DCF (25% bull / 45% base / 30% bear) ≈ $1,085, so the stock is roughly fairly valued with very fat tails. If the advantage lasts only 5 years or capex runs higher, the base falls toward the bear case.
7. Technicals (score 55)
FY27-to-date range: $737.89–$1,168.50; 2026 all-time high $1,255; 2026 low $284.20. The stock fell hard in August/September and then bounced ~15% into earnings, and it barely moved on a big beat. Moving averages, RSI and MACD are not verified for today (latest figures I found were from early August: 50-day ~$965, 200-day ~$518). Support: ~$940 (Aug 26 close area) then $738; resistance: $1,168, then $1,255. Bias: Neutral, low confidence (<50%).
8. Analysts (score 90)
About 45 of 46 rate it Buy; average target ~$1,520 (~+40%), high $2,200, low $361. Post-earnings moves: Davidson $2,100, Rosenblatt $1,900, CLSA $1,700, UBS $1,625, TD Cowen $1,600, Baird $1,520, Mizuho $1,400, Bernstein $1,300, Morgan Stanley $1,200. Goldman is Neutral ($1,250). A 98% Buy rating is itself a crowding warning.
9. News & Sentiment (score 68)
(1) Sep 30 record quarter and $61.5B guide; stock flat afterwards. (2) 10 new supply agreements (26 total). (3) Capex to $50B+ and cleanrooms for late 2028+. (4) Morningstar's $700 fair value. (5) Fed's rate hike. Tone: bullish but skeptical ("priced like everyone knows it won't last"). Unpriced tail risks: AI-capex digestion, and Samsung/SK hynix pulling 2028 capacity forward.
10. Risk (score 40, grade D+)
Systematic: very high sensitivity to AI spending and rates; beta figures I found conflict (stale), so no number is quoted. Liquidity: excellent; ~10M shares a day on average, a trillion-dollar mega-cap. Idiosyncratic: cycle peak, capex, customer concentration, geopolitical/export exposure. The balance sheet removes insolvency risk; the risk is a 50%+ drawdown, not bankruptcy.
Score
| # | Subsection | Score | Weight | Points |
|---|---|---|---|---|
| 1 | Business & moats | 78 | 8% | 6.24 |
| 2 | Financial health | 96 | 10% | 9.60 |
| 3 | Earnings & growth | 88 | 10% | 8.80 |
| 4 | Peers | 70 | 7% | 4.90 |
| 5 | Valuation | 62 | 15% | 9.30 |
| 6 | DCF | 50 | 10% | 5.00 |
| 7 | Technicals | 55 | 7% | 3.85 |
| 8 | Analysts | 90 | 7% | 6.30 |
| 9 | News | 68 | 6% | 4.08 |
| 10 | Risk | 40 | 10% | 4.00 |
| Sum | 90% | 62.07 | ||
Note: the framework's weights add to 90%, not 100%. I rescaled (62.07 / 0.90 = 68.97), then applied the 0.90 Macro Headwind multiplier. Final Score = 68.97 x 0.90 = 62.1 → Buy (60–74), medium conviction.
Ten Core Questions (short answers)
1) AI-grade memory (DRAM, HBM, SSD), sticky through contracts. 2) Revenue ~$270B in FY27 (est.), driven by price and AI demand, not volume. 3) Operating margin ~83% fading toward the 40–60% range after 2028 (est.). 4) ~55–60% of revenue converts to FCF near term. 5) $50B+ capex in FY27. 6) Incremental returns excellent now; they fall as capacity floods in. 7) Duration: tight through 2028 per management; I assume a decline from 2029. 8) Terminal: a cyclical, larger, still profitable oligopoly. 9) Present value ≈ $905 base. 10) Price embeds ~$95B perpetual-ish FCF: believable, not conservative.
Verdict: BUY (medium, bordering Hold)
Thesis: The score says Buy because the company is exceptional and Wall Street is nearly unanimous. But my own DCF finds no margin of safety: probability-weighted value ≈ price. You are paying fair value for peak earnings with a very wide range of outcomes ($460–$2,150). The bull case is contracts and HBM pricing holding through 2028–29. The bear case is a 2028 supply wave.
#1 risk (~30% probability): memory pricing rolls over in 2028 as new capacity arrives, with Micron's own $50B+ capex adding to it. Upgrade triggers: capex steady near $50B with higher committed share; a price pullback toward $900; contract share rising above 35%. Downgrade triggers: FY27 capex above $60B without matching commitments; gross margin falling below 85%; Samsung/SK hynix accelerating 2028 supply.
| Watch (next 1–2 quarters) | Good | Bad |
|---|---|---|
| FQ1 revenue (guide $61.5B ±1.5B) | >$63B | <$60B |
| Gross margin (guide ~86.25%) | Rising after Q1 | <85% |
| Capex (1H FY27 ~$25B) | On plan | Run-rate >$60B/yr |
| Contract coverage (35% / 75% of FY27) | Rising | Flat or stalling |
| Fed / 10-yr yield | Hold, <5% | More hikes, >5.5% |
Trading Strategy (6–18 months)
Sizing: conservative 0–1%; moderate 2–3%; aggressive up to 5% (never more, given 50%+ drawdown history). Entry: scale in thirds: $1,000–1,075, $900–1,000 (near base DCF), and a final third below $850. Momentum alternative: add on a close above $1,255 with volume. Targets: Base 12-mo ~$1,000 (DCF base plus one year of carry); Bull 18-mo ~$2,150; Bear 12-mo ~$600 (bear DCF $460 by 18 months). Stops: hard: thesis break (weekly close under ~$740, the August low, or capex/guidance breakdown); soft: trailing ~20% from the high. Options: covered calls above $1,250 are sensible given the expected range (implied-volatility data not pulled); a protective put or collar suits larger holders; call spreads for defined-risk upside.
| Scenario | Prob. | 12M price / return | 18M price / return | 5% position impact (12M / 18M) |
|---|---|---|---|---|
| Bull | 25% | $1,700 / +58% | $2,150 / +100% | +2.9% / +5.0% |
| Base | 45% | $1,000 / −7% | $1,150 / +7% | −0.4% / +0.4% |
| Bear | 20% | $600 / −44% | $460 / −57% | −2.2% / −2.9% |
| Severe downside | 10% | $400 / −63% | $330 / −69% | −3.2% / −3.5% |
| Weighted | 100% | ≈ −4% | ≈ +10% | −0.2% / +0.5% |
Bottom line: expected 12-month return is roughly flat. The upside is real but arrives later and with a very wide spread. That is why this is a small-position Buy, not a table-pounder.
Sources & disclosure: Micron FQ4/FY26 press release and call (Sep 30, 2026), FQ3/FQ2 8-Ks (SEC EDGAR), CNBC, Benzinga, Investing.com, Yahoo Finance, FX Leaders, Finbold (Morningstar fair value), TD Economics/Schwab/CNBC (Fed, Sep 16), BEA via trade.gov.tr (GDP), Morningstar via InvestSmart (price history). Estimates are labeled; "n/v" = not verified live. I did not retrieve the full 10-K, so balance-sheet items beyond the earnings release (equity, current ratio, share-count detail) are not shown. This is not investment advice and reflects a point-in-time snapshot as of Oct 2, 2026. Past performance does not predict future results.
Comments
Post a Comment