Financebotresearch desk研究台

KLAC › analyze

KLAC · Analyze

Semiconductors

Price: $192.12 (intraday, −5.5% today) · Market cap: $251.0B · Sector: Semiconductors (process control) Verdict: 🟡 WATCH → STAGED BUY, but not today · Conviction [7.5] (from [8.0] on 2026-07-15) ⚠️ REPORTS FQ4 FY2026 TONIGHT — after the close, 2 p.m. PT, 2026-07-28

Full cross-cutting analysis: compare-korea-memory-selloff-CVLT-SNDK-KLAC-AMKR-GLW-2026-07-28.md Prior: compare-ASML-KLAC-LRCX-2026-07-15.md — quality 9.0/10, conviction [8.0], accumulate $180–200 Note: 10-for-1 stock split effective 2026-06-11. All per-share figures post-split.


The headline development

The 7/15 analysis set an accumulate zone of $180–200 when KLAC was near $300 at 44x forward. It is now $192.12 — inside that zone, at ~37.5x forward. Of the five names screened, this is the only one that has actually walked into a previously-established buy range.

Business quality — not in question (§2)

Metric FY2026 (est., ending Jun'26)
Revenue ~$13.5B (+11%)
Gross margin 61%
Operating margin 41.2%
FCF ~$4.05B (31% FCF margin)
ROE 95%
Share count −4.1%/yr
Dividend 0.45% yield, 22% payout

Quarterly revenue has compounded steadily: $3.06B → $3.17B → $3.21B → $3.30B → $3.42B, with operating margin pinned at 41–42% throughout. KLA holds a majority share of semiconductor process control — the inspection/metrology toll booth every fab must pay, and process-control intensity rises with each node. Management expects process control to grow >20% in 2026, outpacing broader WFE.

Adversarial stress-test (§2.2): a well-funded rival cannot meaningfully attack this. Process control requires decades of accumulated defect-library data, fab-floor integration, and leading-edge trust; ASML doesn't compete here, and Applied/Hitachi have tried for years without displacing KLA. The moat is intact. The threat is cyclical and geographic, not competitive.

What changed since 7/15 — and why I tightened the zone

  • Memory is 15–25% of KLA revenue (SK Hynix, Samsung, Micron). South Korea is 15–20% of geographic revenue. That is exactly the customer base whose equity value fell ~$290B in a day, whose ADR (SK Hynix) now trades below its $149 July IPO price, and whose marginal competitor CXMT just IPO'd raising $8.6B (+466% on debut).
  • If memory capex plans get trimmed into 2027, the FY27 consensus EPS of ~$5.13 (+39% on FY26's ~$3.60) is what's at risk. The multiple isn't the fragile part — the estimate is.
  • China export-control exposure remains open.

Valuation (§3)

Metric Value Context
P/E trailing (FY26 GAAP ~$3.60) 53x 5-yr avg trailing 30.5x
P/E forward (~$5.13) 37.5x sector median 29.6x; KLAC 3-yr avg 33.1x
FCF yield 1.6% rich
Graham IV $18.85 — not meaningful (BVPS $4.46, book destroyed by buybacks)
Bogle expected return ~11%/yr (0.45% + 15% growth − 4.1%/yr compression)
Analyst target (28) mean $234, range $150–325

At 30–36x FY27E $5.13 → fair value $155–185 (base ~$170). Still a premium multiple, but ~15% cheaper than 13 days ago and no longer sitting at consensus target.

Action

Do not buy today. FQ4 results land after the close tonight with September-quarter guidance. Buying a 37x-forward name hours before a print — during a sector de-rating that directly hits 15–25% of its customers — is not a risk worth taking for a long-term holder. Let the print land.

  • Accumulate $165–190 (tightened from prior $180–200 — memory-capex estimate risk is materially higher than on 7/15)
  • Strong buy <$160 (200-day ~$156)
  • Trim $260+
  • Thesis break: process-control revenue growth <10% for 2 quarters, memory customers publicly cutting 2027 capex, or FY27 EPS consensus falling below ~$4.50.

A portfolio-specific passage was removed from the public build.