Lam Research Corporation (LRCX) — Coverage File

Data as of: Q4 FY2026 (quarter ended June 28, 2026); FY2026 10-K filed Aug 7, 2026; latest source Goldman Sachs conference, Sep 10, 2026 File last updated: October 10, 2026 (initial build)

---

0. 30-second reopen

• Business: The leading etch and deposition toolmaker plus a 100,000-chamber installed base that sells spares, service and upgrades; FY2026 revenue $23.2B (+26%), 50.5% gross margin, about a third of it from China.
• Core debate: Is Lam's step-up in share and margin structural (vertical scaling raises etch and deposition intensity) or a peak-cycle and China-mix effect that reverses when the AI capex wave and the NAND upgrade cycle end?
• Base thesis: Taller, 3D chips need more etch and deposition per wafer; Lam owns the leading tools and the fleet they run on, so it can outgrow WFE, expand margin toward the mid-50s and compound CSBG on a growing installed base.
• Main bear case: China (~$7.9B of revenue) erodes to local toolmakers, taking above-average-margin mix with it, just as the upcycle peaks and the NAND upgrade wave runs out in 2027; CSBG proved cyclical in the last trough.
• What to watch: China revenue in dollars, non-GAAP gross margin against the 51–52% run-rate, Lam growth vs. WFE in 2027, deferred revenue and receivables, and CSBG growth in the first down quarter.
• Valuation: ~$399B market cap, ~55x FY2026 GAAP EPS, ~1.2% FCF yield (Oct 9, 2026 web price): priced for the upcycle to run well into 2027–28 and for the new margin targets to be met.

---

1. Business snapshot

Lam designs, builds and services wafer-fabrication equipment for etch, deposition and clean, the steps that build and carve the vertical structures in 3D NAND, DRAM and gate-all-around logic. FY2026 (year ended June 28, 2026) revenue was $23.23B (+26%): systems $14.89B (+29.5%) and customer support-related and other (CSBG) $8.35B (+20.2%) [FY2026 10-K]. CSBG covers spares, service, upgrades and the Reliant line of non-leading-edge tools, earned on an installed base of more than 100,000 chambers [Q3 FY2026 call].

Equipment and upgrade revenue was 54% foundry, 39% memory and 7% logic/IDM in FY2026, against 26% foundry and 60% memory in FY2022 [FY2026 10-K; FY2023 10-K]. By geography, China was 34%, Taiwan 22% and Korea 19%. Four customers each took 10% or more of revenue (16%, 15%, 12%, 12%); the 10-K names Micron, Samsung, SK hynix and TSMC as the most significant [FY2026 10-K]. Primary competitors are Applied Materials (deposition and etch), Tokyo Electron and Hitachi (etch), ASM International (ALD) and Screen/Semes (clean) [FY2026 10-K].

2. Core debate

The first question is whether the share gains are structural. Management puts Lam's served market at ~36–36.5% of WFE in CY2026, up from the low 30s, on its way to the high 30s targeted at Investor Day 2025, and says CY2026 is the third straight year Lam outgrows WFE [Q4 FY2026 call]. The bull reading is that vertical scaling in every device type is an etch and deposition problem. The bear reading is that these are company estimates measured during a sold-out upcycle.

The second is how much of the margin and earnings base rests on China and the cycle. China stayed at 34% of revenue in FY2026 (about $7.9B, computed), and the 10-K credits both the FY2024 and FY2026 gross-margin gains largely to "favorable customer mix" [FY2026 10-K; FY2024 10-K]. Domestic Chinese customers declined in Q4 FY2026, when China fell to 26% of revenue [Q4 FY2026 call]. If that mix was carrying margin, the new mid-50s target is harder than it looks.

3. Base thesis

AI is pushing every device type toward vertical scaling: 200-to-500+ layer NAND, 4F² and HBM DRAM, gate-all-around and backside power in logic, and larger advanced packages. Each makes etch and deposition steps longer and more numerous, and Lam leads in the hardest of them (high-aspect-ratio and cryogenic etch, conductor etch with 40,000+ chambers installed, TSV etch and copper plating) [Q4 FY2026 call; Investor Day 2025]. Wins at one node carry into the next through upgrades on an installed base that is rarely retired, and CSBG earns above the corporate operating margin [Investor Day 2025]. Lam has turned that position into a margin step-up: GAAP gross margin went from 44.6% (FY2023) to 50.5% (FY2026), and the 2028 targets from Investor Day ($25–27B revenue, ~50% gross margin, 34–35% operating margin, EPS $6–7) were largely met in FY2026 [FY2026 10-K; Investor Day 2025].

Assumptions this rests on:

1. Lam's SAM rises to the high 30s % of WFE and Lam captures more than half of the increase, so it keeps outgrowing WFE [Investor Day 2025; Q4 FY2026 call].
2. WFE grows again in 2027 from the "low $150B" of CY2026, led by DRAM and leading-edge foundry/logic as 8–10 tier-one cleanrooms open by end-2027 [Q4 FY2026 call; Citi conf. Sep 2026].
3. Gross margin holds at 51–52% and moves toward the mid-50s, with operating margin toward the mid-40s, on Malaysia scale, new products and pricing [Q4 FY2026 call].
4. CSBG grows faster than the installed base in most years and cushions the next trough [Investor Day 2025].
5. China declines as a share of revenue, not in dollars, and does not take margin mix with it [Citi conf. Sep 2026].

4. Main bear case

The bear case is that FY2026 is a peak built on three things that do not last. First, China: export rules bar Lam from some Chinese customers and, per the 10-K, advantage non-US rivals, and the CFO says local toolmakers now have "a captive set of customers" Lam cannot serve [FY2026 10-K; Goldman conf. Sep 2026]. Second, timing: most of the ~$40B NAND upgrade wave lands before end-2027, and NAND greenfield is not expected until "second half of 2028 and beyond" [Q4 FY2026 call; Citi conf. Sep 2026]. Third, the cushion: in the FY2024 downturn revenue fell 14.5%, systems 16.6% and CSBG 11.1%, because upgrades and Reliant behave like capex, and China at 42% of revenue propped up that trough [FY2024 10-K].

How it breaks: domestic Chinese demand keeps shrinking while WFE growth slows in 2028; gross margin drops back below 50% as the mix normalises on a cost base built for peak (headcount +1,800 in one quarter) [Q4 FY2026 call]; and Lam's growth falls behind WFE as Applied Materials and Tokyo Electron win DRAM and gate-all-around positions. Earnings would fall faster than revenue, and the multiple with them.

5. Business quality

High within a cyclical industry. Lam's moat is process know-how embedded in tool-of-record positions: a chipmaker qualifies a tool on a step and keeps it through the node, then upgrades it [Investor Day 2025]. The installed base grew at a ~10% CAGR, CSBG revenue grew faster, and Lam says its tools are rarely retired [Investor Day 2025]. Capital needs are modest: capex was $966M (4.2% of revenue) in FY2026, inside the 4–5% framework [FY2026 10-K; Q4 FY2026 call].

The weaknesses are concentration and breadth. Four customers took 55% of FY2026 revenue [FY2026 10-K]. Lam's own risk factors warn that its narrower product line may make customers see it as "less important" than suppliers with more products, and that some components come from single suppliers [FY2026 10-K].

6. Key value drivers

Revenue = WFE × Lam's SAM share × Lam's share of SAM, plus CSBG on the installed base. WFE: ~$110B in CY2025 and "low $150B" in CY2026, limited by cleanroom space [Q2 FY2026 call; Q4 FY2026 call]. SAM intensity: SAM per wafer guided to double in NAND as stacks reach 500+ layers, rise 1.7x in DRAM and 2x in foundry logic [Investor Day 2025; Citi conf. Sep 2026]. Share: ship share of WFE rose more than 1 point in CY2025 [Q2 FY2026 call]; advanced packaging is growing more than 70% in CY2026 [Q4 FY2026 call]. CSBG: spares and service (driven by ~100% fab utilisation), upgrades (NAND conversions), Reliant (mature-node and China) and new advanced services such as Dextro cobots [Goldman conf. Sep 2026]. Margin: mix, Malaysia volume and a second Malaysian plant, new-product pricing, tariffs [Q4 FY2026 call; FY2026 10-K].

7. Financial profile

FY2026: revenue $23.23B (+26.0%), gross margin 50.5% (48.7%), operating expenses $3.53B, GAAP operating income ~$8.20B (35.3%, computed), net income $7.27B, diluted EPS $5.76 (+38.8%) [FY2026 10-K]. R&D was $2.38B (10.2% of revenue). Operating cash flow fell to $5.86B from $6.17B despite 36% higher net income, as receivables rose $1.96B to $5.34B [FY2026 10-K]. Free cash flow was ~$4.89B (computed: OCF less $966M capex). Lam spent $3.85B on buybacks (including net share settlement) and $1.27B on dividends, ~105% of FCF (computed); cash fell to $5.58B against $3.72B of debt, the earliest maturity a $1.0B note in 2029 [FY2026 10-K]. Deferred revenue fell to $2.43B from $2.68B on lower customer down payments [FY2026 10-K].

Quarterly revenue ran $5.32B, $5.34B, $5.84B and $6.72B through FY2026 (Q1 computed) [Q3 FY2026 10-Q; Q4 FY2026 call]. Q4 non-GAAP gross margin was 52% and operating margin 38.4%. Q1 FY2027 guidance is $8.1B ±$0.4B revenue, 52% gross margin, 39.5% operating margin and $2.15 EPS [Q4 FY2026 call].

8. Valuation

Lam's own frameworks are the anchor. The Investor Day 2025 model for 2028 ($25–27B revenue, ~50% gross margin, 34–35% operating margin, EPS $6–7) was mostly met in FY2026, and management has replaced it with mid-50s gross margin and mid-40s operating margin "over the next several years," plus $30B+ revenue in a trillion-dollar semiconductor industry [Investor Day 2025; Q4 FY2026 call]. At ~55x FY2026 GAAP EPS, the shares discount much more than a normal cycle: annualising the Q1 FY2027 EPS guide gives ~$8.60 and ~37x (inferred; not a forecast), which still assumes the upcycle and new margins hold. The ~1.2% FCF yield is below the dividend-plus-buyback payout rate. The valuation therefore rests on assumptions 1–3; prior-trough multiples are not in sources.

Market data (outside sources/, web, dated) Share price $318.84 at the Oct 9, 2026 close; market cap ~$399.0B; ~1.25B shares outstanding; dividend $1.32 annualised (stockanalysis.com, retrieved Oct 10, 2026). Derived by this file from filings: FY2026 GAAP EPS $5.76, GAAP operating income ~$8.20B, FCF ~$4.89B, net cash ~$1.86B at June 28, 2026. Implied: ~55x FY2026 GAAP EPS (~55x non-GAAP $5.82), EV ~$397B, ~48x GAAP operating income, ~17x revenue, FCF yield ~1.2%. Q1 FY2027 results (due late October 2026) will reset these.

9. Management & capital allocation

Tim Archer has been CEO since December 2018 (ex-Novellus, at Lam since 2012) and Doug Bettinger CFO since 2013 (previously CFO of Avago) [FY2026 10-K]. Their record on stated targets is good: the 2028 model was largely met two years early, and management attributes most of the margin step-up to the close-to-customer manufacturing shift to Malaysia that it set out years earlier [Q4 FY2026 call; Investor Day 2025]. The board added Cadence CEO Anirudh Devgan in February 2026, and two directors retire in November 2026, cutting the board from 12 to 10 [8-K Feb 3, 2026; 8-K Aug 27, 2026].

Capital allocation: the stated policy is to return at least 85% of free cash flow [Q4 FY2026 call]. In FY2026 Lam returned more than it generated, but buybacks were price-sensitive: it bought 9.7M shares at ~$106 in the September 2025 quarter and only ~0.8M at $288–370 in May–June 2026, leaving $4.04B of authorisation [FY2026 10-K]. The dividend declared rose to $1.04 per share from $0.92 [FY2026 10-K]. There is no large M&A.

Pay: CEO total compensation was $28.3M in FY2025 ($30.1M in FY2024). The annual bonus rests on three equally weighted factors (funding, corporate performance, profitability), measured partly on non-GAAP operating and gross margin; 60% of long-term equity is PRSUs on three-year TSR relative to the Philadelphia Semiconductor index, capped at target if absolute TSR is negative; options were dropped in 2025 [2025 proxy]. Nothing in pay rewards returns on capital or through-cycle results. The 2026 proxy and insider-transaction data are not in sources.

Claims to verify over time:

• SAM reaches the high 30s % of WFE and Lam takes >50% of the increase [Investor Day 2025].
• CSBG "should grow nearly every single year" [Investor Day 2025]; FY2024 (−11%) is the counter-example.
• China falls as a share "not because China's going away" [Citi conf. Sep 2026].

10. Risks ranked by damage

1. China erosion and export controls Damage rank: High · Permanence: High · Fixability: Low Mechanism: Export rules and entity listings shut Lam out of Chinese fabs that local and non-US suppliers can serve; each line won locally trains the competitor and stays lost for that fab's life. Evidence today: China 34% of FY2026 revenue (~$7.9B, computed); domestic Chinese customers declined in Q4; the CFO concedes a captive local supplier base [FY2026 10-K; Q4 FY2026 call; Goldman conf. Sep 2026]. What would confirm it: China revenue down in dollars year on year; new entity listings; "customer mix" cited as a margin headwind. Impact: revenue and, if the mix carried above-average margin (inferred), gross margin.

2. Cycle peak and the 2028 air pocket Damage rank: High · Permanence: Low · Fixability: High Mechanism: Cleanroom-constrained spending pulls capacity into 2026–27, the NAND upgrade wave finishes by end-2027 and NAND greenfield waits until 2H 2028+ [Q4 FY2026 call; Citi conf. Sep 2026]. Evidence today: four customers are 55% of revenue; FY2024 showed CSBG falls with systems [FY2026 10-K; FY2024 10-K]. What would confirm it: a revenue miss against guidance, a cut to the 2027 WFE view, deferred revenue falling further. Impact: deep but temporary earnings decline; painful at ~55x.

3. Margin normalisation Damage rank: Medium · Permanence: Medium · Fixability: Medium Mechanism: Part of the margin gain is mix and pricing in a sold-out market; a cost base built for peak (headcount ~22,400 → 23,300 by August 2026) and tariffs deleverage in a slowdown [Q4 FY2026 call; FY2026 10-K]. Evidence today: the 10-K attributes the FY2026 gain "largely" to customer mix, net of tariff costs. What would confirm it: non-GAAP gross margin under 50% at similar volumes. Impact: each gross-margin point is ~$230M of operating profit at FY2026 revenue (computed).

4. Share loss to broader rivals Damage rank: Medium · Permanence: Medium · Fixability: Medium Mechanism: Applied Materials and Tokyo Electron win DRAM, gate-all-around or packaging steps, and customers favour broader portfolios [FY2026 10-K]. Evidence today: none in filings yet; Lam reports gains, but SAM-share figures are its own estimates. What would confirm it: Lam growing slower than WFE in 2027. Impact: slower installed-base growth, which also slows CSBG.

5. Working capital and returns at peak Damage rank: Low · Permanence: Low · Fixability: High Mechanism: Receivables and payouts outrun cash flow late in the cycle. Evidence today: receivables +58% on revenue +26%; DSO 72 days vs. 64 a quarter earlier; payout ~105% of FCF [FY2026 10-K; Q4 FY2026 call]. What would confirm it: OCF falling again while earnings rise, or commercial-paper-funded buybacks (capacity raised to $2.0B in March 2026) [FY2026 10-K]. **Impact:** small with ~$1.9B net cash; a read on discipline.

11. Metrics to monitor

12. Thesis tripwires

13. Open questions

1. How much of the FY2024 and FY2026 "customer mix" margin gain came from domestic Chinese customers? Not disclosed (see §10 risk 1).
2. How big is the second NAND upgrade wave (300–500 layers) after 2027? Management has not quantified it [Q4 FY2026 call].
3. Which customers cut down payments, given deferred revenue fell while demand surged?
4. What share of CSBG is Reliant, and how China-exposed is it? Not disclosed; it drove the FY2024 decline.
5. What changed in the 2026 proxy (pay outcomes for CY2025)? Not in sources.

14. Update log

15. Source map