The ledger remembers what the market forgets. The SEC filing is stark: Third Point LLC, Daniel Loeb's $12 billion hedge fund, has offloaded a portion of its stake in Lam Research. The transaction, executed in late Q4 2024, represents a tactical recalibration, not a panicked retreat. For the macro watcher, this is not a single-stock story. It is a data point on the global capital expenditure cycle, a signal embedded in the supply chain of the AI revolution.
Lam Research is the gatekeeper of the etch and deposition process. Without its high-aspect-ratio etch tools, the 200+ layer 3D NAND stacks and the silicon interposers for HBM simply do not exist. The company commands roughly 40% of the etch market for advanced memory, a position built on decades of process engineering. Its tools are the pickaxes in the AI gold rush. But a pickaxe maker's stock is priced on the expectation of future picks, not the gold already mined.
The context is a semiconductor equipment sector that has rallied aggressively on the AI narrative. From late 2022 to late 2024, the VanEck Semiconductor ETF (SMH) nearly doubled. Lam's P/E multiple expanded from a historical average of 25x to over 30x. This multiple expansion was predicated on a simple thesis: AI infrastructure spending would grow linearly for the next five years. Third Point's move suggests a belief that the market is pricing in a straight line, while the cycle is a curve.
The core insight is the decoupling of AI hype from equipment order velocity. The market has priced Lam as a perpetual growth story. But the data shows that WFE (Wafer Fab Equipment) spending is a cyclical beast. The global WFE market, which hit a record $100 billion in 2022, experienced a 15% correction in 2023. The 2024 recovery was driven by HBM and advanced logic, but the recovery is concentrated. The broad-based memory recovery used to lift all boats. Now, the boat is lifted by a single engine: AI memory.
Based on my experience managing liquidity during the 2022 bear market, I recognize the pattern. The smart money is not selling because the technology is broken. It is selling because the order book visibility is peaking. Lam's guidance for calendar Q1 2025, when it reports, will likely show a sequential decline in orders from the AI-driven spike. This is not a failure of the company. It is the natural rhythm of a capital-intensive industry. The real question is whether the market has already discounted this deceleration.
Consider the customer concentration. The top five customers—TSMC, Samsung, SK Hynix, Micron, and Intel—account for over 60% of Lam's revenue. These customers are not reducing their long-term AI commitments. They are, however, recalibrating their near-term installation schedules. The fab construction cycle has a lead time of 18-24 months. The equipment orders are the final mile. The initial surge of orders for the AI build-out is complete. The next wave, for the next-generation nodes, is still in the design phase. This creates a vacuum in the equipment order book, a period where the easy money has been made and the market waits for the next catalyst.
The contrarian angle is that the decoupling thesis is itself a trap. The narrative posits that AI-driven equipment demand will decouple from the broader semiconductor cycle, providing a floor for Lam's valuation. This is half-true. The AI demand is real and structural. But the equipment companies are not immune to the rhythm of the cycle. They are the most sensitive point in the chain. A 10% slowdown in customer capital expenditure translates into a 20% decline in equipment orders, due to the high fixed costs and the order-to-revenue conversion lag. The decoupling narrative is a comfort blanket that the market uses to justify high multiples. Third Point is removing the blanket.
Geopolitical friction adds another layer of constraint. The U.S. export controls on China, implemented in October 2022 and expanded in 2023, have structurally capped Lam's addressable market. China accounted for 29% of Lam's revenue in FY2021. By FY2024, that figure has dropped to the low 20% range. The service revenue from China, which carries higher margins, remains sticky. But the sale of new tools for advanced process nodes is effectively blocked. The company's long-term growth profile is now more dependent on the rest of the world, which is expanding capacity more slowly. This is not a short-term issue. It is a permanent shift in the company's addressable market. The market has not fully priced this structural headwind, choosing to focus on the AI tailwind.
The takeaway is not about the direction of Lam's stock. It is about the positioning of the portfolio. Third Point's move is a bet on the macro cycle, not against the company. It is a signal that the easy phase of the AI trade is concluding. The market is entering a period of consolidation where fundamentals will be tested against elevated expectations. The next six months will reveal whether the equipment cycle is in a mid-cycle pause or a prelude to a downturn. I have seen this pattern before. The ledger remembers what the market forgets. The winners will be those who can distinguish between a pause and a pivot. The answer lies in the data, not the narrative. We do not build on hype; we build on consensus. The consensus is now being tested.
