Lam Research (LRCX) BofA Securities 2024 Global Technology Conference summary

Industry outlook and growth drivers

• Semiconductor industry is expected to reach $1 trillion in sales, driven by AI and automotive sectors, with capital intensity per wafer increasing at each new process node.

• 3D architectures like Gate-All-Around and advanced packaging are expanding the addressable market for equipment providers.

• AI servers require significantly more silicon, DRAM, and NAND, boosting wafer fab equipment (WFE) demand.

• Government investments in the US, China, Europe, and Japan are expected to drive future equipment orders, though most benefits are yet to materialize.

• The industry remains cyclical, with recent cycles influenced by COVID-related demand pull-forward and inventory corrections.

Market trends and segment performance

• WFE spending is in the low- to mid-$90 billion range this year, with leading-edge foundry and logic showing strength, while NAND remains weak but expected to recover next year.

• China’s investment is growing this year, first half weighted, with a broad customer base in mature nodes and memory, supported by a new $47 billion government fund.

• China’s mix is expected to normalize from 40%+ to low 30s as global recovery, especially in memory, offsets its share.

• Gross margins reached an all-time high due to a favorable customer mix in China but are expected to soften and stabilize at around 48% as the mix shifts.

• Service business (CSBG) remains strong, with spares and upgrades growing alongside the installed base, while Reliant product line is softer this year.

Technology and product innovation

• Lam’s exposure to HBM (high-bandwidth memory) is significant, with proprietary tools for deep silicon etching and metallization, enabling rapid growth in advanced packaging revenue, expected to exceed $1 billion this year.

• Transition from FinFET to Gate-All-Around is driving consistent growth, with $1 billion in revenue expected from this node this year and more investment anticipated next year.

• AI server adoption is incremental to traditional enterprise servers, not cannibalizing existing investments.

• NAND’s role in AI is growing, with SSD demand rising as the industry shifts from hard disk drives to higher-density NAND technologies.