Arm Holdings Plc /Uk (NASDAQ: ARM)

Sector: Technology Industry: Semiconductors CIK: 0001973239
Market Cap 163.91 Bn
Div. Yield 0.00
Add ratio to table...

About

Investment thesis

Bull case

  • The data‑center royalty revenue that arm reported for the quarter surged more than 100% year over year, a figure that reflects the growing dominance of inference‑centric workloads in hyperscale environments. The company’s narrative that CPUs are becoming the backbone of agent‑based AI explains the rapid scaling of core counts in new processors from AWS, Microsoft, Google, and NVIDIA, all of which have announced Arm‑based designs with 150–200 cores. With arm projecting that its share of the hyperscaler market could approach 50%, the implied royalty upside is substantial, especially given the high per‑core efficiency that arm’s Neoverse platform delivers. As cloud providers increasingly favor custom silicon that integrates Arm cores with DPUs and smart NICs, the revenue stream is poised to outpace the traditional mobile segment, which arm already indicates could become its second largest business in the next few years.
  • Arm’s compute subsystem (CSS) has transitioned from a niche acceleration tool to a mainstream IP asset, as evidenced by the jump from single‑digit to double‑digit royalty share last year and the company’s expectation that CSS could capture upwards of 50% of its royalty mix in the next three to five years. The CSS model offers significant integration and risk reduction for silicon partners, reducing design cycle time by about 50% and allowing them to ship faster, which in turn increases the per‑chip royalty base. With five CSS customers already shipping silicon and additional licensing agreements on the books, the momentum is self‑reinforcing, as more customers adopt the higher‑level abstractions and unlock higher margin royalties. The company’s claim that CSS is now in “triple‑digit” growth indicates a hidden catalyst that the market has yet to fully price in.
  • Licensing activity has accelerated in tandem with the rollout of next‑generation architectures, with arm reporting a 25% increase in license revenue to $505 million and an annualized contract value growth of 28% year over year. The recent $200 million deal with SoftBank, driven by that conglomerate’s AI compute strategy and its acquisitions of Ampere and Graphcore, is a testament to arm’s ability to secure high‑value, long‑duration contracts that will generate recurring revenue for years. Importantly, arm’s licensing pipeline is diversified across automotive, robotics, and consumer sectors, each of which is entering a period of rapid innovation that will likely drive further adoption of Arm’s instruction set. This cross‑industry reach provides a robust moat and a steady stream of new deals that can cushion the company against cyclicality in any single market segment.
  • The company’s ecosystem advantage is reinforced by its developer community, which boasts over 22 million developers and controls more than 80% of the global processor silicon market. This widespread adoption ensures that as AI workloads expand into edge, automotive, and physical systems, there is already a talent pool and tooling ecosystem in place to accelerate development. Arm’s focus on energy efficiency, low latency, and flexibility aligns perfectly with the stringent power and safety constraints of on‑device AI, enabling it to capture market share in autonomous vehicles, drones, and industrial robots. As more companies ship AI‑centric products, the per‑chip royalty and licensing income will rise, creating a virtuous cycle that is only beginning to manifest.
  • While arm has ramped up R&D spend by 37% year over year, it has managed to keep non‑GAAP operating margins at 41%, indicating that the company is investing in high‑margin, high‑impact technologies such as chiplets, custom SoCs, and advanced process node migrations. The focus on next‑generation architectures, including v9 and CSS v3, suggests that arm is positioning itself to deliver the performance density required for the next wave of AI and machine learning workloads. If the company can maintain its high development velocity, it will likely sustain or even improve its margin profile, as the incremental cost of R&D is offset by the higher royalty and licensing fees associated with newer IP. This dynamic, combined with a solid pipeline, provides a catalyst for long‑term growth that may be underappreciated by current valuations.
Read more

Bear case

  • The memory supply chain volatility that arm has highlighted during the Q&A could translate into a tangible erosion of smartphone volumes, particularly at the lower end of the market where cost sensitivity is highest. Although the company estimates that this will only lead to a 1–2% negative impact on total royalties, the cumulative effect across multiple product cycles could compress growth, especially if the macro environment weakens consumer spending. Moreover, the uncertainty around future memory pricing may also affect the adoption of CSS and v9 architectures, as these rely on high‑density, low‑latency SRAM blocks that could become costlier or less available. If arm fails to secure sufficient memory supplies, it risks falling behind competitors who may lock in better terms with their own silicon partners.
  • Arm’s aggressive R&D investment, which grew 37% year over year, is putting upward pressure on operating expenses and could erode profitability if the company's growth trajectory stalls. The non‑GAAP operating margin, while currently healthy at 41%, is still susceptible to compression if revenue growth does not keep pace with the rising cost of engineering resources. Management has acknowledged the need for sequential growth to moderate in 2027, yet there is no clear path outlined to offset the expense ramp. Should market conditions deteriorate or demand for new IP slow, arm may find itself in a margin squeeze that could diminish investor enthusiasm.
  • Arm’s revenue concentration is a structural risk; a large proportion of its royalties come from a handful of flagship OEMs and hyperscalers. The company’s reliance on a few high‑volume partners—Android OEMs for mobile, AWS, Microsoft, Google, and NVIDIA for cloud, and automotive leaders like Rivian—means that any renegotiation of licensing terms, shifts in procurement strategy, or market exit could have a disproportionate impact. The Q&A revealed that management expects high‑end flagship shipments to offset low‑end volume dips, but this assumption could be undermined if those flagship OEMs delay product cycles or switch to alternative IP. A sudden reduction in orders from a single OEM would ripple through the royalty revenue stream, highlighting a lack of diversification.
  • While arm touts CSS adoption as a major catalyst, the company also downplayed potential bill‑of‑materials (BOM) cost implications, claiming no visible impact. However, the complexity of integrating new IP subsystems into silicon design is non‑trivial, and any hidden cost or delay could slow the pace of CSS roll‑out. If BOM cost pressures materialize, customers may opt for more mature or lower‑cost alternatives, eroding arm’s projected 50% share of the royalty mix. Additionally, the transition to higher‑core‑count CPUs for inference could lead to diminishing returns if the performance gains do not justify the added silicon real estate, thereby curbing adoption.
  • The licensing pipeline, though growing, is not immune to saturation. High‑value deals such as the SoftBank $200 million transaction represent significant revenue, but they also indicate a concentration of income from a few large contracts. If arm cannot secure new high‑ACV deals at a comparable pace, the license revenue growth could plateau. The company’s current guidance of “high teens” growth in licensing for 2027, while optimistic, does not account for a potential slowdown in the pipeline, especially if competitors offer more attractive licensing terms or if the market shifts toward open‑source alternatives. A contraction in license deals would directly hit the company’s top line and could undermine the optimism of the market.
Read more

Product and Service Breakdown of Revenue (2025)

Segments Breakdown of Revenue (2025)

Peer comparison

Companies in the Semiconductors
S.No. Ticker Company Market Cap P/E P/S Total Debt (Qtr)
1 NVDA Nvidia Corp 4,271.43 Bn 35.65 19.78 8.47 Bn
2 AVGO Broadcom Inc. 1,484.69 Bn 59.26 21.74 66.06 Bn
3 MU Micron Technology Inc 468.64 Bn 17.14 8.06 10.14 Bn
4 AMD Advanced Micro Devices Inc 356.31 Bn 78.73 10.29 3.22 Bn
5 TXN Texas Instruments Inc 341.76 Bn 35.89 19.33 14.05 Bn
6 INTC Intel Corp 239.86 Bn -533.67 4.54 46.59 Bn
7 ARM Arm Holdings Plc /Uk 163.91 Bn - - -
8 ADI Analog Devices Inc 156.51 Bn 58.29 13.31 8.14 Bn