Tech Catch Up – Arm

Indirect competition

• Arm disappointed the market earlier this month with a weak profit forecast, but this is due to increasing investments in advanced products, which should underpin growth but come at the risk of alienating its current customers.
• Arm reported FQ1 25 results that met expectations with revenues / EPS of $1.05bn / $0.35, but guided weakly when it came to profits.
• Here, FQ2 EPS will be $0.29 – $0.35 ($0.32), below consensus of $0.35, which was enough to send the shares down 14%.
• This is a result of an acceleration of investments in “next generation technologies” which, given the commentary on the conference call leads me to believe that this means products and services that are more like full chip solutions than the processor designs that it has historically designed.
• This makes complete sense because this is how Arm can capture more value per chip that is sold, which is what will allow Arm to grow revenues and profits at a rate that its market valuation demands and expects.
• However, this comes at the risk of enraging its existing client base as well as accelerating development of the RISC-V instruction set, both of which could be detrimental to Arm’s long-term business.
• The possibility of Arm making its own chips has been rumoured for some time and first came to light during the Arm vs Qualcomm trial which concluded at the end of 2024.
• At the time, Arm denied that it was intending to go down this route, but Mr Haas admitted during the trial that it was something that the company had considered but had made no firm decision on.
• During the FQ1 25 conference call, Arm made its firmest comment yet that this is a possibility, and although it has announced nothing, I suspect that this is where the increase in R&D is coming from.
• This all begins with the Arm Compute Subsystem (CSS), which is a pre-integrated bundle of IP blocks such as a CPU, a memory subsystem, and a system interconnect which can be taken by a customer and turned into a chipset.
• The key to the CSS is that Arm has already done some of the difficult parts of chip design, making it much easier for customers who are not experts in chip design to make their own chips.
• From here, the next logical step for Arm could be to move one step further up the value chain into making its own chips and selling them directly to hardware makers.
• This is where the problems begin, as this would bring Arm directly into competition with its biggest customers, such as Qualcomm, MediaTek, Nvidia and so on.  
• The last thing any customer wants is to see one of its core suppliers, which has intimate knowledge of its products, begin directly competing with it.
• The supplier would then have a conflict of interest and be incentivised to ensure that its in-house products receive the latest technology first and to skew the roadmap in its favour.
• This is not always the case (e.g. Samsung), but it is a legitimate concern, and Arm’s customers are already feeling uncomfortable.
• Reading the commentary leads me to believe that Arm is not about to enter the merchant market for chips that its customers address but is more likely to focus on the custom market.
• This would be someone like Meta designing its own data centre chip to train and inference its LLMs, meaning that it would be able to reduce its reliance on Nvidia or AMD.
• Arm also cited the examples of Xiaomi launching the XRING 01 and Samsung’s Exynos 2500, which are processors that are used to power smartphones and potentially tablets.
• This is not in direct competition with Qualcomm or MediaTek, as Arm is simply helping companies to make their own chips, but it is indirect competition in that these companies will be less inclined to buy chips from Arm’s customers in the future.
• Arm’s customers will also be incentivised to leave the Arm ecosystem and focus instead on RISC-V.
• The net result is that Arm needs to tread a very delicate balance between extending its business further up through the value chain and alienating its existing customer base.
• This means that it needs to go on a charm offensive and make the case to its customers that its independence will be maintained.
• Failure to succeed will see customers accelerate their RISC-V programs and give the overall RISC-V ecosystem far more air to breathe than it currently has.
• If RISC-V hits critical mass and standards for higher chip functions begin to form and an ecosystem is created, this will greatly increase RISC-V’s ability to compete with Arm.
• Arm currently has complete dominance in many verticals and is growing share, but for this to continue with Arm as a chipmaker, it needs to strike a delicate balance between maximising its revenue opportunity, keeping its customers happy and ensuring that RISC-V stays in its box.
• Making its own chips remains a high-reward but high-risk proposition.