Arm Holdings plc American Depositary Shares (NASDAQ:ARM) Q4 2024 Earnings Call Transcript

Arm Holdings plc American Depositary Shares (NASDAQ:ARM) Q4 2024 Earnings Call Transcript May 8, 2024

Arm Holdings plc American Depositary Shares beats earnings expectations. Reported EPS is $0.36, expectations were $0.3. ARM isn’t one of the 30 most popular stocks among hedge funds at the end of the third quarter (see the details here).

Operator: Good day, and thank you for standing by. Welcome to the ARM Fourth Quarter Fiscal Year Ending 2024 Conference Call. At this time, all participants are in a listen only mode. After the speakers’ presentation, there will be a question-and-answer session. [Operator Instructions] Please be advised that today’s conference is being recorded. I would now like to hand the conference over to your speaker, Ian Thornton, Vice President of Investor Relations. Please go ahead.

Ian Thornton: Thank you very much. Good morning and good afternoon to everybody. My name is Ian Thornton, and I’m the Head of Investor Relations at ARM. I would like to welcome everyone to our earnings conference call for the fourth quarter of the fiscal year ending, 31 March 2024. I’m joined today by Rene Haas, the Chief Executive Officer of ARM; and Jason Child, ARM’s Chief Financial Officer. Hopefully, you will all have downloaded and read the shareholder letter. If not, it is available on the ARM Investor Relations website at investors.arm.com. The shareholder letter provides a rich update on our strategic process in the quarter — progress in the quarter, sorry. Before we begin, I’d like to remind everyone that during the course of this conference call, ARM will discuss forecasts, targets and other forward-looking information regarding the company and its financial results.

While these statements represent our best current judgments about future results and performance as of today, our actual results are subject to many risks and uncertainties that could cause actual results to differ materially from what we expect. In addition to any risks that we highlight during this call, important risk factors that may affect our future results and performance are described in our registration statements on Form F1 filed with the SEC on September 14, 2023. ARM assumes no obligation to update any forward-looking statements, which speak only as of the date they are made. In addition, we will refer to non-GAAP financial measures during the discussion, reconciliations of certain of these non-GAAP financial measures to their most directly-comparable GAAP financial measures and a discussion of certain projected non-GAAP financial measures that we are not able to reconcile without unreasonable efforts and supplementary financial information can be found in the shareholder letter that we released earlier today.

The shareholder letter and other earnings related materials are available on our website at investors.arm.com. And with that, I’ll turn the call over to Rene, who has some prepared remarks.

Rene Haas: Thank you, Ian, and hello everyone. So I’m just going to make a few comments to kick off the call and then I’ll pass it over to Jason. But in summary, this quarter, Q4, obviously being the end of our fiscal year was just outstanding. We have record revenues for this quarter and for our first fiscal year as a public company being completed, also record revenue, exceeding the high end of the guidance range. More specifically, for Q4 revenue was up 47% year-on-year, royalties up 37% year-on-year and this is really driven by acceleration of v9 adoption, which I’ll speak about a little bit more. And also licensing, up 60% year-on-year, which is really a function of increased R&D investment to capture the huge opportunity that is all things AI.

Now in looking back, the expansion strategies that we talked about during our roadshow and at IPO are now all driving growth for the company. As mentioned, we had significant royalty growth in the last quarter, up year-on-year 37%, really driven by v9 adoption. And what we’re seeing is the acceleration of v8 to v9, which drives not only better royalties, but we’re also seeing more CPUs inside the chip, which compounds that royalty growth really across all end markets. And the significant driver for that incline has been around smartphones, but broadly, we also see that in our infrastructure business as well. And v9 adoption will only continue to increase. In the last quarter, we’ve also seen proof points of our diversification strategy. Google, the latest hyperscaler announced their Axion processor based on ARM, a custom chip intended for the data center.

They chose us largely because of our compute efficiency, but also the ability to not only have a high-performing chip, but to design an increasingly performant blade, rack, and system for a fantastic TCO. We also announced our very first autonomous solutions based on v9. This is very, very significant as we’re now bringing v9 performance to the automotive sector with automotive-enhanced features such as functional safety, and we expect huge growth around this area. And we also have introduced the lowest power transformer on the planet, the Ethos-U85 for IoT-based designs. One of the strategies we put in place that we are most comfortable with in terms of its growth but very confident in terms of its trajectory is around our, what we call compute subsystems.

And these are essentially taking blocks of IP, putting them together into a full solution, verified and validated that saves customers huge time to market and also gives them a highly performant solution. So we announced our V3 Neoverse CSS this quarter, which will give increased performance and benefits to customers. The first automotive CSS is now in discussions with our key partners [Technical Difficulty] customers in terms of time to market and efficiency. And our first customer in the Neoverse space doing a design, Microsoft, their Cobalt chip is now ramping. But probably from a more exciting standpoint, we are oversubscribed on this compute subsystem strategy. We have far more demand for the product than anticipated, and we are anticipating growing that significantly over time.

A technician overlooking a circuit board being built and tested for a semiconductor device.

Every end market that we approach has a need for CSSs, and we’re very excited about talking about them in the future. All of this is also being driven by AI. What we are seeing is because ARM has the largest installed base of CPUs on the planet and has over 70% of the world’s population using those CPUs. It’s natural that as these AI workloads are now being moved from anywhere from the edge devices to the training data center, that they need support from an ARM CPU standpoint. So whether it’s from cloud Edge from GPT to Llama, all AI workloads rely and run on ARM, and we only see this increasing. Our licensing activity is probably the best proxy for that. The way to think about licensing revenue as it applies to Al is as software is moving faster than hardware, the hardware designs need to be upgraded quickly to make sure they can capture the needs of these new Al workloads.

So because of that, we have seen huge growth in our licensing activity. We talked about that last quarter and it continued this quarter. So based upon this, we are very, very confident of our growth outlook for the upcoming year. This past year was over 20% revenue growth, and we expect that to be even better than that in this year and the upcoming years. Our growth has been accelerating. Lastly, it’s taken ARM 20 years to get to $1 billion in revenue. It took us 10 years to get to $2 billion. This year, we passed $3 billion in only two years after our first $2 billion year, and we expect to be near $4 billion this year. The future is very bright and we’ll run an ARM, and I could not be more excited about the future that we have. And with that, I’ll turn it over to Jason.

Jason Child: Thank you, Rene. Q4 was a strong end to a tremendous year for ARM. For the quarter, we grew revenue 47% year-over-year to $928 million with licensing revenue up 60% and record royalty revenue up 37%, while also delivering non-GAAP operating margin of 42%. Also — as well as delivering strong revenues, we also have grown our remaining performance obligations, or RPO, by 45% year-over-year to nearly $2.5 billion. The royalty acceleration and record RPO provides a great setup for FY 2025. Turning to guidance. I will briefly touch on both the first quarter and fiscal year ending March 31, 2025. For Q1, we expect revenue of between $875 million and $925 million, representing a 30% to 37% year-over-year increase. Non-GAAP operating expense is expected to be $475 million, a sequential decline due to a change in remuneration as we complete our transition away from cash awards to equity awards, and a onetime Q4 expense.

Resulting non-GAAP EPS is expected to be between $0.32 and $0.36. Unpacking Q1 revenue dynamics a little further, we expect royalty revenue to remain strong with year-over-year growth of approximately 20%. This growth is driven by continued ARMv9 adoption and recovery in smartphones, along with continued share gains in automotive and at hyperscalers. This is partially offset by weakness in loT driven by inventory correction in the broader industrial market, as has been widely reported by many of our semiconductor peers. For Q1 licensing and other revenues, we expect slight sequential growth, driven by revenue from backlog. Looking out to fiscal 2025, we expect revenues of between $3.8 billion and $4.1 billion, representing a 17% to 27% year-over-year increase.

We expect non-GAAP operating expenses of $2.05 billion, representing a 19% year-over-year increase as we continue to invest in R&D to support future growth initiatives. Our full year non-GAAP EPS is expected to be between $1.45 and $1.65. We believe our license business is best understood by the measure annualized contract value, or ACV. Our outlook for ACV remains strong. We expect low double-digit growth for the year, reflecting the durable demand in ARM ‘s latest IP. Licensing revenue, however, will continue to be lumpy from period to period due to the timing of revenue recognition. Based on our current forecast, we expect the first half to represent approximately 40% of our license revenue for the year, with Q2 being the smallest and Q4 being the largest quarter of the year.

We have high visibility through a combination of backlog renewals and new licenses. For royalties, we remain very confident in the demand for ARM-based chips and expect full year growth in the mid-20% range driven by continued v9 adoption, market share gains in cloud and automotive, as well as chips based on our compute subsystems starting to ramp in the second half of the year. Looking out further beyond this year, based on the pipeline for new licenses, agreements already signed, and royalty-bearing chips in development are now shipping, we expect to maintain total revenue growth of at least 20% year-over-year for each of the fiscal years ending in 2026 and 2027. I’ll now turn the call back over to lan.

Ian Thornton: Thank you, Jason. We will now move to the Q&A portion of the call. We request that you limit yourself to just one question each so that everyone gets a chance to ask their most pressing question. If time allows we will go around again. I hand back to you, operator.

Operator: Certainly. [Operator Instructions] First question will come from Ross Seymore of Deutsche Bank. Your line is open.

See also 25 Richest Billionaires in Healthcare Industry and 30 Countries That Receive the Most Aid From ILO.

Q&A Session

Ross Seymore: Hi guys. Thanks for let me ask a question. Congrats on the strong results and guide. Rene, I had a question on your infrastructure business and specifically the cloud side. You rattled off a number of the design wins that are starting to ramp from a number of the hyperscaler customers. What I was really wondering is, during the IPO process you talked about 10% market share in that market rising to roughly, I think you said 27% or so. Are those products that are occurring now and being launched? Are they contemplating their original ramp or are you seeing an acceleration above and beyond the growth rate that you had laid out for all of us during the IPO process?

Rene Haas: I would say I’m still confident in those numbers, Ross. I would say one of the dynamics that has changed a bit since we chatted, and I think it’s a positive for us is the increased investment in the data center around all things AI and that bodes well for ARM for a couple of reasons. First off, when you think about power efficiency and what’s required for the data center, the implementation of an ARM-based design gets very, very interesting relative to not only building a custom chip that will be more power-efficient, but by designing a blade, an interconnecting the storage and an overall rack that will be much more power-efficient. Secondly, and I think we chatted about this during the IPO process with Grace Hopper, but I think now with NVIDIA’s most recent announcement, Grace Blackwell, you are going to see an acceleration of ARM of the data center in these AI applications.

One of the benefits that you get in terms of designing a chip such as Grace Blackwell is by integrating the ARM CPU with the NVIDIA GPU, you’re able to get an interconnect between the CPU and the GPU that allows for a much higher access to memory, which is one of the limiting factors is for training and inference applications. In a conventional system where you might connect to an X86 externally, you have to do that over a PCI bus, which is much slower. So by using a custom bus in the NVIDIA example like NVLink, you get much higher memory bandwidth. So I think what that is going to mean is that ARM adoption in the data center will increase, probably faster than the numbers that we had indicated, but we’re not seeing anything official right now.

Ross Seymore: Thank you.

Operator: And one moment for our next question. And our next question will come from Vivek Arya of Bank of America Securities. Your line is open, Vivek.

Vivek Arya: Thanks so much for the question. It’s on the v9 conversion. If I look at Slide 15 of the chart presentation you have, it shows that it took about seven years or so for v8 to become about half of the base. Rene, how do you expect this v9 conversion? It seems to have gotten off to faster off the gates, 20% or so conversion in a year. What’s your assumption of what that conversion looks like in your fiscal 2025? I think Jason said mid-20% growth or so in royalty, what does that imply for the level of conversion exiting the year?

Rene Haas: Yes, I’ll let Jason touch on that conversion rate. But what I can say from a high level, what is the difference between v8 and v9 is, back into the conversion from v7 to v8, we really didn’t have an infrastructure business at all to speak about. We do today, and that is all v9. So that is going to be a growth driver that we saw in terms of conversion for v9 from v8 that we just didn’t see from v7 to v8. In the smartphone business, we are seeing a fairly rapid acceleration in the premium handset segment that is moving to v9. And I think it’s been pretty well-documented that it’s the premium handset business that has enjoyed probably better growth than some of the other segments, which has accelerated as well. So those are two drivers for v9 that will look, I think a bit different than v8, and why it should happen more quickly.

I’ll let Jason comment on what we think the potential exit rate might be and particularly how it applies to our year-on-year revenue growth projection.

Jason Child: Yes. I would say the 500 basis points of growth or as a percentage of total going from 10% to 15% last quarter and then this most recent quarter from 15% to 20%. I think that’s a pretty steady clip. It could be plus or minus, but that somewhere around 5-ish percent growth — 500 basis points of share growth per quarter is probably a reasonable assumption. So that puts you probably out somewhere in the two to three-year timeframe that we probably get to the high end, which is probably somewhere maybe around 60% to 70%, because there is still going to be some v8 and v9 out there even as we see growth in the v9 adoption. Keep in mind, I think when we talked about this back at IPO and even last quarter, you do also see an increased royalty that’s almost double from v9 when you go to subsystems.

And so we start to see our subsystems come online at the end of this year — late in the back half of this year and then that will start to take hold next year. That also probably has somewhere in the kind of three to four-year timeframe of becoming a large share of our royalties. So that’s the reason why we have high confidence in a royalty forecast that should be pretty, pretty strong and in that kind of out-of-protein or in that 20% range for years to come.

Vivek Arya: Thank you.

Operator: And one moment for our next question. And our next question will be coming from Matt Ramsay of TD Cowen. Your line is open.

Matt Ramsay: Thank you very much, everybody. Good afternoon. Rene, I wanted to follow-up a little bit on the infrastructure business question that Ross sort of started us off with. And one of the questions I’ve been getting from folks is, I can see very, very clearly how things like Grace Hopper and Grace Blackwell will be accelerants of pulling ARM into the data center and you’ve announced with Amazon, Google, and Microsoft other design wins that will do the same thing. On the flip side, in the accelerated data center, if you just look at a percentage of CapEx that goes to CPUs versus accelerators, that accelerator portion is going to be a vast majority, it seems like and that trend is accelerating pretty rapidly. So maybe you could speak a little bit to ARM’s role, not just in the head node of accelerated servers, but across the accelerator portion of that sort of exciting growth in CapEx? Thanks.