Not NVIDIA, but more expensive than NVIDIA! Is Arm worth $52 billion?

Arm, with a valuation of $52 billion, would you dare to take it?

On Tuesday, September 5th, Arm submitted its IPO application to regulatory agencies. According to the application, Arm's ADS will be priced between $47 and $51 per share, with a total issuance of 95.5 million ADS. This means that, based on the median price range, Arm's valuation is approximately $52 billion.

SoftBank has conveyed a clear message to skeptical analysts and investors that Arm's revenue will grow rapidly in the coming years after raising patent fees by about 40%.

When SoftBank recently acquired the shares of Arm that it did not directly hold from its Vision Fund, the valuation of Arm was $64 billion. Although the valuation did not meet Masayoshi Son's expectations, Arm is still a hot commodity in the chip industry this year and will become the largest IPO globally. Moreover, this target valuation indicates that the market is quite optimistic about Arm's prospects and that it is positioned in the middle of the market value of companies in the Philadelphia Semiconductor Index.

However, many analysts do not buy into SoftBank's optimistic expectations. They believe that Arm's valuation of $52 billion may be too high. Arm is not NVIDIA, and from certain perspectives, Arm is almost as expensive as NVIDIA. However, its business growth rate, revenue, and P/E ratio are difficult to support Masayoshi Son's vision and a valuation of $52 billion.

If Arm is priced at the midpoint of the IPO guidance range, its P/S ratio would be around 18 times, which is only exceeded by NVIDIA among the constituents of the Philadelphia Semiconductor Index.

Analysts believe that the fundamental reason why companies such as Apple, Google, NVIDIA, Samsung, Intel, and TSMC agreed to become cornerstone investors in this offering is to prevent Arm from being acquired by other cornerstone investors. Protecting Arm's neutral position in the industry also means that after going public, the shareholder structure may become a situation of confrontation among various forces, making it difficult to determine the ultimate control.

Arm will also bet on the "computing power race" in the booming AI market in the future. However, analysts believe that Arm is on the edge of the AI trend and may not become a core player in the future. Moreover, the investment return cycle of Arm is controlled by other players, which may make it difficult for those "short-term profit-oriented" investors in the long-dry IPO market to bear or wait.

SoftBank: Revenue will accelerate

According to sources familiar with the matter, Arm told investment bank analysts in Cambridge, UK last month that after using the new version of chip technology, they will increase the patent fees charged to smartphone companies by about 40%, which will accelerate their revenue growth in the coming years. Arm executives told analysts that with technological improvements, they have been able to increase their royalty rates from around 2.3% to 3.2%. Insiders say that Arm has already secured over 80% of its patent revenue for the fiscal year ending in 2026 through signed contracts, and this increase in revenue will further drive its profit growth.

According to forecasts from Goldman Sachs and Jefferies, Arm's sales are expected to exceed $4 billion by the fiscal year ending in 2026, representing a growth of over 56% compared to the fiscal year 2023:

The growth in patent fees will help Arm's revenue growth recover to a level similar to that of the fiscal year 2022, which saw a 33% year-on-year increase in revenue.

Media reports indicate that these forecasts were made in March and were kept confidential until the IPO was announced.

Sara Russo, Senior Analyst at AB Bernstein, expressed confidence in Arm's new chip architecture during a media interview, stating that the new version of the chip is expected to bring improved security and computational performance, putting Arm's revenue back on a fast growth trajectory.

Increasing Revenue Pressure on Arm

However, many investors seem skeptical of Arm's optimistic revenue expectations.

According to SoftBank Group's financial report, Arm's quarterly revenue for the latest fiscal quarter ending on June 30 decreased by 2.5% to $675 million compared to the same period last year, with net profit declining by over 50% to $105 million.

In the fiscal year ending in June 2023, Arm's revenue was slightly below $2.7 billion, placing it at the bottom of the Philadelphia Semiconductor Index component companies when measured by this indicator.

Therefore, analysts point out that Arm's moderate market value has led to a premium compared to many other chip companies such as Marvell Technology Group, GlobalFoundries Inc., and Onsemi, all of which have significantly higher revenue than Arm.

In terms of revenue composition, Arm has two main sources of income: licensing (accounting for 40% of total revenue) and patent royalties (accounting for 60% of total revenue).

Among them, patent royalties are the most profitable business for Arm, despite their relatively low pricing. This is also a significant reason why Arm maintains a market share of 99% while its revenue and profit are not particularly large. In 2022, the total value of chips using Arm technology reached $98.9 billion, accounting for nearly half of the market share. However, Arm's profit was only $1.68 billion, representing 1.7% of the chip value.

At the same time, this "unusual" revenue model of Arm has also diluted the benefits brought by AI. Moreover, the operating costs of this business model are not low: in the most recent fiscal year ending in March, the ratio of research and development expenses to Arm's annual revenue was 42%. This is compared to chip companies in the Philadelphia Semiconductor Index, which have an average ratio of about 16%.

However, analysts point out that investors may not be concerned about the increased R&D expenses associated with chasing the AI wave.

The nature of Arm's business also means that it takes several years for investments to pay off. With a valuation of $50 billion, Arm's investors may not have such patience.

Arm typically focuses on the development of commercially viable products that have been validated. However, the investment return cycle, from conceptual design to selling the program code to licensees, is actually controlled by other players. It usually takes Arm eight years to complete this cycle.

If we also consider the R&D cycle of cloud computing companies like Amazon, the production cycle of chip foundries like TSMC, and the replacement cycle of old chips in major cloud computing giants' data centers, even if Arm has obtained a ticket to enter the AI field and become one of the players, it cannot become a core player.

Can Arm become a core player?

Analysts point out that since SoftBank's acquisition, Arm has "missed many opportunities" and may only be able to sit on the periphery of the arena without becoming a core player.

Arm has bet on the "computational race" in the thriving AI market as its high-growth point after going public. This may mean an increase in demand for chips designed by Arm, such as those used in autonomous vehicles. Arm's prospectus states, "Arm will become the core company after the transformation brought by artificial intelligence."

Rene Haas, who took over as Arm CEO last year, is also focusing on more advanced computing, especially chips for data centers and AI applications:

We believe that if we invest in certain specific instructions, such as SME and SVE - which are vector extensions for specific workloads on super-scale processors - we can make a big impact in the hyperscale server market.

However, ARM itself admits that its CPU chips based on its own architecture are not the most suitable for running the latest AI algorithms. ARM's design architecture prioritizes speed and computational simplicity over raw processing power. It is precisely because of the low cost, low power consumption, and high efficiency brought by its chip architecture that ARM quickly opened up the mobile chip market and gained a dominant position.

ARM estimates that it currently holds a 10% share of the $18 billion cloud processor market, slightly higher than 7% in 2020, and expects this market to grow at a double-digit percentage in the coming years.

However, James Anderson, one of the UK's most famous technology investors, believes that Arm has missed out on many areas, including cloud computing. He believes that although Arm is strong in the smartphone market, it is unclear whether Arm will play a key role in the expanding era of artificial intelligence, and he does not believe that Arm has a unique advantage in AI development.