How SoftBank’s expensive bet on the ‘internet of things’ backfired on Arm

When Masayoshi Son tried to persuade investors of the wisdom of buying one of the world’s most successful chip companies in 2016, the head of SoftBank had a clear message: “For the ‘internet of things’ era, I think the champion will be arm.

But the concept of connecting billions of everyday and industrial devices to the Internet has taken much longer to materialize than anticipated.

Son’s push to capture the Internet of Things (IoT) chip design market was the first bet he made on Arm that didn’t pay off. The second was a $66 billion sale of the company to Nvidia that unraveled last week.

Arm remains the dominant player in smartphone chip design, still the most ubiquitous form of computing but a source of much slower growth in recent years. Ahead of an initial public offering that could come this year, the company is rushing to consolidate its position in new markets it hasn’t tapped to date, while trying to boost profits to attract a new group of investors.

Rene Haas, Arm’s incoming chief executive, told the Financial Times that its products were now “much more competitive” in data centers and cars than when SoftBank bought the Cambridge-based company.

“Make concessions on where to invest and where not to invest. . . those are the trade-offs that public companies and even private companies have to make every day,” he said. “The company is in very good shape.”

When Son led the $31 billion purchase de Arm, saw it as a commitment to the future of the entire technology industry, which was crystallizing at that time around the IoT concept. He proceeded to firmly push the executive team toward designing chips for this future of machine connectivity.

Five and a half years later, it has become increasingly clear that the IoT gamble was a costly misadventure. Furthermore, it distracted Arm from attacking Intel’s dominance in the much larger data center market.

When Son’s vision collided with reality, SoftBank quietly revised its market calculations. A 2018 presentation forecast that by 2026, the IoT controller market would be worth $24 billion and the server market $22 billion.

But, a similar presentation from 2020 predicted that by 2029, the IoT chip market would reach just $16 billion, while the server market, of which Arm had so far only captured a 5 percent share, would reach $32. billion. The Japanese tech group also revised down its estimate of the value of the IoT market, from $7 billion in 2017 to $4 billion in 2019.

Tudor Brown, who co-founded Arm in 1990 and was an executive with the company for 22 years, described its heavy investment in IoT as “strange” given that “there was never going to be any money in that market.” He added: “Focusing on that, they didn’t focus on the grand prize, which was the server.”

In Arm’s regulatory filings in December, the company made a strong case. against seeking an initial public offering and in favor of an Nvidia sale, describing how shareholder pressure could stifle the company’s ability to invest in PC and data center markets, which had been “difficult to penetrate” and where it had made only “limited inroads.” “. Public market investors would “demand profitability and performance,” meaning reduced costs and a lack of financial clout to invest in innovative new businesses, Arm’s presentation added.

“We always felt that the Nvidia acquisition would provide us with a fantastic opportunity to invest and do more,” said Haas. “Now that we are in the [IPO]I feel very good about our prospects.”

Son also underestimated how expensive it can be to deliver semiconductor innovation, even though Arm doesn’t make its own silicon. Arm’s costs rose from $716 million in 2015 to $1.6 billion in 2019, according to data from SoftBank. Revenue rose 20 percent to $1.9 billion, while profit plummeted nearly 70 percent to $276 million in 2019.

Line bar chart showing revenues and costs in millions of dollars along with a line showing the number of employees '000

Arm has recently started to correct course, investing more heavily in the growing server and PC market over the past four years, gaining allies such as Amazon Web Services, now in the third generation of its Arm-based Graviton chip, and Apple. , which is switching its entire range of Mac computers from Intel to its own M1 processors, was based on Arm’s designs.

Haas admitted: “While IoT is still a very important area for us, we are very, very focused on the computer space,” he said, referring to chips for servers and PCs. He declined to reveal how much of Arm’s revenue came from areas outside of its core mobile business, citing the “heavy regulatory process” surrounding the Nvidia deal.

Arm executives argue that they are only now beginning to reap the rewards of strategic investments made several years ago. Arm’s chip designs are licensed to semiconductor companies and electronics manufacturers as they begin to develop new products; It may take several years for the initial profits from the design to translate into royalties from product sales.

The company’s royalty income, which accounts for more than half of its total sales, has risen 22 percent in the last nine months, supporting Haas’s claims of a turnaround. These were “different numbers than Arm has seen before and higher than before SoftBank,” he said.

“Masa had always said that making Arm a public company one day was definitely the goal,” Haas said, adding that now that the Nvidia deal had fallen through, Arm was “back to the original Plan A.”


#techFT brings you news, commentary and analysis on the big companies, technologies and issues that shape this fast-moving sector of specialists from around the world. Click here to receive #techFT in your inbox.

Leave a Reply

Your email address will not be published.