Why would Nvidia want to pay tens of billions for Arm in the biggest chip deal in history and why now? First, Nvidia has attained substantially more financial firepower, which gives it the resources to do so. It has thrived this year as demand for its products that serve the data-center and video-gaming markets have soared amid the pandemic. As a result, the company’s share price has more than doubled, making it 2020’s second-best performing stock in the entire S&P 500 Index after Carrier Global Corp. With Nvidia’s market value now at about $300 billion — roughly 50% more than the previous semiconductor king Intel Corp. — the stock part of any transaction is more palatable.
Nvidia also may be trying to take advantage of SoftBank’s string of turmoil and strategy changes, which has included a large-scale disposal of assets amid pressure from investors as well as controversy over its recent aggressive derivative bets in the public markets. And now, the Japan-based conglomerate’s very structure may be in a state of flux as well. This weekend, the Financial Times reported SoftBank’s executives have revived discussions about taking the company private, following frustration over the discount in its market value versus the aggregate value of its holdings.
As to the rationale of the deal, an Nvidia-Arm combination has the potential to remake the entire semiconductor industry over the next decade. Nvidia has always pitched itself as a chip company built for the big ideas of the future, a provider of the key components required for all the large high-growth opportunities such as cloud-computing, artificial intelligence, robotic automation, mobile computing and internet of things — a vast array of millions of internet-connected sensors and devices.
While Nvidia’s graphics chips are deemed essential for a large swath of those applications, the company has a huge hole in its lineup. The company lacks a general-purpose processor chip, or central processing unit (CPU), that serves at the main conductor in all computing devices. Every time it sells a graphics processor chip, its customers need to use another company’s CPU. To illustrate, earlier this year when Nvidia launched its latest artificial intelligence computer system, it was forced to use the server chip made by rival Advanced Micro Devices Inc. because it didn’t have a competitive offering.
With the Arm acquisition, which provides chip designs and technologies for general processors, Nvidia would gain the ability to own the whole chip stack, across mobile phones, computers and cloud-computing data centers. Nvidia CEO Jensen Huang has already hinted at Arm’s potential. He swooned over Arm’s chip architecture prospects, saying it was “incredibly energy efficient” and has “the headwind to scale into very high performance levels over time” during an investor conference call last month.
The acquisition will also help Nvidia extend its technologies into new markets. Sure, the company could have just licensed an Arm chip design like everyone else, but by owning the asset it can shape its future. For example, it can leverage Arm’s dominant position in the smartphone market and nascent Arm-based server processors by incorporating the use of its own software and programming interfaces into Arm’s designs, increasing the likelihood other vendors will standardize on Nvidia’s chips and technologies.