Nvidia still has some homework to do on various regulatory approvals to close its $40 billion acquisition of ARM from Softbank. The acquisition needs the approval of regulators in Japan, the United Kingdom, and United States - the directly affected parties - since Softbank (owner of ARM) is a Japanese company, ARM is based in the UK, and Nvidia is based in the United States.
The approval of antitrust regulators in the European Union and China is also required due to the significant market presence in both jurisdictions. The deal was supposed to be closed by March 2022, but can be extended to September 2022. I would expect an extension due to the up-to-now slow pace of regulatory filings and the expected lengthy approval process.
In the UK, the government asked the independent Competition and Markets Authority (CMA) to determine the jurisdiction, national security, and competitive implications of the proposed acquisition. The CMA returned its findings 10 days ahead of its deadline of June 30th, whether the acquisition should be approved with or without conditions or if the acquisition warrants a more thorough investigation.
Considering that this is the CMA’s first merger review since the UK left the European Union, and it’s the CMA’s first opportunity to showcase its capabilities as a real competition watchdog, it is quite likely that the CMA recommended a more thorough investigation. Another dimension to consider is that the approval process is likely to become politicized as ARM is the crown jewel of the United Kingdom’s high-tech industry.
The British government would be under severe pressure if the sale of ARM to Nvidia were viewed as selling out British ingenuity to an American company, especially if jobs would move from the UK. It is therefore paramount for Nvidia to alleviate any concerns of the British government and public that such a brain drain could occur, and to instead demonstrate that Nvidia would strengthen and expand ARM’s operations in the United Kingdom.
For Japanese antitrust authorities, this combination is probably the easiest to approve since it would allow Japan’s Softbank to sell a division it wants to monetize to Nvidia, headquartered in the United States, a close ally of Japan. Considering the high likelihood of a Japanese approval, it is not surprising that Nvidia has not yet filed its notice to Japanese authorities that it plans to acquire ARM from Softbank.
In the United States, the FTC opened an in-depth probe into the acquisition, asking Nvidia and ARM twice for information about the deal as well as asking third parties for information. According to Reuters, Google, Microsoft, and Qualcomm among others have complained to the FTC about the deal, whereas Broadcom, MediaTek, and Marvel have expressed support of the transaction after they received assurances from Nvidia.
The FTC recently had a change in leadership when Lina Khan was elevated to FTC Chair hours after being confirmed as FTC Commissioner. Lina Khan is an outspoken critic of platform-based businesses, and it will be interesting to see if the new FTC Chair views the combination of Nvidia, the leader in cloud-based AI technology and ARM, the leader in device-based AI technology, to be the emergence of such a dominant tech platform.
The acquisition of Xilinx by AMD will be an interesting test case to see how much the new antitrust axis of Tim Wu in the White House, Lina Khan as FTC Chair, and Jonathan Kanter as head of the Antitrust Division of the Department of Justice are going to flex their muscles and bring more rigorous oversight to mergers and acquisition, as the AMD/Xilinx is a much more unproblematic combination than ARM’s acquisition by Nvidia. If AMD/Xilinx is challenged by the U.S. authorities, the Nvidia/ARM combination will definitely be challenged as well, whereas the opposite does not give us any indication of how the US antitrust authorities will react to the NVIDIA deal. The FTC and the Antitrust Division of the Department of Justice have time on their side to wait and see how the other global antitrust authorities will react before they decide whether they want to sue against the Nvidia acquisition of ARM. The companies do not need any approval to merge under US law, but it is up to the government to sue against such a combination.
Nvidia has not yet filed its documents with the European Commission. With the customary August vacation of the European Commission (EC), the earliest the European authorities will look at the documents will be in September. The EC has been acting as a more assertive antitrust force on American companies than their American counterparts.
A large reason is that the EC does not have to fear significant economic blowback since the regulated parties are not European companies who pay significant taxes or employ a large number of people in the European Union. They are easy targets for enforcements since they predominantly just profit from selling their products and services in the EU. The same will be true for Nvidia and ARM. If US antitrust authorities lean against the Nvidia acquisition of ARM, they might wait for the EU to block the merger, essentially letting it do the US government’s dirty work of scuttling it. If the EU approves the merger, then US authorities can still file against it if they have antitrust concerns.
Nvidia filed for the ARM merger approval back in March 2021. Under Chinese law, the Phase I review period takes only 30 days and Phase II lasts up to 90 days, which can be extended to a 60- day Phase III. However, the clock only starts ticking when the antitrust authorities are satisfied that they have all the information they would like to have. This makes the Chinese merger approval process extremely flexible.
As of July 2021, the Chinese antitrust authorities have not yet indicated that they have begun their review process. There are rumors that Chinese companies have been consulted as part of the preparatory work with companies like Huawei expressing their concern. Considering that Chinese antitrust investigation is part of China’s industrial policy, it is quite likely that the Chinese authorities will wait until the other antitrust authorities have come to their decision before announcing theirs.
In 2018, Softbank sold the majority of ARM China to a consortium of local companies, including China Investment Corp., the Chinese government’s sovereign investment fund. In 2020, there were conflicts between the Chinese majority owners of ARM China and the British parent company where ARM unsuccessfully tried to oust the Chinese CEO of ARM China. One can only speculate, but it wouldn’t be surprising if this conflict will negatively impact the sale of ARM to Nvidia.
We also must consider that any merger that involves western companies must be viewed through a political lens where the relationship between China, the United States, Japan, and the United Kingdom will play a role in the approval process. The tenser the relations between China and the other countries, the more likely it is for China to simply wait out the time allotted for the merger approval and scuttle the merger by being passive aggressive, similar to what happened when the Chinese antitrust authorities did not rule on the Qualcomm acquisition of Dutch-based NXP.
It has been almost one year since Nvidia signed the agreement to acquire ARM from Softbank. Nvidia has basically passed the halftime mark for the approval process, and until now has little to show for it. The company needs to gain positive momentum to get all the approvals by its September 2022 deadline.
Time is ticking, and this is especially true for Softbank. Softbank investors must view the deal with Nvidia quite critically as the S&P 500 has increased by 30% between September 13, 2020, and July 27, 2021. This increase in the general stock market is now larger than the profit Softbank would receive for buying ARM for $31 billion and selling it for $40 billion to Nvidia. The same $31 billion invested in an S&P 500 index fund would now be worth $41 billion. If investors would have invested the $31 billion on September 15, 2016, when Softbank closed the ARM acquisition in an S&P 500 index fund, their investment would now be worth about $62 billion.
Roger Entner is the founder and analyst at Recon Analytics. He received an honorary doctor of science degree from Heriot-Watt University. Recon Analytics specializes in fact-based research and the analysis of disparate data sources to provide unprecedented insights into the world of technology. Follow Roger on Twitter @rogerentner.