By Lawrence B. Landman*
Many observers were surprised when the United Kingdom’s Competition and Markets Authority (“CMA”) recently prohibited Microsoft’s proposed $69 billion acquisition of Activision. But they should have seen this coming. After the CMA issued its provisional report1 it said that it saw no competition problems in the game console market, but still saw competitive problems in the cloud-gaming market.
The CMA saw competitive problems, not in the market for currently existing cloud-gaming products, but in the market for future cloud-gaming products. In other words, the CMA saw competitive problems, not in the current market, but in the Future Market, the market for products which do not exist yet.
Further, the CMA acted although Activision would not itself participate in the Future Market. Activision does not, and will not, sell cloud-gaming services. It sells games other companies include, and will include, in their cloud-gaming offerings. Yet the CMA forbid the transaction. The CMA found that the transaction was likely to vertically foreclose competition in the future cloud-gaming market.
In Protecting Competition to Innovate is Protecting Competition in Future Markets: Ten law review articles leave no doubt2 I provide an overview of the ten law review articles I have written which examine how the antitrust authorities in the U.S. and Europe (including the CMA) protect competition in Future Markets, markets for products which do not exist yet. Observers sometimes call these nascent markets, and indeed the CMA called the cloud-gaming market nascent.3 In that article I describe, in summary form, the Future Markets Model, the Model all antitrust authorities, in reality, always use when they decide whether they should protect competition in markets for products which do not exist yet.
A quick application of the Future Markets Model shows clearly that the CMA applied the Model to the cloud-gaming market, and did so aggressively. Competitors in this market offer games participants play directly in the cloud, not on their own computers. This market is nascent, it is still developing. For example, different competitors allow participants to play games using different equipment (phones, low-end computers, Smart TVs, and so on).4 The market is also still developing, among other reasons, because the relevant technology, such as stable, affordable, internet connections, is still developing.5
The Future Markets Model requires an authority to answer four questions, which the CMA did:
A. Does a current product exist?
B. How many firms are trying to develop a future product?
The CMA listed many other firms it described as “potential entrants or are already active in cloud-gaming to some extent.”8
C. For each possible future product, is it sufficiently developed that the authority will consider it a possible future product?
No. The CMA called the market “concentrated.” It thus found that the firms trying to develop future products were not sufficiently likely, in the foreseeable future, to apply significant future competitive pressure on Microsoft.9
D. How broad will the authority define the Future Market? Will the authority consider future products which are similar, but not identical, as future competing products?
The CMA defined the market broadly. Not only do Microsoft,10 Amazon,11 Sony,12 Boosteroid,13 and NVIDIA14 offer different products, but the products themselves are evolving.15 The CMA nevertheless recognized one broad cloud-gaming market.16
Activision however, did not, and would not, compete in the cloud-gaming market. The CMA feared that if Microsoft controlled Activision’s games, its very popular Call of Duty in particular, it would keep the games from its competitors, and thus not allow them to compete effectively in the future cloud-gaming market.17
The Federal Trade Commission (“FTC”) 18 and the European Commission19 both relied on a similar vertical foreclosure theory when they challenged Illumina’s purchase of Grail. As I said while examining this transaction, competition authorities should, in the proper case, fear that a firm may indeed use a vertical foreclosure strategy to harm competition in a Future Market.20
The CMA feared that Microsoft would do just that. Microsoft, the CMA recognized, already enjoys several advantages in the cloud-gaming market including: its massive cloud infrastructure; its control of the most popular computer operating system; and its control of Xbox, one of the three main gaming consoles; and its control of popular computer games such as Minecraft and Halo. Microsoft is already the largest firm in the market.21 Given its current size and advantages, if it were also able to keep Activision’s popular games from its competitors, the CMA concluded, it would very likely be able to dominate the Future Market. Particularly given the probability that network effects would exacerbate Microsoft’s ability to dominate the Future Market,22 the CMA concluded, the risk was just too great.
The key is that the CMA acted in the face of uncertainty. It does not know what products any company will make in the future. It does not know how technology, such as the ability to offer inexpensive, stable, internet connections, will develop. It does not know how, if at all, network effects will shape the Future Market. But, in the face of this uncertainty, it acted. It acted aggressively. And it did so, not to protect current competition, but to protect future competition.
The CMA also rejected the remedy Microsoft offered, which it claimed would assure its competitors sufficient access to Activision’s games. The CMA found this remedy difficult to implement. But more importantly, the CMA felt this arrangement would distort future competition. 23 It wanted to keep the market open and thus allow innovation to flourish.
In this respect the CMA’s actions are comparable to the FTC’s actions when the American enforcer reviewed Ciba-Geigy’s merger with Sandoz, which created Novartis. The FTC required Novartis to license key gene therapy intellectual property. It too wanted to keep the market open, and thus allow innovation to flourish.24
The CMA’s action has brought attention to other authorities’ review of this transaction. In the United States, which after all is the home of both Microsoft and Activision, the FTC has sued to block this transaction.25 The FTC has also alleged that the transaction will harm competition in the future cloud-gaming market.26
But the FTC’s and CMA’s actions are different. In the U.S., no court has ever found that the enforcement agencies may protect competition in Future Markets. In 1996 the FTC acted to keep the gene therapy market open, which shows that the enforcers have in fact been protecting competition in Future Markets for many years. But to get their transaction approved — and probably also because the antitrust problems were so obvious — the relevant firms, such as Novartis, simply acquiesced; they did what the enforcers required, which was typically to license intellectual property.
The exception is Illumina/Grail. These companies challenged the FTC. The Administrative Law Judge (“ALJ”) found that because he could not know with sufficient certainty what products (cancer screening tests) would exist in the future he would not block the transaction.27 The ALJ thus answered “No” to the Future Markets Model’s question C. But the FTC, accepting more uncertainty than did the ALJ, answered “Yes” to this question and prohibited the transaction.28
Illumina is now challenging the FTC’s decision.29 The Court of Appeals may accept the FTC’s reasoning, and, despite the uncertainty, act to protect competition in the Future Market. But the Appeals Court may, like the ALJ, find that because it cannot know with sufficient certainty what products may exist in the future, it too will not block the transaction. Or, saying something slightly differently, the Fifth Circuit may say that the law requires a level of certainty which the FTC has not shown.30
On the other hand, it may not matter what the American courts say. First, regarding Microsoft/Activision, if Microsoft loses its expected appeal, then the CMA will have blocked this transaction. If so, then the FTC’s efforts to do the same thing — and its legal authority to do so — will, at least in this case, be irrelevant.
Similarly, it may not matter if the Court of Appeals allows Illumina to buy Grail. The European Commission has already said in cannot.31 If Illumina loses its various challenges to this ruling, then, even if it prevails in the American court, it still will not be able to devour its prey.
And this shows the true significance of Microsoft/Activision. Not only does the CMA have the authority to protect competition in Future Markets, but it will do so — aggressively. It will, in the future, aggressively protect future competition.
And in Illumina/Grail the European Commission has shown that it too will do this. Thus, any transaction which either the European Commission or the CMA reviews will have to meet the exacting standards of both of these authorities.
In one very important way this is not new. Competition authorities on one side of the Atlantic have been, in effect, protecting competition in the market on the other side of the Atlantic for decades. In Ciba-Geigy/Sandoz, to pick one of many possible examples, the FTC kept the European gene therapy market open. In 1996 no one complained that the FTC’s decision infringed on European sovereignty.
What is new is that thanks to Brexit Europe now has two competition authorities. And the new one has shown that it will be at least as aggressive as the older one. Boris Johnson and his friends may not have campaigned to give the world another aggressive competition authority, but that is the fruit of their efforts.
* B.A., Stony Brook University; Juris Doctor, University of California, Berkeley; M.B.A., Columbia University; Ph.D., Roskilde University, Denmark. Partner, The Interagan Technology Group. The author thanks his son Ben for explaining the computer game market and thus showing that playing computer games can, indeed, be a form of work.
1 CMA, Anticipated acquisition by Microsoft of Activision Blizzard, Inc, Provisional findings report, Feb. 8, 2023.
2 CPI Antitrust Chronicle, Feb. 2023, Vol. 2(2), pp. 55-60.
3 CMA, Anticipated acquisition by Microsoft of Activision Blizzard, Inc, Final report, April 26, 2023, para. 4.32.
4 Id. para. 5.82.
5 Id. paras. 6, 5.97 and 7.73.
6 See infra notes 21-22 and accompanying text.
7 Id. para. 8.416.
8 CMA, supra note 3, paras. 8.81 and 8.82.
9 Id. paras. 8.399 and 8.422.
10 Id. para. 8.79(c). See also infra note 21.
11 Id. paras. 8.79(b) and 8.417.
12 Id. paras. 8.79(d) and 8.419.
13 Id. paras. 8.79(e) and 8.420.
14 Id. paras. 8.79(a) and 8.418.
15 Id. para. 4.35.
16 Id. para. 5.97.
17 Id. e.g. paras. 8.434 to 8.441.
18 In the Matter of Illumina and Grail, Opinion of the Commission, FTC Docket No. 9401 (March 31, 2023).
19 European Commission Press Release, Commission prohibits acquisition of GRAIL by Illumina, Sept. 6, 2022. (As of this writing the Commission has not made public its full decision.)
20 Lawrence B. Landman, Nascent Competition and Transnational Jurisdiction: the future markets model explains the authorities’ actions, 43 E.C.L.R. 294, 302 (2022), Lawrence B. Landman, Competition to Innovate and Future Potential Competition, 103 Journal of the Patent and Trademark Office Society 177, 199-203 (2023).
21 Id. e.g. paras. 8.219-8.224.
22 Id. e.g. paras. 8.342 and 8.384.
23 Id. paras. 11.129 to 11.132.
24 Lawrence B. Landman, Did Congress Actually Create Innovation Markets? 13 Berkeley Tech. L.J. 721, 787-794 (1998).
25 Complaint, In re Microsoft and Activision., FTC Docket No. 9412 (Dec. 8, 2022).
26 Id. e.g. paras. 39-42 and 73-95.
27 Initial Decision, In re Illumina Inc., Docket No. 9401 (F.T.C. Sept. 9, 2022), p. 176. See also Lawrence B. Landman, Competition to Innovate and Future Potential Competition, 103 Journal of the Patent and Trademark Office Society 177, 202 (2023).
28 See supra note 18.
29 Illumina and Grail v. FTC. No. 23-60167, (Fifth Circuit, Filed Ap. 5, 2023).
30 See Lawrence B. Landman, Competition to Innovate and Future Potential Competition, 103 Journal of the Patent and Trademark Office Society, 177 (2023).
31 See supra note 19.