Wednesday, October 12, 2022

UK antitrust authority CMA publishes full text of Phase 1 decision on Microsoft's acquisition of Activision Blizzard: more detailed--but no more accurate--than the short version

Today the Competition & Markets Authority (CMA), the competition authority of the United Kingdom, published the full text (76 pages) of its September 1, 2022 Phase 1 decision not to grant fast-track clearance to Microsoft's acquisition of Activision Blizzard. Meanwhile, the merger review has already been in Phase 2 (a more comprehensive review) for well over a month. I don't know whether the CMA has changed its perspective on some of the questions, but the document published today is just a public redacted version of something that is six weeks old. In a merger review, six weeks is a long time. Sort of old news.

In the meantime, there have been other developments: Brazil's competition authority CADE doesn't see a need for imposing any conditions on the deal, and the CEO of a major games company--Grand Theft Auto maker Take-Two Interactive--actually welcomes the transaction because it's going to be good for the industry. (In the post on Take-Two's statements I mentioned my emotional attachment to those Blizzard franchises, including the fact that my first online article ever--which simultaneously happened to be my first published English-language article--was about how exactly the spells in Warcraft II - Tides of Darkness worked at the algorithmic level.)

When the CMA's Phase 1 decision came down on September 1, I studied the summary that was released that day and commented on it: UK antitrust authority gets basic facts wrong as it declines to approve Microsoft's purchase of Activision Blizzard on fast track: extensive review was expected, but reasoning is nonsensical

I have now read the 76-page version the CMA published today. Theoretically, there is always the possibility of a detailed ruling stating hard facts or providing useful explanations in light of which an unconvincing short-form decision all of a sudden makes sense. Not so here. It's just more words as opposed to more substance. That's why I have nothing to nuance--much less retract--that I wrote in the September 1 post I just linked to.

Paragraph 19 clarifies that the standard has meanwhile changed:

"At phase 1, the CMA needs to establish whether there is a realistic prospect of an SLC which merits a reference to an in-depth phase 2 investigation. This is a lower threshold than that used during a phase 2 investigation, which requires the CMA to conclude that a merger is likely to result in an SLC in order to prohibit a transaction." (emphases in original)

At least one of the theories of harm doesn't even meet the "realistic prospect" standard. That's the one according to which the combination of all of Microsoft's assets, with particular references to Windows and the Azure cloud, gives it an advantage that could result in a substantial lessening of competition. As I already explained in the September 1 post, Windows is simply not an input to game makers or cloud gaming services; and Azure is a general cloud service. That is a total commodity. Almost like electricity.

Activision Blizzard's games use Microsoft's competitors' cloud services--which is presumably a secondary reason why Google is lobbying against the merger

It took me only a couple of minutes to find out that Activision Blizzard isn't even using Microsoft's Azure cloud in the operation of its online games.

Google aggressively promotes its Google Cloud Platform to game makers. It points to key customer references including Activision's Call of Duty and Blizzard's World of Warcraft (WoW), but also major non-ActivisionBlizzard titles such as Fortnite, League of Legends, Dota 2, and Final Fantasy XIV.

In August 2018, King.com confirmed that it was moving its servers (for Candy Crush and its other titles) to the Google Cloud.

In April 2020, WoW players "noticed that IP addresses from Amazon's pool are popping up for world servers [servers that deliver the maps and units relevant to a particular WoW universe] now."

Google is opposing the transaction to avoid--not to protect--competition. As I noted, Google paid Activision Blizzard and other major game makers a lot of money to ensure their loyalty to the Google Play Store (Google's Android app store) when Epic Games took its first initiative (a couple of years before the "Fortnite hotfix" that activated an alternative payment system). The fact that Microsoft wants to open up game distribution and would likely be much less receptive to such proposals is almost certainly the primary reason why Google doesn't want the deal to go through.

A secondary reason may be that Google is concerned about the possibility of losing--further down the road, certainly not overnight--Activision Blizzard (including King) as a customer that generates significant revenues for the Google Cloud and some of whose products are great use cases and references that help acquire other customers from the gaming industry. However, even if a migration away from Google Cloud were to occur post-transaction, Google would still be a highly competitive cloud service provider, would still be able to point to the fact that the Google Cloud for many years satisfied Activision Blizzard King's exacting requirements, and that the servers of various other major games still run on the Google Cloud.

My own app development company has actually used AWS, Google Cloud, and Azure (in order of scale) for different backend purposes in different projects. There is nothing unique to Azure that "gives Microsoft an advantage that is difficult for rivals to replicate" (para. 238 of the CMA decision) or that suggests a substantial lessening of competition in the games business in the (unrealistic) scenario of "[d]Denying access to Microsoft's cloud platform to rival cloud gaming services providers (or offering it on worse terms, including price, location, and/or processing power)."

While the CMA's theories of harm regarding the impact on the console market and on cloud gaming services are more traditional theories than the one involving Windows and Azure, the purpose of competition enforcement is not to protect market leaders like Sony. The Brazilian decision was quite pragmatic: Microsoft doesn't even have incentives not to make future versions of Call of Duty available for the PlayStation, but even if it did and if a limited number of players migrated from the PlayStation to the Xbox, that wouldn't mean a substantial lessening of competition (oddly, it might actually mean more competition).

It's debatable whether those theories of harm were realistic, but at least there is no likelihood of a substantial lessening of competition.

For example, anyone who knows how the games business works can easily identify flaws in the part on cross-platform play, such as para. 196:

"The CMA considers that, given the high number of CoD players, high-quality matchmaking would continue to be possible even if cross-play were compromised. Cross-play would also be less important if a foreclosure strategy succeeded in attracting more gamers to Xbox (an effect that would be strengthened by the existence of strong direct network effects). In any event, the CMA considers that there are mechanisms of partial foreclosure that are compatible with cross-platform play, such as exclusive in-game items or price differences."

The number of players for any successful game--not just CoD--is more than sufficient that players would always find others to play with or against, even in the middle of the night, even without cross-play. Critical mass is a non-issue, but if players want to play with particular people, such as real-life friends, it's not a numbers game. And then the likelihood of Xbox users being disappointed that they can't play with their friends (who are more likely than not to have a PlayStation) is far greater than the other way round, so Microsoft would be shooting itself in the foot.

Also, if there is one company in the industry that wanted to limit cross-platform compatibility, that's Sony. That is a fact, and it was also mentioned during last year's Epic Games v. Apple trial.

The Verge's Tom Warren has published (on Twitter) excerpts of a Microsoft response to the CMA's Phase 1 decision:

Having read the 76-page document released today, I would agree that the CMA's Phase 1 decision "adopt[e] Sony's complaints without the appropriate level of critical review":

But the CMA has every opportunity to get the facts (and, of course, the law) right in Phase 2. I still very much like the work that other people at the CMA are doing on mobile platform issues such as the iOS browser monopoly. If that Phase 1 decision on Microsoft-ActivisionBlizzard remains an outlier, the CMA can still be a force for good in the tech sector. And at some point the agency may even find the time to question Sony's restrictive PlayStation developer terms...