U.S. appeals court finally decides Microsoft legitimately acquired Activision Blizzard: FTC loses by ultrawide margin

What the appeals court had to decide (and what not)

For those who follow this matter, but are not already familiar with how litigation works, some basic and simplified explanations:

A U.S. appeals court does not repeat an entire trial with witnesses. That job has already been done by the court below. The appeals court looks at whether the appeals court made any mistakes. That does not mean that only because the appeals court views a question differently than the lower court, it uses its authority to overrule. Depending on the type of issue, the court below will be afforded some deference, meaning that it may take a rather strong disagreement for the appeals court to overrule. Simply put, if a question is very much about the facts, the lower court gets more deference (only “clear error” will be corrected), and if it’s not about the facts but generally about how to interpret the law, then there is no deference at all (“de novo” meaning that the appeals court addresses those questions as if it was the first court to do so). Therefore, appellants (parties appealing a decision) will try to portray as a question of law whatever they can: it is easier to prevail on appeal if the appeals court simply makes its own decision as opposed to saying that what the lower court did may or may not have been correct, but at any rate, it wasn’t clearly erroneous, so it stands.

In the ABK case, the FTC (under its previous Democratic leadership) had an additional reason for prioritizing pure questions of law: it wanted to make it easier for itself to win future merger cases in the Ninth Circuit, a very powerful appeals court to which all appeals from cases on the U.S. West Coast (and from some other places such as Hawaii) go. The number one priority for the FTC was to lower the standard it has to meet when it tries to block a merger in terms of whether the need to raise a “serious question” (about a substantial lessening of competition in one or more relevant markets) was more like “an interesting question that deserves a closer look” or means that the court must feel that the FTC will ultimately (after a full trial with everything it involves) win its case against the merger. A PI proceeding is never about a final decision, and only about an initial assessment.

Another rather high priority for the FTC was to obtain a decision according to which any remedial measures by merging parties, such as Microsoft’s offer of a 10-year Call of Duty deal to Sony (which Sony, in fact, ultimately accepted) or agreements with Nintendo and various cloud gaming providers, should matter in the PI analysis.

The FTC achieved none of that as will be explained more specifically further below. The goal posts remain in their place for the next merger challenge that the FTC will bring.

Now, what is important to understand is that neither the district court nor the appeals court made a definitive decision on whether or not the merger was anticompetitive. The reason for which that question was not on the agenda is simply the procedural context:

Even though the PI would have practically killed the merger in all likelihood (as Microsoft and especially ABK couldn’t have accepted protracted uncertainty), the FTC was “only” asking for some more time (just a few years, which is an eternity for business, especially in a fast-placed industry like games) to conduct its in-house trial, make a decision and (if it had been a blocking decision) defend it on appeal, all of that without the merger closing in the meantime.

So the question back in 2023 was only practically, but not technically, a thumbs up or thumbs down for the deal. Judge Corley had to decide whether the FTC deserved more time. One should say “even” more time, because when she heard the PI motion, almost a year and a half had passed since the announcement of the merger, and the FTC started investigating shortly thereafter. So if the FTC didn’t have a solid case in place then, when would it ever have one?

Theoretically anything could happen, but the result in this case is pretty devastating for the FTC’s case. Even though all the decision formally says is that Judge Corley was not wrong to be unconvinced of the FTC’s arguments, it says it so loud and clear that even if the FTC held a multi-week trial and did the best job it possibly could (with all the lessions it can learn from today’s decision), its “case” would most likely never impress those judges. The law is not that the government can arbitrarily block any merger just because it doesn’t like it. Normally companies can buy and sell as they want. There must be some profound competition concerns to block such deals.

Judge Corley got the legal standard for a merger PI right

The appeals court recalls that Judge Corley’s conclusion was that the FTC had “not raised serious questions regarding whether the proposed merger is likely to substantially lessen competition in the console, library subscription services, or cloud gaming markets.” What the FTC then told the appeals court was that the judge had made it too hard for the agency to win.

Even the FTC did not dispute that Judge Corley had laid out the correct legal standard at the start of her decision, which was that the FTC needed to “raise[] questions going to the merits so serious, substantial, difficult and doubtful as to make them fair ground for thorough investigation, study, deliberation and determination by the FTC in the first instance and ultimately by the Court of Appeals.”

The FTC claimed that Judge Corley was too demanding and wanted the FTC to show not just a reasonable likelihood of success, but that competition “would probably be substantially lessened” by the deal. What’s the difference? It’s a question of degree: if something is reasonably likely to happen, then that’s still much less than saying it probably will happen. Here, the correct standard is that the questions must be serious in terms of there being a reasonable probability that the FTC will win when all is said and done.

The appeals court doesn’t agree with the FTC that Judge Corley departed from the correct standard as laid out at the start and again at the end of her decision. The FTC tried to make that argument just based on the fact that Judge Corley didn’t repeat the complete definition of the standard each and every time throughout her decision. The appellate judges applied common sense. They took into account that the decision had to be made under enormous time constraints, and that the answer is not in “flyspeck, out-of-context, isolated phrases.” A reasonable reader can see, at the beginning and at the end, that Judge Corley knew what she had to do and what she was doing.

The FTC tried something else (which equally failed). It wanted the appeals court to hold that a district court faced with a merger PI motion must side with the FTC on any question of fact where there is conflicting evidence. In other words, even if a party that wants to complete a merger puts forward very strong evidence supporting its defenses, the FTC wants to win only if it shows some evidence (albeit flimsy) to the contrary. Under that standard, the FTC would automatically win most cases. Mergers of the kind that the FTC reviews are large transactions involving reasonably large companies. There would always be some conflicting evidence. The Ninth Circuit rejects the FTC’s proposal that conflicting evidence would mean that there is, by definition, a “serious question” of the kind that justifies a merger-blocking PI.

Those PI decisions are, as the Ninth Circuit notes, based on preliminary assessments of the FTC’s case. It’s not about final decisions, it’s not about conclusive evidence of the kind that would be required later on, but that doesn’t mean the district court can’t also assess facts in a way that is appropriate at that stage of proceeding.

There is a sentence at the end of the part on the legal standard that is difficult to understand without some litigation-related experience, but those who do have such experience will find it interesting:

“The FTC’s proposed construe-everything-my-way standard is more suited for defending against a summary-judgment dismissal of claims than it is for obtaining provisional affirmative injunctive relief.”

Here’s an attempt to explain it in simpler terms. What the Ninth Circuit means is that the FTC wants a ridiculously low hurdle for getting a PI against a merger. It wants a cakewalk. A “summary-judgment dismissal of claims” is when a U.S. court throws out a case, or part(s) of a case, at a relatively early stage, finding that the case is too weak to go to trial. In other words, the case is a joke, a waste of time, so the judge throws it out at an early stage. The hurdle to defend against that is low because one doesn’t have to show that the case has merit: one just has to show that it’s not laughably meritless. To do that, it’s typically sufficient to show that there is some evidence somewhere in the record that could potentially support a case. As the appeals court accurately notes, the FTC was advocating a legal standard for a merger PI that is similarly low: as long as it has any evidence in its favor (which it will, as mentioned before, almost always have), it should get a PI. In the end, that would mean total governmental control, without serious checks and balances.

Remedial measures are a non-issue because the merger is a non-issue in the first place

Even at the June 2023 pre-hearing conference in Judge Corley’s courtroom, the FTC already tried another thing to make it easier to win merger PIs: it argued that anything a merging party does in order to address actual or hypothetical competition concerns should be ignored in the PI context.

In this case, Microsoft had done a lot. It did a deal with Nintendo in late 2023 (which was finalized a few months later), it reached agreements with various cloud gaming providers, and above all, there was an offer on the table all the time for the loudest complainant, Sony Interactive Entertainment, to get a 10-year Call of Duty deal. Sony had argued that Microsoft would (ab)use CoD to monopolize the video game console market. Beyond denying that allegation, Microsoft offered a formal contract that would make it impossible for what is an eternity in that business: 10 years. Sony still had a couple of years left under an old contract with ABK anyway, but this way it would have had long-term legal certainty and peace of mind. Sony, however, wanted to keep complaining in hopes of blocking the deal, and accepted Microsoft’s offer only after it saw that the merger was going to happen anyway. At that point, the FTC had lost in Judge Corley’s court and Microsoft had worked out a solution with the UK Competition & Markets Authority (CMA), which temporarily blocked the deal but ultimately allowed it to close.

From a policy point of view, the FTC’s position that such procompetitive deals should be ignored when a district court decides on whether to grant a PI raises serious issues. The only argument that the FTC had for such artifical ignorance was that only the FTC should have the right to determine remedies. It said that it had the expertise to do that as part of its in-house trial (before a judge who is an FTC employee), but district courts shouldn’t form opinions. The first policy issue with this proposition is that it’s actually a good thing if parties take proactive measures to address concerns (whether or they believe those concerns are valid). Anything procompetitive should be welcomed by a competition authority. The second issue is that if such remedial measures by parties solve problems, regulatory agencies can use their limited resources to deal with problems that actually need to be solved through further administrative proceedings. No one stands to gain, other than the FTC getting more power over mergers by being able to block (and practically kill) them, if good things don’t happen because they don’t count if they happen.

The Ninth Circuit didn’t decide on this because it didn’t have to. The question of whether Microsoft’s remedial measures such as the offer to Sony were correctly considered by Judge Corley or should have been ignored turned out irrelevant or, as courts call it, wasn’t reached. In order for there to be a need for a remedy, there must be something to be cured in the first place. If everything’s great, nothing must be cured and remedies don’t matter.

It’s actually a disaster for the FTC that this point wasn’t reached. It means that all the things Microsoft voluntarily and proactively did were, in the appeals court’s opinion, not necessary to fend off the FTC’s PI motion. Where there is no problem there is no need whatsoever for a solution.

No meritorious case for the FTC in any of the three relevant markets

The FTC’s mistaken focus in this case has been on the video game console market from the outset. To be fair, the FTC’s proceedings work a little differently from those of competition regulators in some other jurisdictions, so its peers like the European Commission’s Directorate-General for Competition (DG COMP) or the UK CMA were able to refocus their cases on cloud gaming at a time when the FTC had already made its bed and was forced to lie in it. And that bed was and still is console-centric. Cloud gaming was an afterthought. All of that is so because the FTC brought its in-house administrative complaint over the deal in December 2023 with a total focus on Sony’s fears that have meanwhile been proved to be baseless at best.

At this stage, where the FTC could honorably drop the case out of respect for the Ninth Circuit decision, it should also think hard about whether there is any point in carrying on with a case that has all the wrong priorities.

In fact, the Ninth Circuit decision reflects that the FTC never focused on cloud gaming. This is not to say that there is an issue in cloud gaming: even those regulators who focused on that part cleared the deal. It took a detour in the UK, but the CMA’s CEO has meanwhile given an interview that was the next best thing to expressing regrets over having been quite so obsessed with that merger: she said the CMA doesn’t always need a seat at the table when deals like ABK are reviewed.

The FTC has a problem that is called “frontloaded proceedings.” That means a procedural framework where at an early stage a party (here, the FTC) has to make its whole case, and can’t practically amend it later. The FTC proceedings are not frontloaded at the investigative stage: Microsoft agreed with the FTC to give it enough time, and it had almost a year until it brought its administrative complaint. But the administrative proceedings are frontloaded: they start with the complaint, and the complaint, including the priorities it sets, remains key. In this case, the complaint places the emphasis on an obviously nonsensical theory. There is (as the Ninth Circuit decision also indicates) no issue in multi-game (or “game library”) subscription services or cloud gaming either. But at least those are newer markets and there’s potentially more of an opportunity for a regulator to come up with wild theories concerning those markets. Video game consoles are an established market (by the way, the Ninth Circuit decision describes the way the games business works pretty well, at a high level), and it never would have made sense for Microsoft not to offer CoD, a game that costs a lot of money to develop, for other consoles than its Xbox. The offer to Sony showed it, and now that Sony has a ten-year deal, it’s a non-issue even for contractual reasons.

The Ninth Circuit decision reflects how low a priority the cloud gaming market was for the FTC when it filed the December 2023 complaint with which it was still stuck when it asked Judge Corley for a PI. The cloud gaming-specific part of the 39-page decision spans only a couple of pages.

With respect to video game consoles, the Ninth Circuit finds no fault in Judge Corley’s conclusion that Microsoft lacked an incentive to foreclose Sony. And rightly so: by now it’s clear, even regardless of the 10-year CoD deal Microsoft signed with Sony, that this acquisition never was about vertical foreclosure in terms of withholding ABK games from Sony’s PlayStation. Since the closing of the deal, Microsoft has actually broadened the availability of its games on various platforms (October 13, 2024 games fray article). The Ninth Circuit decision shows that some of what happened after Judge Corley’s decision, such as the CMA’s approval of the deal, was known to the judges. But none of that was needed to find that the FTC never had a strong case in the first place.

FTC officials may see a straw in the fact that the Ninth Circuit, at various points, just finds that Judge Corley didn’t clearly err or abuse her discretion by finding that the FTC’s evidence had shortcomings. Maybe some people at the FTC now believe that they can put on a stronger case at the time of their in-house trial than at the June 2023 PI hearing. Obviously, they can try to do a better job. They can try to learn from the Ninth Circuit decision. But they are still stuck with their December 2023 complaint and the evidentiary record they built around that one. Microsoft still has a strong argument that foreclosure wouldn’t make sense for it in any of those three markets. The FTC can do what it wants, but if it wants Microsoft to divest (such as through a spin-off) ABK, the decision will be appealed. The angle from which the appeals court will look at it will be a different one. But all in all, it’s clear that the FTC’s “case” has glaring holes. The hurdle to get a vertical merger blocked is very high.

The Trump Administration’s approach to tech regulation and what it means for this merger

It’s impossible to predict what the new FTC, with its Republican majority, will do. But based on what senior Trump Administration officials have recently said about the Make America Great Again (MAGA) and America First approach to antitrust enforcement, it’s hard to see how continued litigation over this merger is a good use of government resources. The Department of Government Efficiency (DOGE) won’t come in and intervene. But a waste of resources is still a waste of resources.

The Trump Administration has made it abundantly clear that it distances itself from a liberal anti-business agenda, yet wants to avoid that technology companies use their monopoly power to restrict freedoms, particularly the freedom of speech.

That is not an issue in the ABK context. None of this has anything to do with censorship. Unreasonable mobile app review guidelines that are arbitrarily applied may raise such concerns. Search engines with their autocomplete suggestions could be (and may in fact have been) used to advance left-wing political objectives. But whether or not Microsoft owns ABK has nothing to do with any of that.

Today’s decision doesn’t talk about it, but it is well-documented how the Biden-Khan FTC pursued the opposite of an America First policy in 2023. During the PI hearing Judge Corley’s court, there were situations where FTC litigators acted as if they were (to be clear: they were not, but their ideological agenda led them act as if they were) paid by Sony to extract concessions and get Sony a better deal than the one that was on offer at the time (and apparently good enought hat Sony ultimately took it). It was bizarre. There was the U.S. government going against U.S. company buying another U.S. company and asking questions that, whether intended or (probably) not, were tantamount to pressuring Microsoft to make promises to Sony for which there was no legal basis.

It would be consistent with an America First policy to drop a case that was originally brought over concerns that didn’t age well. Not one of the Biden-Khan FTC’s fears has been validated to the slightest extent. There is arguably more competition now than there used to be, and undoubtedly more consumer choice.

Why it took the Ninth Circuit so long to decide will remain a mystery, normally forever as court deliberations are secret. They may have had other things to do. The quality of the decision they put out is very high. The judges thought it all through, and distilled a complex case into a relatively short appellate opinion that has opened the door to an exit for the new FTC. The previous FTC was, as the opinion shows, lightyears away from obtaining a PI. It is now even further away from meeting the standard for a forcible divestiture.

No one can blame the FTC for drawing the appropriate conclusions from the clarity of today’s Ninth Circuit decision.

The document

Finally, here’s the ruling: