Thread by @FOSSpatents

#Microsoft and #Activision will file their UK appeal of the @CMAgovUK blocking decision with the @CATribunal in the next approx. 90 minutes.

It can take a few days or a week--or could even take more than a week--before the CAT will publish the summaries of those appeals.

🧵1/X

https://twitter.com/diane_alter/status/1661373525372878852

There are major misconceptions out there concerning the likelihood of that appeal (I'll refer it as one, though technically they're two) succeeding.

Linklaters, one of the largest UK law firms, says the CMA has won 67%--i.e., NOT WON 33%--of all merger appeals since 2010.

🧵2/X

It's wrong to say that the CMA can simply reach the same conclusions again upon remand, thereby creating an endless loop.

legislation.gov.uk/ukpga/2002/40/…

"refer the matter back ... with a direction to reconsider and make a new decision in accordance with the ruling of the [CAT]"

🧵3/X

In this case, there are multiple ways in which the merging parties can defeat the CMA for good, making a remand either unnecessary or at least making the outcome a foregone conclusion.

For instance, if CAT says rejection of proposed remedy was irrational: GAME OVER then.

🧵4/X twitter.com/i/web/status/1…

I tweeted earlier today about a CAT judge having said publicly that the CMA should be more receptive to behavioral remedies in vertical merger cases: that is EXACTLY the remedy issue in the ABK case. See

https://twitter.com/FOSSpatents/status/1661163607864799232

🧵5/X

Remedy determination is only the last step. But there are very serious issues with the CMA's decision starting with market definition and market share. Even the aggressive class-action lawyers suing over the deal in the U.S. did not dare to allege a 70% share ("only" 40%).

🧵6/X

It will be easy for the CATribunal's judges to see that the CMA's Inquiry Group (a) did not understand the technical aspects (the President of the CAT and one other CAT judge are patent judges, so they understand technology very well) and (b) that they were totally biased.

🧵7/X

The CMA is not "immune" to appeals -- otherwise it would win 100% of the time, not 67%.

There are misconceptions regarding how high a hurdle the so-called Judicial Review standard is. Yes, it's high. No, it's not insurmountable. Let me explain:

🧵8/X

The standard is also called "Wednesbury unreasonableness":
uk.practicallaw.thomsonreuters.com/6-200-9152?tra…)

"A reasoning or decision is Wednesbury unreasonable (or irrational) if it is so unreasonable that no reasonable person acting reasonably could have made it."

What does it mean?

🧵9/X

It means that the appeals court can't just say "we believe our position makes more sense than that of the CMA." It takes something more.

But legal errors can always be overturned. And even factual determinations if they are clearly unsupported by evidence or arbitrary.

🧵10/X

In the Compare the Market case, the CAT overruled the CMA on 5 out of 6 counts. It was not a merger, and in mergers the CMA benefits from the fact that it's a lot about predicting the future as opposed to deciding on the basis of what has already happened. Still.

🧵11/X

There are similar standards in the United States and while I haven't been following UK law nearly as closely as U.S. law, it does help to compare.

The JMOL standard and the "arbitrary or capricious" standard from the U.S. are the closest points of reference.

🧵12/X

1) Judgment as a Matter of Law (JMOL): if factual determinations in the U.S. are made by a jury, the jury can only be overruled--or the jury trial avoided in the first place--"if no reasonable jury could find (or could have found) otherwise."

Sounds much like Wednesbury.

🧵13/X

So if a judge or an appeals court in the U.S. overrules a jury verdict (and juries are a constitutional cornerstone of U.S. law), the court has to say that the jury did something that no reasonable jury could've done, i.e., the jury acted unreasonably. It does happen!

🧵14/X

2) Then there's the "arbitrary or capricious" standard from which certain U.S. government agencies (NOT the FTC in a merger case though) benefit. The chances of success are 8% or so (will check on it again one of these days). Compared to that, the CMA performs poorly.

🧵15/X

The CAT apparently likes to avoid calling the CMA['s decisions] irrational by sometimes saying a finding was either against the law or irrational.

In the end, if the CAT judges believe the CMA is clearly wrong, they're absolutely able to rule accordingly.

🧵16/X

The CMA has another problem here: it has not really respected guidance from prior CAT cases. For instance, in Meta-Giphy the CAT said that anything more forward-looking than 5 years is not an "expectation": it's unpredictable. CMA cited stakeholder input 10-15 years out.

🧵17/X

The CAT also told the CMA that it needs to perform a "cross-check", i.e., look at the potentially negative effects if it is wrong. The CMA ruling does not really do that. There are negative effects such as less competition among big players in consoles, app stores.

🧵18/X

When we see the summary of the two appeals, we'll be in a better position to discuss it. I have no doubt that the appeals court will see that the CMA shoehorned previously failed theories of harm (console gaming services, ecosytem/conglomerate) into the cloud theory.

🧵19/X

The EC ruling will not be binding precedent. Psychologically and politically, it obviously does help. But the CAT can't rely on the EU finding when concluding that the CMA irrationally rejected a workable, administrable, procompetitive, consumer-friendly remedy.

🧵20/X

The CMA is in the best position to defend itself on appeal when it has MORE THAN ONE theory. Then it can say that there's a bit of concern here, a bit of concern there, and all in all there's enough concern that there's a 50%+ risk of bad things happening.

Not so here.

🧵21/X

The CMA's case here is down to one flimsy theory of harm about a nascent, currently minuscule if not obscure, market.

Yes, they enjoy wide latitude when assessing a deal's potential effects. No, PARANOIA is not rational and can be overruled.

🧵22/X

The parties have some of the best lawyers in Britain: Daniel Beard KC for Microsoft (defeated the CMA twice last year, also defeated the EC), and for Activision Blizzard the Queen's lawyer Lord Pannick KC as well as the CMA CEO's former firm (Slaughter & May).

🧵23/X

The first key thing for the CAT to decide will be timing. When will they hold a hearing? I believe CAT hearings never take place in August. Can they do it before? Or right after? We should know soon.

I'm glad the parties didn't request an extension of the filing deadline

🧵24/X

The European Commission has a lot more experience in dealing with high-stakes high-profile merger and other antitrust cases. I know that EC officials are always extremely cautious about taking on the likes of Apple with their vast resources. The CMA is experiencing it now

🧵25/X

For example, there's the Spotify complaint over Apple's 30% app tax on subscriptions that run more than a year. The EC even retracted its original Statement of Objections and narrowed its case because Apple uses an army of armies of top-notch lawyers and can win appeals.

🧵26/X

Let's be realistic: this appeal, even if the hearing was scheduled for July, won't be decided before the July 18 closing deadline under the current merger agreement.

This means they must either "close over" the CMA during the appeal or extend the merger agreement.

🧵27/X

I hope the CAT will, as courts typically do in these situations, urge the parties to agree on a settlement.

For the CMA, the institutional damage from losing this appeal (especially if it's a humiliating loss making the CMA look like fools) is not an option. Need an exit

🧵28/X

The UK government won't just overrule the CMA; it would've had to become involved earlier for that. But "the government has levers" as the CMA's own chairman said at least week's Parliament hearing. It also makes political sense for the CMA to figure out a solution.

🧵29/X

Rarely has a regulator had to defend itself on appeal from such a weak position. They claim to protect competition in cloud gaming, but the real players like @NVIDIAGFN and @Boosteroid_main want the deal to go through! It's preposterous. Who knows better, Nvidia or CMA?

🧵30/X

38 countries (or 37 if one doesn't count South Africa until the June 21 hearing) have approved the deal. Approximately 2.4 billion people and an aggregate GDP of $42 trillion. See

https://twitter.com/FOSSpatents/status/1659554348404277250

This deal will go through, be it before or after a CAT ruling.

🧵31/31