Antitrust Matters Episode 13: Epic v. Apple Ninth Cir. Appeal: Reactions to and Analysis of the Decision
Antitrust Matters provides engaging and timely conversations about competition policy in the digital age. Antitrust has always mattered to consumers and businesses, and to antitrust lawyers and economists, but today it also is in the political and public discourse more than ever. From the prices we pay for food, travel, financial services, payments to the way we interact daily using digital apps and platforms, antitrust touches each and every one of us in ways we may not even realize. Antitrust Matters brings you you perspectives of experts and visionaries in the field who discuss where antitrust law has been, where it is going and why it is so important to our current political discourse.
In this episode of Antitrust Matters, Matt Cantor, Ankur Kapoor, and David Golden analyze and react to the April 24, 2023, decision of the Ninth Circuit Court of Appeal in the Epic Games v. Apple case. This episode is a follow-up podcast to Episode 9, Epic v. Apple Ninth Cir. Appeal: Reactions to and Analysis of the Oral Argument.
SUBSCRIBE TO OUR PODCAST
Episode Transcript and Show Notes:
Jeff Shinder:
Welcome to Antitrust Matters, a Constantine Cannon podcast where we have engaging and timely conversations about competition policy in the digital age. My name is Jeff Shinder and I’ll be your host. Antitrust has always mattered to consumers and businesses, but today it is also in the public discourse more than ever. From how we get our food on our plates, to how we travel, to the way we interact daily using digital apps and platforms, antitrust touches each and every one of us in ways we may not even realize. In Antitrust Matters, we bring you perspectives of experts and visionaries in the field who discuss where antitrust law has been, where it is going, and why it matters today more than ever before.
David Golden:
Welcome back, everyone. I’m David Golden, a partner in Constantine Cannon’s antitrust practice. We’re here to discuss the Ninth Circuits April 24th decision in the Epic Games, the Apple case. We’ll be discussing a few of the issues raised in this decision, but I first wanted to introduce again, two of my partners, Ankur Kapoor and Matt Cantor. Welcome, gentlemen.
When we last discussed this case, it was in November ’22 just before the oral argument, and that was in episode nine of the podcast. Now that the Ninth Circuit has spoken, we are recording this episode to follow-up. So let’s get into it. Matt, there was a lot in the decision about the topic of market definition in the Ninth Circuit. Can you summarize what the circuit held and how did that analysis impact the court’s substantive antitrust analysis?
Matt Cantor:
Sure. The Ninth Circuit obviously reviewed the district court’s finding on market definition, and of course, the district court found that Apple had not violated section two or section one of the Sherman Act. And a good reason why it said that that did not occur was because the market pursued by the plaintiff there, Epic Games, was not cognizable.
As you may recall, for those who listened into the previous broadcast that Ankur, Dave and I did on this case, the plaintiff there tried to prove a so-called single brand of market, that there were markets, for example, for app stores and in-app payments that were solely predicated on the iOS operating system.
Interestingly, the reason why that was pursued and the Ninth Circuit references a statistic that the district court did in that regard, is that there was a factual finding that Apple only makes up for 15% of the smartphones worldwide, and therefore 15% of the smartphone operating systems. In my view, that’s just wrong.
I have never seen statistics that low. I’ve seen statistics that are much higher. And in the United States, the statistics are that Apple has well over half of the smartphones. But nonetheless, that was the factual finding that was made, and it was the cause that Apple’s market share was deemed to be so low in operating systems that the plaintiff felt compelled to sue on a single brand market theory.
And a single brand market, just so everyone understands, is a market, just as I said, based only on that brand. In other words, in and of itself, it is a market. That generally happens when there’s some type of unique product that’s involved. And when you’re talking about an aftermarket theory, which this is talking about, this is talking about iOS is the for market or the principle market, if you will, but market powers being leveraged according to the plaintiff in this case, in the aftermarket, which is the iOS App Store and the iOS in-app payment facilitation. Those are derivatives, if you will, of iOS.
So, the plaintiff down below said, “Hey, there’s this iOS App Store market,” for example. And the court said, “No, no, no, that’s not true. There’s a broader market for all types of gaming services” effectively. In order to prove that there are these iOS aftermarkets, you have to show that there’s some kind of lock in effect, that people really didn’t appreciate the fact that they weren’t going to be able to use, for example, another app store when they purchased an iPhone. And basically the Circuit Court said, “Well, there’s a dearth of evidence on that.”
In fact, the district court below said there was no consumer survey evidence or anything of the like that demonstrated that consumers were unaware of this fact and therefore were locked into the iOS ecosystem. That’s the principle thing that you have to prove when you’re trying to prove an aftermarket theory under the Supreme Court’s opinion in Kodak v. Image Tech.
But the thing that’s really curious here is that the majority opinion in this case, because there was two opinions, there was a majority opinion authored by Judge Smith, George W. Bush appointee, and there was a concurrence ordered by Judge Sidney Thomas, who is a Clinton appointee. And basically the majority opinion, which was joined by a district judge sitting by designation, said she got the market definition stuff wrong, particularly because she should have given more analysis on whether or not consumers were locked in and determined whether or not there was a single brand market. But it doesn’t matter.
It doesn’t matter because, under the rule of reason, the court made enough findings to demonstrate that Apple conduct was not anti-competitive. Well, Judge Thomas said, “Well, I agree with the majority. I agree with the majority that there was errors in the market definition analysis.” But that’s where you start under the rule of reason. And if you sort of mess that up, then how can you say that the rule of reason analysis that was done was somehow kosher, that it was done correctly? You can’t do that.
So, while the Ninth Circuit affirmed the district court on the section one and section two analysis, Judge Thomas would’ve reversed on the rule of reason finding in particular under section one because he said that the entire analysis had to be wrong if the market definition analysis was wrong as well. So, that is something that is likely to be thought in this litigation as it goes forward.
David Golden:
So, Ankur, Matt just mentioned the courts section one analysis under the rule of reason. Section one analysis under the rule of reason. Can you take us through what the analysis normally entails?
Ankur Kapoor:
Sure. So the classic formulation of the rule of reason dates back to the Chicago Board of Trade case in, I think, 1912. So it’s well over 100 years that antitrust law has been using this formulation. First, the plaintiff has to show that there are any competitive effects from the alleged restrain of trade. And the analysis is similar under section two of the Sherman Act, where the plaintiff has to show monopolization, but in terms of proving whether the conduct is anti-competitive, it’s the same analysis. The plaintiff has to come forward with a showing of anti-competitive effects.
Then the burden shifts to the defendant to come forward with proof of pro-competitive effects. Because the defendant being the business that put in place the challenge restraints, it knows why it did it, it has the proof for whether it’s beneficial, and so the burden naturally rests on the defendant to come forward with these pro-competitive effects. If the defendant does that, then the plaintiff has two options.
One, the plaintiff can challenge those pro-competitive effects as pretextual. Either the evidence just doesn’t support them, or the defendant is only making this argument up for litigation purposes. Second, the plaintiff can show, well, those pro-competitive effects are all well and good, but it wasn’t necessary to implement the restraint in order to achieve those pro-competitive effects. The defendant could have done, could have implemented a number of alternatives that were less restrictive of competition. So that’s the classic formulation of rule of reason inquiry.
In terms of where the dissent and the majority in the court court of appeals differed, the dissent thought, as Matt said, that precisely defining the relevant market is absolutely essential to demonstrating market power, and therefore showing any competitive effect under the first step of the rule of reason. That’s not entirely accurate.
So there are a couple of ways that a plaintiff can show any competitive effects. There’s rules called direct proof of anti-competitive effects, and there’s indirect proof of anti-competitive effects defining a market, showing the defendant’s market share, establishing market power, those are all indirect ways of showing anti-competitive effects. So let’s say Apple, in a relevant market had a 100% market share. Total monopolies. That would certainly inform the inquiry as to whether or not what Apple was doing and what it could do could exclude competition, but it’s an indirect means of showing any competitive effects.
There are also direct ways of showing any competitive effects. Principally, higher prices, lower output, and harm to innovation. The district court found, and the Ninth Circuit affirmed that finding is grounded in the evidence, that Epic had put forward sufficient proof, direct proof of anti-competitive effects. That Apple’s 30% commission was super competitive, it was in other words, substantially above Apple’s costs. Apple did have a large share in a market for mobile gaming, and that Apple was an extremely profitable company. Those are all well recognized ways of showing market power and proof of anti-competitive effects. So it’s not entirely … It’s certainly one way to show anti-competitive effects. And in that sense, the descent was correct. Look, there was no definition of the relevant market and demonstration of Apple’s market share in that market, and therefore using that methodology, Epic failed to show anti-competitive effect. But that’s incomplete. The district court found that there was direct proof, higher prices of anti-competitive effect.
The Ninth Circuit also holds, I know this is really interesting, that precisely defining irrelevant market or to be accurate, not defining irrelevant market using the accepted tests for doing so in the antitrust laws, not doing so is harmless error. Which is the first time I can recall a court of appeals ever saying … Economists have been saying it for a long time that you don’t have to engage in these economic tests to precisely delineate what the relevant market is. It includes wet cat food, but not dry cat food, which is actually a case I litigated once. You just have to show that there’s a market for something. Cat food apps and prices have gone up. You don’t have to engage in these econometric tests all the time in order to very precisely draw these market boundaries. You’re looking for the harm that the antitrust laws are intended to prevent. Higher prices, lower output, less innovation. And if you find that, it doesn’t matter that you didn’t do these market definition tests correctly.
So in the rule of reason analysis on direct evidence, principally higher prices, the district court found in the Ninth Circuit affirm that Epic had met its burden of proof. And so then the debate occurred over Apple’s pro-competitive justifications and whether or not Epic had put forward any less restrictive alternatives in order to achieve those pro-competitive effects. That’s where the case turned. Apple’s proffered pro-competitive justifications, which really weren’t disputed, or that Apple’s wall garden approach make its ecosystem much more secure than others. And Epic really didn’t have an answer to that. The less restrictive alternatives that they offered, Epic’s proof at trial didn’t fully develop how that might look. And I think the Ninth Circuit reason that Epic didn’t come forward with any evidence showing that Apple could have achieved the same levels of security using these other means that Epic had postulated. And it’s Epic’s burden to do so, so that’s where Epic failed in the rule of reason analysis.
Matt Cantor:
I want to respond to a couple things that Ankur said. I agree with most of what he said. I don’t agree with everything. He is right on the market definition generally is not always needed in a particular antitrust case, particularly in a case concerning horizontal restraints. But the Supreme Court, although economists may differ, the Supreme Court in Ohio versus Amex specifically held that in a vertical case, market definition is necessary, which was something that Judge Thomas pointed to. And why again, he felt reversal was required with respect to Epic. I agree what Ankur said exactly what the ninth Circuit held that Epic did not come back with the least restrictive alternative, or lesser restrictive alternative. Epic did though, at least with respect to the security justification, point out that Apple has something called a notarized system, which is if you download content from something other than the app store and you have an iMac, there’s a little legend that’s put down by Apple that says, “We can’t verify the security of this download.” And they said they could do the same thing on the iPhone if we were allowed to have an app store on the iPhone and people downloaded conduct from our app store. As Ankur said correctly, that that was deemed to be insufficient. I’m spellbound as to why that was deemed to be insufficient, but that was deemed to be insufficient.
Also, with respect to the intellectual property claim that Apple made, which by the way, Apple did not present a lot of evidence, saying that its security justification was supported by fact or that its IP justification was supported by fact. It just basically said, “Well, consumers like security and we have an IP right here.” That generally has not historically been enough, but the Ninth Circuit accepted that. Coming back with the way Epic Games responded to that, Epic Games with respect to the way Epic Games responded to that, Epic Games with respect to IP, said, “Well, we don’t think we should have to compensate them for it, but we’ll engage in negotiations and we’ll compensate them.” That was deemed to be insufficient because there was no specific number that was placed by Epic Games on how Apple would be compensated either by them or by others who wanted to have app stores on the iPhone. I agree with Ankur. Epic Games did not provide persuasive proof, but at least a security justification. They did provide some proof as to how a lesser restrictive alternative can be brought about based on what Apple was already doing.
On market definition, one last thing I want to say. I didn’t add this to my answer, so forgive me, but it’s important that everyone recognizes. The Ninth Circuit on the market definition principles basically set forth or reaffirmed the standard for market definition. There has to be substitution. There has to be the ability of a consumer to substitute for an alternative product in the wake of a price increase. The trick with a single brand market is you are saying that, well, those consumers cannot substitute. Here it was, consumer consumers, regular person consumers, again, not OEMs. They couldn’t substitute for other operating systems, for example, in the wake of a price increase.
That was the issue, or other app stores, in the wake of a price increase because they wouldn’t necessarily go substitute with a Google Android phone. That was the test that was laid out. That market definition is informed by substitution and the ability to substitute and what options are available to the relevant consumer. That was reaffirmed by the Epic Games case.
David Golden:
The decision was widely seen as a win for Apple, but it was not entirely left off the hook because it was… The finding of liability for violating California’s unfair competition law was affirmed. Ankur, can you describe what the court held there?
Ankur Kapoor:
Sure. This finding related to Apple’s requirement that developers use or that consumers use Apple’s system for processing in-app payments or in-app purchases, and that Apple’s rules prohibited developers from pointing consumers to other ways that they could pay for those purchases. For example, in Epic’s flagship game, Fortnite, you can buy lots of virtual goods and if you’re on iOS, if you were playing on your iPad or your iPhone and you wanted to buy one of those goods, consumers had to use and developers had to use Apple’s system for in-app payments. Epic could not tell consumers, “Hey, go to Epic’s website and you’ll be able to buy these goods for less.” Consumers could buy them for less on Epic’s website because possibility was that because Epic wouldn’t have to pay Apple’s 30% commission on the in-app purchases, it could charge consumers less for whatever good they were buying.
But Apple’s rules prohibited Epic from doing that, prohibited all developers from doing that. That’s in fact what led to the… It was the genesis of the dispute when Epic created a hot fix that enabled consumers to go around Apple’s in-app payment system, Apple then kicked Epic off the platform. What the district court held under California’s unfair competition law was Apple’s prohibitions were unfair under the California law and the Ninth Circuit affirm.
My view of it, both the district court’s ruling and the Ninth Circuit deferments, was they really didn’t see a pro-competitive justification for Apple’s restrictions. In fact, the district court specifically noted that other payment services, for example, PayPal or Google Pay or Apple Pay itself would offer better secure… There are number of sophisticated payment systems that have been around for decades. They’ve established a number of security protocols in order to make online or in-app purchases secure. The district court held that some of those competing providers could actually provide better security than Apple. I think that motivated the district court to want to do something about the IAP restrictions. They couldn’t do it under the Sherman Act, but the more flexible California unfair competition law enabled the district court to find that to be on balance an unfair [inaudible 00:20:43] of competition. The Ninth Circuit affirmed, I think without much additional discussion, compared to the district court.
David Golden:
Matt, final question to you, and Ankur for you after Matt, is what are the next steps in Epic v. Apple? Where do we go from here? Then, also, what do you think will be the response, if any, by government enforcers or legislators to the Ninth Circuit’s opinion?
Matt Cantor:
Yeah, but Ankur did a great job in responding to the last question. The only thing I wanted to add to it is I never could figure out why the unfair competition case that was brought by Epic was only brought with respect to these steering restrictions, restrictions that precluded Epic from saying, “Hey, you can go somewhere else to buy our product,” and why it didn’t relate to the gamut of Apple’s conduct. I couldn’t find why that was the case in either the district court or the circuit court opinions. I find that to be curious.
But in any event, there’s two things I’d wanted to say on the question that you stated, Dave, or the questions. One, there is going to be a petition for rehearing on bonk. I agree with you that I think this has generally been perceived as a win for Apple, although I think Apple is an aggressive litigator and is going to certainly fight on the 17 200 aspect, and the fact that there was a split in the circuit does raise the possibility of an on bonk petition being taken. You’re talking about the Ninth Circuit. That is very, very possible.
That’s going to be interesting. I’m not going to predict whether they’ll take the petition or whether the petition would be ultimately successful, but I don’t think this is a petition that will be dismissed out of hand. I think it’ll be considered, and I wouldn’t be surprised if there were some votes on the Ninth Circuit bench for taking the matter on rehearing.
With respect to government enforcers and legislators, look, I think there’s two things to say. One is that there’s been a lot in the last few years, particularly during the first two years of the Biden administration, there was a lot of talk about revising the antitrust laws because particularly when it comes to tech players, the antitrust laws may not have all the quivers that are necessary to adequately enforce the law. I know that there are many people who disagree with that. That was nonetheless the mantra that you heard from many people, including Senator Klobuchar and Representative Cicilline.
This case certainly indicates that it’s hard to get a monopolist to pursue a claim under section two of the Sherman Act against tech monopolist. One of the things that’s interesting here is that there was also a tie-in claim. Again, the Circuit Court said, “Well, this is a form of technological tie, and those types of ties really should be adjudicated under full rule of reason inquiry instead of a more shortened per se standard.” That seems to be a holding that is now almost well established if not well established. It seems very few tie-in cases against tech providers will be adjudicated under the per se standard. I think government enforcers and legislators will look at the case. Again, it’s still ongoing, and say, “This is just another example of how difficult it is to…”
Another example of how difficult it is to pursue antitrust against some of the more incumbent tech “monopolists”, whether it’s Apple or Facebook, for example. And that could at some point on a legislative front spur another legislative charge with respect to enforcers, look, chairwoman Lena Kahn and the FTC, which now does not have any Republican members, only three Democratic ones have been very consistent that the FTC has quivers beyond the Sherman Act and the Clayton Act to enforce competition law.
And that is in particular section five of the FTC Act, which prohibits unfair methods of competition. I would expect that the FTC has taken note of this case, will certainly consider it in its enforcement posture, but will see this as just another reason for why it should have an expansive view of section five of the FTC Act and why it should be in proceedings under section five of the FTC Act. They really haven’t done that yet. Yeah, there was recent case against Amgen and with respect to its acquisition of Verizon that mentioned section five of the FTC, but that’s really a Clayton Act section seven merger challenge. But I just think that given the leadership at the current FTC, that’s going to add more incentive for them to bring a section five case and we’ll see if they do.
Ankur Kapoor:
Yeah, I agree with what Matt said. I think that this kind of case is a prime candidate for the FTC to bring under section five. I think also in terms of enforcement beyond this case, the law is pretty well established that if you’re pursuing an antitrust claim under the Sherman Act based on a single brand market, or Apple in this case is a monopolist, even in a market that consists entirely of Apple’s own products. Or think of Tesla thinking of cars, right? Tesla is a single market because it’s maybe a dominant provider of electric cars. Under current antitrust of the United States, the only way for that kind of claim to succeed is if the plaintiff can show that they and others like the plaintiff were locked in by virtue of something else that the defendant did. In other words, not just that lots of consumers and developers use the Apple platform, but in this case it was that Apple had always told consumers and developers that they could only buy apps on the app store. And consumers and developers freely signed onto the platform, they bought into it, knowing that.
And under the current law, that means that there is no single market claim. It hasn’t changed the fact that you have an entire ecosystem on which a number of developers depend for their livelihoods and a number of, a very large number of developers, very large number of consumers use the platform and spend a great deal of money on it. Whether that law should change is I think a legitimate question for debate. Right now, there’s nothing else the Ninth Circuit could have done, right? It’s constrained by the law and this Supreme Court is not going to alter that law in any way, shape, or form. So I think that’s exactly right that it’s going to fall to the FTC Act and FTC enforcement to bring these kinds of cases.
But I think we should seriously consider whether that should continue to be the law. And in Europe, for example, it’s not. In Europe, the plaintiff, whether it’s the European Commission or the UK competition markets authority or private plaintiffs do not have to show that they didn’t know about this logging. And either it arose, this theory arose, this law arose in the context of, it’s a very specific context of photocopiers. Kodak, the Kodak case, Supreme Court in 1992. In which, the plaintiffs alleged that Kodak had monopolized in the aftermarket for the service and the parts of Kodak copiers, very large, high-end copiers.
That was the cutting edge technology at the time. And the Supreme Court and subsequent cases in the court feels how that, well, look, we’re talking about very sophisticated business customers. They can evaluate not just how much it costs to buy the copier, but how much the parts are going to be, how much the service is going to be, and knowing all that if they still buy into the Kodak platform and they’re logged into it, that’s not anti-competitive, that’s voluntarily done. So they knew well what the costs were going to be. And that’s just not the case with the Apple ecosystem. Even if Apple puts the disclaimer in its developer agreements and consumer agreements, which nobody reads, that you can only buy apps on the app store and use all these restrictions, it just doesn’t change the fact that there’s this logging and it’s a serious question as to whether or not that law should be reformed to fit a 21st century market or 21st century markets more generally.
Matt Cantor:
I think that’s a great way to end. I think that Ankur raises excellent points. I also think that Ankur’s particular point contrasting the European Commission is a very interesting one because very quickly, European commissions, antitrust standards are more lenient and the Digital Markets Act is going into effect, which will also impact the ability of large technology players to take anti-consumer action. So what we’re going to end up happening is you could have substantial enforcement in the EC and changes made in the European economic community, but those changes aren’t made in the United States and American consumers and merchants and app developers are all left on the sidelines. So that would be obviously a very disjointed situation.
David Golden:
Thanks, Matt. Thanks Ankur. We’ll continue to monitor the Epic, the Apple case, and business issues in general.
A postscript to this podcast episode. Since recording, the 9th Circuit has rejected both Epic’s and Apple’s requests for rehearing. Apple has indicated that it will seek Supreme Court review.
Jeff Shinder:
That’s all for our show today. If you like the podcast, make sure to subscribe to Antitrust Matters and leave us comments on how we were doing or on the topics you would like us to cover going forward. You can also follow us on Twitter, or follow the Constantine Cannon antitrust team on LinkedIn. Until next time, be well, and remember antitrust matters.