For those not steeped in antitrust law’s treatment of single-firm monopolization cases, under the rule-of-reason framework, a plaintiff must first demonstrate that the challenged conduct by the defendant is anticompetitive; if successful, the burden shifts to the defendant in the second or balancing stage to justify the restraints on efficiency grounds. According to research by Professor Michael Carrier, between 1999 and 2009, courts dismissed 97 percent of cases at the first stage, reaching the balancing stage in only two percent of cases.
There is a fierce debate in antitrust circles as to what constitutes a cognizable efficiency. In April, the Ninth Circuit upheld Judge Yvonne Gonzalez Rogers’ dismissal of Epic Game’s antitrust case against Apple on the flimsiest of efficiencies.
A brief recap of the case is in order, beginning with the challenged conduct. Epic alleged Apple forces certain app developers to pay monopoly rents and exclusively use its App Store, and in addition requires the use of Apple’s payment system for any in-app purchases. The use of Apple’s App Store, and the prohibition on a developer loading its own app store, as well as the required use of Apple’s payment system are set forth in several Apple contracts developers must execute to operate on Apple’s iOS. The Ninth Circuit found that Epic met its burden of demonstrating an unreasonable restraint of trade, but Epic’s case failed because Apple was able to proffer two procompetitive rationales that the Appellate Court held were non-pretextual and legally cognizable. One of those justifications was that Apple prohibited competitive app stores and required developers to only use Apple’s payment system because it was protecting its intellectual property (“IP”) rights.
Yet neither the District Court nor the Ninth Circuit ever tell us what IP Apple’s restraints are protecting. The District Court opinion states that “Apple’s R&D spending in FY 2020 was $18.8 billion,” and that Apple has created “thousands of developer tools.” But even Apple disputes in a recent submission to the European Commission that R&D has any relationship to the value of IP: “A patent’s value is traditionally measured by the value of the claimed technology, not the amount of effort expended by the patent holder in obtaining the patent, much less ‘failed investments’ that did not result in any valuable patented technology.”
Moreover, every tech platform must invest something to encourage participation by developers and users. Without the developers’ apps, however, there would be few if any device sales. If all that is required to justify exclusion of competitors, as well tying and monopolization, is the existence of some unspecified IP rights, then exclusionary conduct by tech platforms for all practical purposes becomes per se legal. Plaintiffs challenging these tech platform practices on antitrust grounds are doomed from the start. Even though the plaintiff theoretically can proffer a less restrictive alternative for the tech platform owners to monetize their IP, this alternative per the Ninth Circuit must be “virtually as effective” and “without increased cost.” Again, the deck was already stacked against plaintiffs, and this decision risks making it even less likely for abusive monopolists to be held to account.
Ignoring the Economic Literature on IP
In addition to bestowing virtual antitrust immunity on tech platforms in rule-of-reason cases, there are important reasons why IP should never qualify as a procompetitive business justification for exclusionary conduct. Had the Ninth Circuit consulted the relevant economic literature, it would have learned that IP is fundamentally not procompetitive. Indeed, there is virtually no evidence that patents and copyrights, particularly in software, incentivize or create innovation. As Professors Michele Boldrin and David Levine conclude, “there is no empirical evidence that [patents] serve to increase innovation and productivity…” This same claim could be made for the impact of copyrights as well. Academic studies find little connection between patents, copyright, and innovation. Historical analysis similarly disputes the connection. Surveys of companies further find that the goals of patenting are not primarily to stimulate innovation but instead the “prevention of rivals from patenting related inventions.” Or, in other words, the creation of barriers to entry. Innovation within individual firms is motivated much more by gaining first-mover advantages, moving quickly down the learning curve or developing superior sales and marketing in competitive markets. As Boldrin and Levine explain:
In most industries, the first-mover advantage and the competitive rents it induces are substantial without patents. The smartphone industry-laden as it is with patent litigation-is a case in point. Apple derived enormous profits in this market before it faced any substantial competition.
Possibly even more decisive for innovation are higher labor costs that result from strong unions. Other factors have also been found to be important for innovation. The government is responsible for 57 percent of all basic research, research that has been the foundation of the internet, modern agriculture, drug develop, biotech, communications and other areas. Strong research universities are the source of many more significant innovations than private firms. Professor Margaret O’Mara’s recent history of Silicon Valley demonstrates how military contracts and relationships with Stanford University were absolutely critical to the Silicon Valley success story. Her book reveals the irony of how the Silicon Valley leaders embraced libertarian ideologies while at the same time their companies were propelled forward by government contracts.
In an earlier period, the antitrust agencies ordered thousands of compulsory licensing decrees, which were estimated to have covered between 40,000 and 50,000 patents. Professor F.M. Scherer shows how these licenses did not lead to less innovation. Indeed, the availability of this technology led to significant economic advances in the United States. In his book, “Inventing the electronic Century,” Professor Alfred Chandler documents how Justice Department consent decrees with RCA, AT&T and IBM, which made important patents available to even rivals, created enormous competition and innovation in data processing, consumer electronics, and telecommunications. The evidence is that limiting or abolishing patent protection has far more beneficial impact than its protection, let alone allowing its use to justify anticompetitive exclusion.
Probably the weakest case for the economic value of patents exists in the software industry. Bill Gates, reflecting on patents in the software industry said in 1991 that:
If people had understood how patents would be granted when most of today’s ideas were invented and had taken out patents, the industry would be at a complete standstill today…A future start-up with no patents of its own will be forced to pay whatever price the giants choose to impose.
The point is that there is very little support for antitrust courts to elevate IP to a justification for market exclusion. The case for procompetitive benefits from patents is nonexistent, while much evidence supports an exclusionary motive for obtaining IP by big tech firms.
As Professors James Bessen and Michael Meurer show, patents on software are particularly problematic because they have high rates of litigation, are of little value, and many appear to be trivial. In particular, Bessen and Meurer argue that many software patents are obvious and therefore invalid. Moreover, the claim boundaries are “fuzzy” and therefore infringement is expensive to resolve.
When asserted in a rule-of-reason case under the Apple precedent, software patents would seem to escape all scrutiny. The defendant would simply assert IP protection without any obligation to reveal with specificity the nature of the IP. The plaintiff then would have no way to challenge validity or infringement or to be able to demonstrate an ability to design around the defendant’s IP. Instead, they must show, per the Ninth Circuit’s opinion, that there is a less restrictive way for the plaintiff to be paid for its IP that is “virtually as effective” and “without increased cost.” This makes no sense at all. It would make far more sense to force any tech platform that seeks to exclude competitors on the basis of IP to simply file a counterclaim to the antitrust complaint alleging patent or copyright infringement and seeking an injunction that excludes the plaintiff. In such a case, the platform’s IP can be tested for validity. The exclusion by the antitrust defendant can be compared to the patent grant, and patent misuse can be examined.
Ignoring Its Own Precedent
It is unfortunate that the Apple court did not take seriously the Circuit’s earlier analysis in Image Technical Services v. Eastman Kodak. There, Kodak defended its decision to tie its parts and service in the aftermarket by claiming that some of its parts were patented. The Court noted that “case law supports the proposition that a holder of a patent or copyright violates the antitrust laws by ‘concerted and contractual behavior that threatens competition.’” The Kodak Court’s example of such prohibited conduct was tying, a claim made by Epic. Because we know that there are numerous competing payment systems, and because nothing in the Ninth Circuit’s opinion addresses the specifics of Apple’s IP that must be protected, it is likely the case that Apple does not have blocking patents that preclude use of alternative payment systems. And if this is the case, Epic alleged the very situation where the Ninth Circuit earlier (citing Supreme Court precedent) found that patents or copyrights violate the antitrust laws. Moreover, the Ninth Circuit thought it was significant that Kodak refused to allow use of both patented or copyrighted products and non-protected products. This may also be true of Apple’s development license in the Epic case. The Court didn’t seem to think that an inquiry into what IP was licensed by these agreements to be significant.
In sum, use of IP as a procompetitive business justification has no place in rule-of-reason cases. There is no evidence IP is procompetitive, and use of IP as a business justification relieves the antitrust defendant of the burden to demonstrate validity and infringement required in IP cases. It further stacks the deck in rule-of-reason cases against plaintiffs, and unjustly favors exclusionary practices by dominant tech platforms.
Mark Glick is a professor in the economics department of the University of Utah.