Tuesday, March 6, 2012
Posted by Daniel Crane
I am already on record finding Creation Without Restraint a must-read book for anyone interested in innovation, antitrust, and intellectual property. Herb and Christina have done a wonderful job at pulling together the key contemporary themes in innovation policy and discussing them in a clear and accessible way. So kudos to the authors.
My brief comments for this symposium will focus on a central theme of the book—the idea of an “IP injury” requirement. Herb and Christina coined this concept in their Boston College Law Review article in 2010 and develop it more fully in Creation, particularly in Chapters 3 and 7. Borrowing from the concept of “antitrust injury,” which limits private antitrust suits to those that seem redress for the kinds of injuries the antitrust laws were intended to prevent, Herb and Christina propose than IP remedies should be limited to situations in which the IP holder has suffered suffer actual or likely harm conceptually linked to the central purpose of patent and copyright law, which is to incentivize innovation. To put it the other way, according to Herb and Christina’s proposed test, an IPR holder could not assert an infringement claim that did rest on the assumption that the infringing act had the tendency to diminish innovation incentives.
At first blush, the idea is appealing. Certainly, the patent and copyright systems have become unmoored in many ways from innovation incentives, partly because of litigation strategies and partly because of Congressional tinkering (mostly in the copyright space). Legal doctrines that can refocus the IP system on producing innovation are most welcome.
But I have some doubts about deploying an IP injury requirement modeled on the antitrust injury requirement. The problem is that proving harm to the competitive process and proving harm to innovation incentives are two very different kettles of fish. Proving harm to the competitive process requires examination of generally concrete and observable facts, whereas proving harm to innovation incentives requires examination of much more speculative and theoretically contestable questions.
To address competitive injury, one generally looks at questions addressed to the facts of a particular case: what is the nature of the competition at issue (call this relevant market definition or something like it); who were the competitors; how strong were they before and after the conduct; how differently did they behave because of the challenged conduct; what consequences to consumers arose from the challenged course of conduct? To be sure, answering these question isn’t always easy, but at least we know the questions to ask and the facts to study.
To address harm to innovation incentives, one has to consider a much broader and contested range of questions, many of which lie outside of the facts of a particular case: what social, economic, and personal factors contribute to the generation of new ideas; what factors contribute to the successful transmission of new ideas from the human brain to instantiations in technologies or products; under what circumstances does uncompensated borrowing of new ideas tend to spur or retard innovation; what level of incentive reward is necessary to spur investment in innovation; how does innovation in one product retard or advance innovation in another product; how much innovation is good for consumers? And that list goes on.
I greatly admire people who have answers to the foregoing questions—I wish I did. Given the extreme contestability of many of these questions, it seems that individualized proof of harm to innovation incentives on a case-by-case basis may not be a great idea. Requiring the creator to prove that the defendant’s conduct caused harm to her innovation incentives (or, more precisely, that if the creator had known ex ante that the defendant would engage in this conduct, she would have engaged in less creative activity, unless she had been sure that the courts would compensate her for her full loss from the infringement), may set up an impossibly high hurdle in many cases.
One illustration: The European Court of Justice’s 2007 decision in Microsoft v. Commission, which concerned Microsoft’s refusal to supply Sun and other competitors with specifications for protocols implemented in Windows work group server operated systems. Microsoft defended its refusal on the grounds that if it were required to give third parties access to its proprietary technology, this would reduce Microsoft’s incentives to innovate. The ECJ rejected Microsoft’s assertion, finding that Microsoft “merely put forward vague, general and theoretical arguments on that point” and offered no concrete proof that its innovation incentives would have been adversely affected.
True enough, but that’s exactly the point. Our current state of knowledge about what sorts of legal rules and economic conditions produce innovation is “vague, general and theoretical.” Requiring proof of harm to innovation incentives to the standards of proof of civil litigation may make for good theater—as did the Scopes monkey trials—but it’s hardly likely to produce satisfying results. Structuring IP litigation rules around an “IP harm” ideal makes a lot of sense, but requiring individualized proof of “IP injury” may not.
To be clear, I don’t think that Herb and Christina have in mind an open-ended “IP injury” requirement that would require the IPR holder to prove actual or likely harm to her innovation incentives in every case. Many of their specific proposals for litigation rule reforms would advance an IP harm ideal without requiring such individualized proof. My comments here should simply be understood as a mild admonition against trying to model an IP injury requirement—as a legal doctrine—on the antitrust injury requirement.