The procompetitive interest in intellectual property law.

• Author: Cotter, Thomas F.

ABSTRACT

When government recognizes intellectual property (IP) rights, it is often viewed as sanctioning the existence of private "monopolies," in contrast to the general antimonopoly thrust of the antitrust laws. And yet, on occasion IP law itself condemns conduct on the part of IP owners--or excuses otherwise infringing activity on the part of IP defendants--expressly for the purpose of promoting competition. It does so even though antitrust law--if one were to apply it at all under analogous circumstances--would not find anticompetitive harm without conducting a more thorough analysis of whether the antitrust defendant possesses power over a well-defined market. Salient examples include the misuse doctrines in patent and copyright law; some applications of merger and fair use in copyright; and trademark law's functionality doctrine. In this Article, Professor Cotter develops a theoretical explanation for that divergence between antitrust and IP. Specifically, he argues that in some limited contexts the expected social costs, including error costs, of ruling for IP defendants may be low in comparison with the expected anticompetitive harm from ruling for IP plaintiffs. As a result, IP courts sometimes may enhance welfare if they are less concerned than antitrust courts about the expected costs of "false positives," that is, cases wrongly decided against the party defending the allegedly anticompetitive conduct. Put another way, it sometimes may be appropriate for courts to excuse IP defendants from liability, in order to avert relatively speculative threats of anti-competitive harm. Professor Cotter further contends that such cases probably are more common in the copyright than in the patent law context, and that even in copyright contexts courts should be cautious about casually inferring anticompetitive harm; but that his analysis provides a rationale for a relatively expansive definition of trademark functionality.

TABLE OF CONTENTS INTRODUCTION I. THE ANTITRUST CONTEXT II. IP's PROCOMPETITIVE DOCTRINES A. Misuse B. Merger and Fair Use C. Functionality III. ANALYZING THE COSTS AND BENEFITS OF ALLEGED ANTICOMPETITIVE CONDUCT A. A Thought Experiment B. [H.sub.1] and [H.sub.2] in Antitrust Law C. [H.sub.1] and [H.sub.2] in IP Law IV. IMPLICATIONS FOR IP's PROCOMPETITIVE DOCTRINES A. Procompetitive Tools in Patent and Copyright 1. Two Problems with Misuse 2. Restrictions on Reverse Engineering 3. Restrictions on Price Discrimination 4. What Is To Be Done? B. Functionality CONCLUSION INTRODUCTION

Conventional wisdom holds that antitrust and intellectual property (IP) law seek to maximize social welfare in opposite ways, with antitrust law condemning monopolies to attain this goal and IP law granting temporary monopolies to achieve the same end. (1) This "wisdom" is an oversimplification insofar as (1) contemporary antitrust enforcers, recognizing that IP rights typically do not give rise to economically meaningful monopolies, (2) approach IP-related conduct and transactions with much less inherent suspicion than often was the case in years past; (3) and (2) antitrust law by itself does not condemn the mere possession of monopoly power, but rather certain exercises of or efforts to obtain it. (4) Nevertheless, there is a kernel of truth to the conventional characterization, to the extent that IP rights sometimes enable the exercise of a degree of market power, (5) and sometimes even permit IP owners to engage in conduct that would be unlawful absent the IP right. (6) Though subject to exaggeration, the recognition and enforcement of IP rights might be seen as an exception to the antimonopoly policy embodied in the antitrust laws.

There is another side of the coin, however, that is less frequently commented on. (7) On occasion, IP law condemns conduct on the part of IP owners--or excuses otherwise infringing activity on the part of IP defendants--for the express purpose of promoting competition. It does so even though antitrust law--if it were to apply at all--typically would not condemn similar conduct on the part of the IP owner, or require the IP defendant be given free access, absent thorough analysis of (1) the markets within which the parties compete, and (2) whether the IP owner possesses market power. (8) To illustrate, consider these examples of what are referred to here as IP's "procompetitive doctrines": (9)

* In patent and copyright cases, courts sometimes apply the misuse doctrine to render IP rights unenforceable until the plaintiff "purges" that misuse. (10) Courts continue to state the principle that certain practices on the part of IP owners are sufficiently anticompetitive to constitute misuse, even though the same conduct would result in antitrust liability, if at all, only upon proof of additional elements including market power. (11)

* In copyright law, courts sometimes excuse defendants from liability by applying the merger or fair use doctrines in contexts in which the copyright owner's exercise of exclusive rights, often in some aspect of its computer software, might impede competition in a related market. (12) Courts, however, do not require the defendant to define what that market is, or to quantify the harm, in any way comparable to the plaintiffs burden in analogous antitrust litigation.

* In trademark law, the functionality doctrine permits defendants to copy a trademark owner's distinctive product design, notwithstanding the potential for consumer confusion, if inter alia the copying is justified by competitive need. (13) As in the above two examples, however, courts rarely articulate what they mean by "competitive need," (14) and do not apply antitrust standards to determine whether according exclusive rights would impede competition in any well-defined market.

Two possible justifications for IP law's comparatively loose conceptualization of "markets" and "competition" are what might be referred to as the "beyond the scope" and "adjudication cost" rationales. The "beyond the scope" rationale posits that courts correctly excuse IP defendants from liability when not doing so would enable the IP owner to control subject matter that the IP laws themselves, presumably for reasons bearing some relation to competitive need, relegate to the public domain--"beyond the scope" of the IP owner's rights. I will argue that this justification is useful up to a point, but that ultimately it depends upon a pre-existing baseline--a principle that IP rights extend only up to Point X and no further--which is only sometimes readily discernible. (15) Everyone may agree, for example, that copyright does not subsist in ideas; but exactly what distinguishes an "idea" from "expression" is often unclear, and in many instances the distinction must be made in light of its predictable consequences. Put another way, formalistic application of IP's procompetitive doctrines to prevent the expansion of IP rights beyond their scope might undermine consideration of how best to interpret that scope to attain an optimal accommodation of monopoly incentives with competition. Alternatively, under an "adjudication cost" rationale, one might justify IP law's less rigorous approach to competitive need on the ground that it would be unduly expensive in some contexts to prove competitive harm in the same way that antitrust often requires. Potentially turning every merger doctrine or trade dress case into a miniature antitrust dispute might seem extravagant. (16) Like the "beyond the scope" argument, however, this argument also rests on a baseline assumption--in this instance, that the social benefits of protecting the IP rights at issue are not sufficiently large as to warrant investment of substantial resources into more accurately evaluating competitive need. The assumption may be correct, but it would be useful to articulate why, when the costs of greater certainty are prohibitive, some cases involving speculative competitive harms should be resolved in favor of IP defendants rather than IP plaintiffs.

Notwithstanding the preceding reservations, my thesis is that IP law, on occasion, may increase social welfare by promoting competition in ways that antitrust law does not address, and may do so based on evidence that would be insufficient in an antitrust context. Context, in other words, is crucial; although forced sharing to attain optimal competition might seem unwarranted in most antitrust contexts, absent clear proof of market harm, it might constitute good IP policy--even in the presence of ambiguous evidence. A major premise of this argument is that both antitrust and IP law can be viewed as seeking to minimize the expected cost of three things: "false positives," that is, cases incorrectly (17) decided in favor of the IP defendant or the antitrust plaintiff; "false negatives," that is, cases incorrectly decided in favor of the IP plaintiff or antitrust defendant; and adjudication. In seeking to attain this goal, however, antitrust and IP may diverge insofar as certain types of "false positive" harms that are likely in the antitrust context may, on occasion, be nonexistent in the IP context; and certain types of "false negative" harms that are not cognizable in the antitrust context are worth taking into account in the IP context. Apropos of the latter point, I will argue that IP law sometimes should be less tolerant than antitrust of (1) harms stemming from the exercise of small-scale market power; (2) harms that are individually of little competitive significance, but which in the aggregate threaten to reduce social welfare; and (3) threats to dynamic efficiency stemming from the reduction of competition in so-called "innovation markets." In addition, although it often may be good antitrust policy to incur substantial adjudication costs to define the market and ascertain anticompetitive consequences, in the IP context it may enhance welfare to avoid these costs if the stakes are sufficiently low, or...