Antitrust and the Patent System: A Reexamination

98 Pages Posted: 27 Aug 2014 Last revised: 28 May 2015

Herbert J. Hovenkamp

University of Iowa - College of Law

Date Written: June 27, 2015

Abstract

Since the federal antitrust laws were first passed they have cycled through extreme positions on the relationship between competition law and the patent system. Previous studies of antitrust and patents have generally assumed that patents are valid, discrete, and generally of high quality in the sense that they further innovation. As a result, increasing the returns to patenting increases the incentive to do socially valuable innovation. Further, if the returns to the patentee exceed the social losses caused by increased exclusion, the tradeoff is positive and antitrust should not interfere. If a patent does nothing to further innovation, however, then any amount of social loss from increased monopoly is harmful. Since only a subset of patents are worthless, this naturally invites the question whether considerations of patent social value are a reasonable element of competition policy. As a general matter, the answer is no, but there are exceptions.

At various points in their history both antitrust law and patent law have engaged in considerable overreaching. Beginning in the late 1970s, however, antitrust law went through a lengthy, ongoing process of court-imposed discipline that has brought its rules more closely into alignment with its stated concern, which is increasing consumer welfare by promoting competition. Antitrust cases are far more difficult to win, the per se rule is less frequently used, and we have considerably heightened the requirements of allegation and economic proof. By contrast, patent law has continued on an expansion course that is only now showing signs of abating. Patent law largely still awaits the consumer welfare revolution that has already occurred in antitrust doctrine. Writing in the mid-sixties Ward S. Bowman and Robert H. Bork warned of a "crisis in antitrust," presaging the significant reform that was to follow. Today we are facing a crisis in patent law.

One particularly important problem is the uneven way antitrust and patent policy accommodate one another. Beginning with a set of tools for assessing economic performance, antitrust law has developed a strong tradition of taking innovation effects into account, even in cases where specific IP rights are not invoked. One good example is the strong rule that innovation as such cannot be an unlawful exclusionary practice. In sharp contrast, courts construing the Patent Act typically disregard competition effects or sometimes even treat increased competition as part of the "harm" that patent policy should limit. The discussion here uses "no authorized generics" provisions in Hatch-Waxman pharmaceutical settlements as one example.

This article also examines antitrust and patent law as regulatory institutions with legislative mandates to regulate in their given areas, but subject to limitations that all regulatory institutions face -- namely, high cost, imperfect information, and special interest capture. One failed approach to this regulatory enterprise is the view that the patentee acts improperly when it engages in activity "beyond the scope" of the patent. The flip side is that activity that is not "beyond the scope" is permissible. A more sensible way to view the interaction between the patent and antitrust regulatory systems is to divide patent activity into two parts: pre-issuance and post-issuance conduct. Secondly one must look for explicit statutory authorization of the conduct in question. Post-issuance conduct that is not statutorily authorized is generally amenable to antitrust scrutiny. Next I examine the antitrust and patent systems as regulatory institutions, finding that today the presence of special interest capture is far stronger in the patent system than the antitrust system, although that may not always have been the case. After that I turn to the very different ways that antitrust and the patent system approach economic policy. The antitrust system is empirical, market based, and acutely sensitive to market diversity. In sharp contrast, the patent system is dominated by a much more myopic set of queries concerned with the boundaries of individual property rights and largely indifferent to market performance and diversity. Then comes an analysis of the appropriate roles of antitrust's rule of reason, per se rule, and a possible "quick look" approach.

Finally this article develops a set of rules for evaluating specific disputes that implicate both antitrust and the patent system, focusing mainly on the differences between pre-issuance patent conduct, which is intensely regulated, and post-issuance conduct; as well as the differences between practices that are expressly authorized by the Patent Act and those that are not. These principles are applied to a number of practices, including product price-fixing in patent licenses, pooling and cross-licensing, vertical practices, pay-for-delay settlements and other naked market division agreements, and improper patent enforcement actions, including actions by patent assertion entities. Contrary to widespread belief, courts rarely need to determine patent validity or infringement in order to assess the antitrust consequences of a settlement agreement or other license. In any event, often the appropriate fix must come, not from the antitrust laws, but rather from the patent court's own equitable powers.

Keywords: antitrust, patents, restraints of trade, monopolization, exclusionary practices, licensing, actavis, authorized generics

Suggested Citation

Hovenkamp, Herbert J., Antitrust and the Patent System: A Reexamination (June 27, 2015). 76 Ohio State Law Journal 468 (2015); U Iowa Legal Studies Research Paper No. 14-27. Available at SSRN: https://ssrn.com/abstract=2486633 or http://dx.doi.org/10.2139/ssrn.2486633

Herbert J. Hovenkamp (Contact Author)

University of Iowa - College of Law ( email )

407 Boyd Law Building
Iowa City, IA 52242
United States
319-335-9079 (Phone)

Paper statistics

Downloads
502
Rank
43,621
Abstract Views
3,345