Based on a speech before the American Bar Association's Section of Antitrust Law Program, "Intellectual Property and Antitrust: Navigating the Minefield"
The interface between intellectual property law and competition is a challenging subject that is a relatively new one for me. For a variety of reasons my individual private practice did not focus on patent antitrust issues.
One reason was that patent lawyers and antitrust lawyers tend to be segregated in the private bar and in corporate legal departments. When I worked in General Motors, for example, the Legal Staff reported to the General Counsel and the Patent Staff had a different reporting line; communications between the two staffs were sporadic. In private practice, I found that other client companies were often organized in much the same way and their inside patent lawyers tended to have relationships with different outside firms.
My personal history is, of course, unimportant but the fact that patent lawyers and antitrust lawyers have been segregated from one another is, I think, profoundly significant. We deal with overlapping issues but we have been schooled in different traditions. Programs like this one can help to bridge the gap. There are a number of other speakers on this program who are genuine experts in the patent-antitrust interface and they have produced program materials of very high quality.
If you think about it, the American Bar Association, among other activities, functions as an enormous research joint venture. The amount of research material that is made available to members in section publications and in conferences like this one is remarkable. The concept of the ABA as a research joint venture is something that I will return to later.
The subtitle of this program refers to navigating the patent-antitrust minefield. I will leave it to others to talk about navigation. What I want to talk about here today are proposals to reconfigure the minefield or perhaps remove most of the mines altogether. Because of time and space limitations, the views I will express are necessarily selective and suggestive rather than comprehensive and conclusive.
II. Proposals For Special Antitrust Rules To Apply To "High-Tech" Industries
As you well know, a number of articles have been written in recent years that point to the special characteristics of so-called "high tech" industries and advocate special antitrust treatment or broad antitrust exemptions for industries that fit into this category.(2) The writers claim that innovation proceeds at a vastly accelerated rate in the high tech sector, and that the useful life of any particular innovation is therefore relatively short -- perhaps lasting for only three to five years. They argue that any market dominance that lasts for such a short period of time is not worth worrying about and, in any event, will have evaporated before meaningful antitrust remedies could be brought to bear.
This is an issue that clearly needs to be addressed when considering remedies.(3) Substantive standards are different. One problem with different substantive standards is that the proposals are not backed up by specifics. We do not have a precise definition of the industries which would fall within the definition of "high tech" and hence appropriately be subject to special rules. Some have suggested that high tech industries are those characterized by so-called "network effects" -- defined as a situation where "the utility that a user derives from consumption of the good increases with the number of other agents consuming the good."(4) There is no agreement, however, on the appropriate antitrust rules for these industries.(5) It is not obvious, for example, that high tech industries, thus defined, will evolve more quickly than other industries. Moreover, there are many industries with an equal, or greater intellectual property content that do not display network effects at all. As a matter of law and policy, it is not possible to even consider special substantive rules or special exemptions without a better idea of the areas to which they would apply.
Some argue that in the high tech field the most important area of competition is competition for monopoly.(6) The argument is that we need not be concerned if a particular company has a dominant position at any particular point in time because that dominant position is precarious, and other companies will be constantly striving to topple the dominant rival and to substitute their own product as the dominant one for an equally brief period of time. I recognize the force of this argument, but I am not sure that this kind of competition is a recent phenomenon, uniquely associated with high tech businesses.
I worked from 1971 through 1982 in the automobile industry, which is much more high tech than people realize but which is probably not considered one of the family of high tech industries in this day and age. The president of my employer, General Motors, once had in the corner of his office a large rectangular object made out of Lucite -- perhaps a foot and a half on either side and three feet tall. Embedded in it were a number of what looked like marbles of different sizes and different colors. The first time I saw it, I thought it was some kind of art object, but I was told that it was a display of competing automobiles. It could display five different attributes -- things like model type, cost, or drive train -- in the three spatial dimensions of the object and the size and the color of the embedded marbles. The president explained that the display helped him to visualize the structure of competition in the business. "What I'm always looking for are the relatively empty product spaces and I am trying to figure out the best products that might go in there." That sounds a lot like competition for some kind of short-term dominant position in a market space that other people had not exploited before. There are many examples of this competitive strategy in the history of the automobile business; Chrysler's introduction of the minivan is perhaps the most conspicuous recent example. I tell this story because I am not sure that there is anything really new or different about the concept of competition for a transitory dominant position.
Furthermore, even if we were to acknowledge that the phenomenon exists in a wide variety of industries, I am not sure what the consequences are for antitrust enforcement. For example, if companies compete for transitory monopolies and if those monopolies are short lived, does that mean we should be more or less tolerant of exclusive dealing arrangements that last for a relatively short period of time? For another example, consider merger law. If the achievement of a substantial market position can be accomplished quickly and lasts for only a short period of time, what impact should it have on the time frames that we apply to evaluate whether potential entry will be timely and sufficient to defeat merger-related market power? I think questions like this have to be explored in some detail before we glibly assume that we know the appropriate antitrust treatment for high tech industries with supposedly special competitive characteristics -- assuming that we can identify them in the first place.
Consider other policy implications on the intellectual property side. At the moment patent protection is granted for a period of twenty years.(7) This period of time is purely arbitrary. We really do not know whether twenty years is any better than fifteen or twenty-five. What conclusions would you draw about the reasonableness of this period if you assumed that the useful life of an invention was only five years? After all, one of the tradeoffs that justifies the grant of a patent monopoly is the fact that a patent holder has to disclose the particulars of the invention in a way that it will help other people in the world to copy it and get in the business themselves at the conclusion of the patent monopoly period. If the invention has such a short useful life, disclosure may not be worth much, as a practical matter. Part of the "consideration" for the monopoly has failed. Does this mean that we should consider a shortened time frame for patent protection in the so-called "high tech" area?(8) I am not aware that anyone has suggested this, but it would be a logical question to ask if indeed innovations in high tech businesses have an especially short useful life.
Please do not misunderstand. I do not advocate special patent terms for high tech industries. I am simply pointing out that an argument for special antitrust rules, based on short life spans, could suggest that we also examine the length of the patent period. At the present time, I do not think there is a factual basis for special rules in either area.
III. Proposals To Change The "Balance" Between Intellectual Property and Antitrust Values
Others do not claim that there is a need for special antitrust exemptions applicable to high tech industries, but argue that the balance between intellectual property law and competition law has been tilted in a way that tends to subordinate intellectual property values. It is true that the apparent balance between the two has shifted back and forth over the years.(9) In the 1960s, for example - a time when antitrust law was compared to the Magna Carta(10) -- patent law tended to be regarded by the antitrust community as a alien intrusion, which undercut much more exalted competition values and which should be given the narrowest possible application. On the other hand, about twenty years later, the tilt was sharply in the other direction. The view expressed in the Department of Justice's International Guidelines,(11) for example, was that all patent license restrictions were presumably lawful, because the patent holder can legally exclude all competitors and therefore a license allowed more competition than could have otherwise taken place.(12) Notwithstanding the later withdrawal of these Guidelines, substantial elements of this view are retained in the new Guidelines on Licensing Intellectual Property.(13)
Some argue today that intellectual property rights should be given more exalted status because they are essential for innovation, whereas competition values may be based on fuzzy predictions of the future and therefore inherently less credible. I question whether this distinction makes sense. Patent law and the incipiency elements of antitrust law are similar in that they both are ultimately based on inherently uncertain predictions of what is going to happen in the future. The difference is that in the antitrust regime, we sometimes are concerned about conduct that in the short term may be benign or even helpful to consumers, but that may be harmful in the long term, whereas in the patent regime we are willing to tolerate immediate consumer harm in the expectation that in the long run it will benefit consumers by encouraging innovation.(14) In both regimes, there really is no rigorous way to compare the immediate effects with the ultimate effects. In other words, there is no reason to suppose that one is more solidly grounded than the other.
There is fertile opportunity here for more focused research into the balance between immediate competitive effects and future competitive effects. We might want to look at whether we have the right balance on the patent side as well as the competition side. Bob Pitofsky, the Commission's former chairman, gave a speech a few months ago in which he called attention to a looming controversy over the excessively anticompetitive potential of some patents that are being granted today, particularly in the area of so called "business methods."(15) The Federal Trade Commission, of course, does not determine patent policy, but we do have the responsibility, indeed the obligation, to point out the competitive consequences of policies that are determined by others. We are not in a very good position to evaluate whether patents are or are not improvidently granted, but we can point out what the effects might be if they are. It also might be of some interest for researchers to evaluate the incentives that exist in the patent office.
It has been claimed that patent examiners have an incentive to grant patents improvidently because it is a lot easier for them to meet their workload quotas that way.(16) If an examiner grants the patent, that is the end of it for the examiner; if the examiner denies the patent and the inventor does not abandon the application, the examiner is involved in extensive appellate processes. If differential incentives really exist, it is an appropriate matter for public debate. For a comparable example, imagine the uproar if the antitrust agencies had financial incentives to turn down mergers rather than to clear them.
People who claim that antitrust enforcers are not paying sufficient attention to intellectual property rights today also point to a number of recent cases, which they claim should not have been brought. The one Federal Trade Commission case that is often cited is Intel.(17)
This case was brought and settled before I joined the agency and I do not know enough about the facts to have an opinion on the merits. You may remember that Intel had denied access to certain data, directly or indirectly, to three companies that were pursuing their own intellectual property claims against Intel. There may well be a legitimate issue as to whether the denial of access to this information merely caused harm to individual competitors or whether it caused harm to the competitive system sufficient to justify antitrust intervention.(18) But, I do not think the Intel case is good evidence for the proposition that the antitrust agencies have devalued intellectual property rights. After all, there were intellectual property rights on both sides of the equation.
In addition, the case seems to have been settled with a decree that makes a lot of sense. I compare it to a nuclear disarmament pact. The decree basically says that Intel cannot deny access to competitively important information (a "nuclear" attack) merely because it is engaged in an intellectual property dispute with another company. If, however, that company itself launches a "nuclear" strike against Intel by seeking a broad injunction, Intel is entitled to retaliate by denying information. It seems to me that the decree is one way of encouraging people to litigate genuine intellectual property disputes without draconian consequences on either side and, as such, is procompetitive and pro-intellectual property.
IV. The Current State of Patent-Antitrust Law
My skepticism about certain critiques of the present balance between intellectual property rights and competition law does not mean that I am dug in in defense of the status quo. I have an open mind on a number of questions and would certainly welcome discussion of specific issues. I will say that I believe the 1995 Intellectual Property Guidelines are generally useful and correct. There are three fundamental principles stated in section one of the Guidelines. First, intellectual property is "comparable to other" property; second, there is no presumption of market power simply from the possession of intellectual property; and third, there is an express recognition of the fact that intellectual property licensing is generally procompetitive. I support each one of those basic principles. I also subscribe to the general test for evaluating intellectual property licenses in Guidelines Section 3.1, which states that horizontal licensing arrangements are suspect when they harm "competition among entities that would have been actual or likely potential competitors" in the absence of the license. This is a very useful way to look at the question; the statement briefly captures some longstanding principles of patent antitrust law. For example, a license that purports to restrict dealing in a non-infringing product will be suspect because it has adverse consequences on competition that might otherwise occur in the area of the non-infringing product.
Issues can become extremely difficult, of course, if we do not really know whether competition could have taken place in the absence of a license restriction. This is the problem presented in some recent cases involving the settlement of intellectual property disputes between the holder of a pharmaceutical patent and a potential generic entrant.(19)
There is a risk of collusive settlements because the monopoly rents to be shared are likely to be in excess of the potential combined profits of both in a competitive regime. There is an incentive to settle a dispute with some licensing arrangement that splits the monopoly rents. In this situation, there may never be a judicial resolution of the patent dispute, so we cannot be sure whether competition would have been possible but for the license. At the Commission, we are presently groping for some proxy tests to distinguish between settlements that are likely to be procompetitive and those that are likely to be anticompetitive, without embarking on an independent inquiry into the merits of the underlying patent case. This is a subject for another talk,(20) but it is also an illustration of how the Intellectual Property Guidelines may be easier to state than to apply.
Incidentally, the test in Section 3.1 of the Guidelines is expressly limited to arrangements between possible competitors at a horizontal level; it does not facially apply to purely vertical relationships. I surmise that this limitation was not based on rigorous logic but rather on a perceived need to recognize existing law on vertical arrangements in the intellectual property area and in other areas. For example, there is a fairly well settled body of law today holding that resale price maintenance is illegal in a patent licensing arrangement.(21) In addition, it seems settled that royalty terms extending beyond the term of the patent are suspect.(22) There are situations where arrangements of this kind could be benign or procompetitive, and a consistent application of the principles in section 3.1 of the Guidelines would recognize them. However, it will take some time for the law to evolve in this direction, and courts will need to lead. An agency like the Commission cannot simply disregard a well-settled body of law because we do not think it makes sense.(23)
If courts were willing to reconsider the unqualified hostility to resale price maintenance, in particular, it would eliminate the need for other artificial distinctions based on semantics more than on reason. I refer specifically to the artificial distinction between a licensor's outright refusal to deal with people unwilling to maintain resale prices and licenses that contain r.p.m. provisions. This distinction, of course, derives from the Colgate case,(24) which sanctions similar "unilateral" refusals to deal in a broader context. The Colgate doctrine is one way to mitigate the adverse effects of an overly rigid rule, but it does so in a way that does violence to fundamental notions of contract law. We all learned in law school that a unilateral offer can be accepted by performance, and that is exactly what happens when a Colgate program is running well. The fiction that there is no "agreement" in these situations means that judicial scrutiny is misdirected to nuances of verbal communication rather than to competitive effects.
Finally, a word about research joint ventures. I personally believe that the discussion of them in the recently published Guidelines for Collaborations Among Competitors(25) may be misunderstood and may cause people to avoid research ventures that are entirely benign and helpful. In my view, the focus when evaluating cooperative research should be on the nature of the collateral restraints, if any. These are the issues that were highlighted in the statute that governs collaborative research and indeed collaborative production.(26) The so-called structural issues, which are highlighted in our Guidelines,(27) are very unlikely ever to be applied to research. There has never been an antitrust prosecution of a pure research joint venture, without collateral restraints,(28) and I think it is unfortunate to suggest that we are likely to attack one now.
At the beginning of this talk, I referred to the Antitrust Section of the American Bar Association as a research joint venture. That is exactly what it is. People have the impression that the model research joint venture employs a bunch of people who wear long, white laboratory coats in front of gurgling retorts. Of course, some research joint ventures perform experiments, but a lot of research ventures simply seek out and organize, for the benefit of the members, some elements of the vast amount of knowledge already available in the world. That is the service provided by the Antitrust Section of the American Bar Association. I think it would be silly to suggest that the "market share" of the Antitrust Section is remotely relevant in determining whether its research activities are procompetitive, and the same could be said for a lot of other research joint ventures.
In conclusion, I appreciate the opportunity to share with you some of these general observations on the configuration of the "antitrust minefield." Others today will actually guide you through the field, but I welcome your thoughts on the underlying issues that I have tried to raise in this talk.
1. I would like to acknowledge the assistance of my advisors, Holly Vedova and Thomas Klotz, in preparing this written version of my remarks before the American Bar Association Section of Antitrust Law Program, "Intellectual Property and Antitrust: Navigating the Minefield," Philadelphia, Pennsylvania (May 3, 2001). The opinions expressed, however, are my own and are not necessarily shared by any other commissioner.
2. See, e.g., Richard M. Brunell, Appropriability in Antitrust: How Much Is Enough?, 69 Antitrust L.J. 1 (2001); Richard J. Gilbert & Willard K. Tom, Is Innovation King at the Agencies? The Intellectual Property Guidelines Five Years Later, 69 Antitrust L.J. 43 (2001).
3. A government agency has a great deal more discretion when it comes to remedies than it does when it comes to substantive law. There have been situations where the special characteristics of certain industries were taken into account in the selection of particular remedies. For example, in the recent AOL-Time Warner merger, the negotiated decree lasts for only five years, while the typical Commission decree lasts for ten. Federal Trade Comm'n, Statement of Policy with Respect to Duration of Competition and Consumer Protection Orders, (Aug. 9, 1995), reprinted in 4 Trade Reg. Rep. (CCH) ¶ 13,320. In addition, considerations of this sort could affect whether the Commission decides to embark upon a relatively protracted administrative proceeding, or decides to seek an immediate court resolution of the controversy. In the recent Microsoft opinion, the court noted the problem of devising remedies when the competitive situation has already "changed drastically." "Conduct remedies may be unavailing. . . , because innovation to a large degree has already rendered the anticompetitive conduct obsolete (although by no means harmless). And broader structural remedies present their own set of problems, including how a court goes about restoring competition to a dramatically changed, and constantly changing, marketplace." United States v. Microsoft Corp., No. 00-5212, 2001 U.S. App. LEXIS 14324, at *15 (D.C. Cir. 2001) (emphasis in original). Nonetheless, the Microsoft court did not suggest that prosecutors should abdicate on that account.
4. Michael L. Katz and Carl Shapiro, Network Externalities, Competition, and Compatibility, 75 Am. Econ Rev. 424, 424 (1985).
5. United States v. Microsoft Corp., 2001 U.S. App. LEXIS 14324, at *16-20. In light of this controversy, the Microsoft court decided to apply traditional antitrust standards.
6. See, e.g. United States v. Microsoft Corp., 2001 U.S. App. LEXIS 14324, at *17-18
7. 35 U.S.C. § 154(a)(2) (1994).
8. It could be argued, of course, that disclosure still has value because it facilitates follow-on innovation. It also could be argued, however, that even a short-term monopoly on a short-term invention can also have follow-on competitive effects.
9. See Michael S. McFalls, Basic Antitrust Issues Involving Intellectual Property, Materials from A.B.A. Section of Antitrust Law Program "Intellectual Property and Antitrust: Navigating the Minefield," Philadelphia, Pennsylvania (May 3, 2001) at 73-74.
10. United States v. Topco Assocs., Inc., 405 U.S. 596, 601 (1972).
11. United States Dep't of Justice, Antitrust Enforcement Guidelines for International Operations (1988), reprinted in 4 Trade Reg. Rep. (CCH) ¶ 13,109, withdrawn, United States Dep't of Justice & Federal Trade Comm'n, Antitrust Enforcement Guidelines for International Operations (1995), reprinted in 4 Trade Reg. Rep. (CCH) ¶ 13,107.
12. A similar argument was made by Microsoft and rejected by the Court of Appeals. See United States v. Microsoft Corp., 2001 U.S. App. LEXIS 14324, at *58-59.
13. United States Dep't of Justice & Federal Trade Comm'n, Antitrust Guidelines for the Licensing of Intellectual Property (1995), reprinted in 4 Trade Reg. Rep. (CCH) ¶ 13,132.
15. Robert Pitofsky, Antitrust and Intellectual Property: Unresolved Issues at the Heart of the New Economy, Remarks before Antitrust, Technology and Intellectual Property Conference, Berkeley Center for Law and Technology, Berkeley, California (March 2, 2001), </speeches/pitofsky/ipf301.htm>.
16. See Robert P. Merges, As Many as Six Impossible Patents Before Breakfast: Property Rights for Business Concepts and Patent System Reform, 14 Berkeley Tech. L.J. 577, 607-09 (1999); Brian Kahin, The Expansion of the Patent System: Politics and Political Economy, 6 First Monday (Jan. 2001) <http://www.firstmonday.dk/issues/issue6_1/kahin/index.html>. The quotas are an important factor in promotion and bonus decisions.
18. Timothy J. Muris, The FTC and the Law of Monopolization, 67 Antitrust L.J. 693 (2000).
19. In the Matter of Abbott Labs. and Geneva Pharmaceuticals, Inc., FTC Docket Nos. C-3945 and C-3946 (March 16, 2000) </os/2000/03/index.htm#16>; In the Matter of Hoechst Marion Roussel, Inc., Carderm Capital L.P., and Andrx Corp., FTC Docket No. C-9293 (March 16, 2000) </os/2000/03/index.htm#16>; In the Matter of Schering-Plough Corp., FTC Docket No. 9297 administrative complaint filed Apr. 2, 2001)
20. See Thomas B. Leary, Antitrust Issues in Settlement of Pharmaceutical Patent Disputes, Prepared Remarks before Sixth Annual Antitrust Healthcare Forum, Northwestern University School of Law, Chicago, Illinois (Nov. 3, 2000) </speeches/leary/learypharma.htm>; Thomas B. Leary, Antitrust Issues in the Settlement of Pharmaceutical Patent Disputes, Part II, Remarks before ABA's Antitrust Healthcare Program, Washington, D.C. (May 17, 2001) forthcoming in Journal of Health Law (2001).
21. See United States v. Line Material Co., 333 U.S. 287 (1948); United States v. Huck Mfg. Co., 382 U.S. 197 (1965); Newburgh Moire Co. v. Superior Moire Co., 237 F.2d 283 (3d Cir. 1956); but see United States v. General Elec. Co., 272 U.S. 476 (1926).
22. See, e.g., Brulotte v. Thys Co., 379 U.S. 29 (1964).
23. In the Matter of Sony Music Entertainment, Inc.; In the Matter of Time Warner, Inc.; In the Matter of BMG Music, In the Matter of Universal Music & Video Distribution Corp.; In the Matter of Capitol Records, Inc., FTC Docket Nos. C-3971, C-3972, C-3973, C-3974, and C-3975 (Statement of Chairman Robert Pitofsky and Commissioners Sheila F. Anthony, Mozelle W. Thompson, Orson Swindle, and Thomas B. Leary, at n.2) </os/2000/09/musicstatement.htm>; In the Matter of Nine West Group Inc., FTC Docket No. C-3937 (Statement of Commissioners Orson Swindle and Thomas B. Leary) </os/2000/04/ninewestswindleleary.htm>.
24. United States v. Colgate & Co., 250 U.S. 300 (1919).
25. Federal Trade Comm'n & United States Dep't Of Justice, Antitrust Guidelines For Collaborations Among Competitors (2000), reprinted in 4 Trade Reg. Rep. (CCH) ¶ 13,160. See Concurring Statement of Commissioner Thomas B. Leary, Antitrust Guidelines For Collaborations Among Competitors (2000) </os/2000/04/antitrustguideleary.htm>.
26. National Cooperative Research and Production Act of 1993, 15 U.S.C. §§ 4301-05 (1994).
27. Antitrust Guidelines For Collaborations Among Competitors, Sections 3.33, 4.2.
28. The government charges in the so-called Auto Smog cases, often cited as precedent, contained a (subsequently discredited) claim that the venturers suppressed the sales of anti-smog devices. See United States v. Automobile Mfrs. Ass'n, 307 F. Supp. 617 (C.D. Cal. 1969), appeal dismissed sub nom. City of New York v. United States, 397 U.S. 248 (1970), modified sub nom. United States v. Motor Vehicles Mfrs. Ass'n, 1982-83 Trade Cas. (CCH) ¶ 65,088 (C.D. Cal. 1982).