After two administrations in which the antitrust laws were not a high priority, many expected that the current administration's antitrust enforcers would bring sorely needed competence and energy to the high technology sector. Disappointment was widespread with the Antitrust Division's first major high-technology action -- the Microsoft stipulated judgment. United States v. Microsoft, Civ. No. 94-1564 (D. D.C., July 15, 1994), reprinted in 5 CCH Computer Cases ¶ 47,058 (1994). Representative of the reaction was an article by the general counsel of Sun Microsystems, which said:
The present source of Microsoft's domination in the PC world derives from its status as the standard-holder, not the practices the Justice Department condemned and which will now be prohibited under the settlement. Microsoft understands this perfectly well, which, of course, is why Bill Gates is letting the settlement happen. Nothing in the proposed settlement is likely to have anything other than the most marginal effect on Microsoft's future.
Michael Morris, "Microsoft Deal: Too Little, Too Late," San Francisco Examiner, July 24, 1994, C-6 col. 1.
With the publication of the Antitrust Division's draft Antitrust/Intellectual Property Guidelines, disappointment will justifiably deepen.[2] The Guidelines, written by a team headed by an economist, represent minimal change, do not address the critical questions of market definition, address problems with bromides and superficial examples rather than with the necessary factual and technical depth, create a new type of market (innovation market) that may be indefinable, and do not even cite the case of Eastman Kodak v. Image Technical Services, 112 S. Ct. 2072 (1992), which is the Supreme Court's most recent pronouncement on many of the relevant issues.
History, Role and Format of the Guidelines
The Department of Justice issued the first set of Guidelines including intellectual property issues in 1977. U.S. Department of Justice Antitrust Guide for International Operations (Jan. 26, 1977), reprinted in 4 Trade Reg. Rep. (CCH) ¶13,110. The role of the Guidelines then and now is to give interested parties insight into how the government analyzes cases. However, the current Guidelines themselves state that the Antitrust Division is not necessarily bound by them: "these Guidelines cannot remove judgment and discretion in antitrust law enforcement." Guidelines at ¶ 49,063.[3] The Guidelines were last revised during the Reagan administration in 1988. The seeming differences in antitrust philosophy between that administration and the current administration created an expectation of substantial change from the 1988 Guidelines.
The format of the Guidelines is similar to those that have been issued in the past. The Guidelines cover various topics, including an overview of the intellectual property and antitrust laws, antitrust concerns and modes of analysis, general principles concerning the department's evaluation of licensing agreements, application of those principles, and enforcement of invalid intellectual property rights. Throughout the Guidelines, short hypothetical examples are given of behavior that might be scrutinized by the Antitrust Division. These examples are followed by an analysis of how the Antitrust Division would likely handle the situation.
III. Comparisons With The 1988 Guidelines
A. Minimal Changes in the Guidelines
There are a few additions and changes to the Guidelines. The most significant of these are: (1) a definition of "innovation markets," (not yet recognized by the courts) where research and development are conducted; and (2) a so-called "safety zone," which allows a licensor and a licensee to enter into agreements largely without fear of antitrust scrutiny by the Justice Department as long as they make up no more than 20 percent of each market affected by the restraint.
Regarding innovation markets, according to the Guidelines, "[i]f the capacity for research and development activity that likely will produce innovation in technology is scarce and can be associated with identifiable specialized assets or characteristics of specific firms . . . it may be appropriate to consider separately the impact of the conduct in question on competition in research and development among those firms." Under the "safety zone" provision, "[t]he Department will not challenge a restraint in a licensing arrangement if (1) the restraint is not of a type that normally warrants condemnation under the per se rule and (2) the licensor and its licensees collectively account for no more than twenty percent of each relevant market affected by the restraint." According to the Guidelines, this new provision "is designed to provide owners of intellectual property with a degree of certainty, so as to encourage pro-competitive licensing arrangements."[4]
B. Unchanged aspects of the Guidelines
Other aspects of the Guidelines, while relatively unchanged, continue to have application to intellectual property that relates to computers, and particularly to computer software. For example, the Guidelines state: "antitrust concerns may arise when licensing arrangements impede competition that likely would have taken place in the absence of the license." Such arrangements would include "restrictions on goods or technologies other than the licensed technology, [and] contractual provisions that penalize licensees for dealing with suppliers of substitute technologies...." This would include licensing of computer software.
Finally, according to the Guidelines, the antitrust laws apply to intellectual property just as they apply to other forms of property. Thus, "[i]ntellectual property is neither particularly free from scrutiny under the antitrust laws, nor particularly suspect under them."
The problem with such generalized formulations is that they do not provide the sort of guidance needed to provide legal advice to a client with a problem in a complex, fact-intensive area where the interests underlying the antitrust laws and the sometimes conflicting intellectual property laws collide. What clients and litigants want to know - and what the Guidelines do not tell them -is where legitimate protection of intellectual property ends and where an antitrust problem begins in the real world.
IV. Deficiencies of the Guidelines
- Market definition is not seriously addressed.
In the high-technology sector, these questions become much more complex. For example, are hardware and operating system software, which are functionally dependent on one another, one or two markets?
Market definition is critical in the high-technology sector because companies that set de facto industry standards - as IBM did in hardware and as Microsoft is doing in operating system software - have tremendous economic power in that they can control the interfaces to the products for which they set the standard.
If competitors cannot interface with the standard-setting product - for example if a maker of applications software cannot interface with Microsoft's operating systems software - then that competitor cannot compete effectively. Interfaces may well define markets, but the Guidelines do not address this important aspect of high-technology market definition.
Indeed, although there are many places in the Guidelines where markets are mentioned, the difficult questions inherent in defining high-technology markets are ignored. It is difficult to understand, for example, why the Guidelines do not even cite the Kodak case, which is the most recent Supreme Court case on this critical issue. It is true that the U.S. Government filed an amicus brief opposing the prevailing party in Kodak, but, whether the Antitrust Division agrees with Kodak or not, the case certainly merits citation.
Kodak, for example, discusses the definition of aftermarkets and the problem, common in high technology, of products which do not function except in conjunction with other products (like operating system software and hardware). Indeed, the Supreme Court rejected Kodak's argument that because there is no demand for parts separate from service, no separate markets for parts and service may exist. Kodak, 112 S. Ct. at 2080. The Supreme Court held that "'We have often found arrangements involving functionally linked products at least one of which is useless without the other to be prohibited tying devices.'" Id. (quoting Jefferson Parish Hosp. Dist No. 2 v. Hyde, 466 U.S. 2, 19 (1984)).
Kodak also discusses other important market definition factors such as aftermarkets, interbrand competition and the lock-in theory. The Court stated that "on the occasions when the Court has considered tying in derivative aftermarkets by manufacturers, it has not adopted any exception to the usual antitrust analysis, treating derivative aftermarkets as it has every other separate market." Id. at 2089 n. 29. The court also explained the difference between interbrand and intrabrand competition: "[T]his case does not concern vertical relationships between parties on different levels of the same distribution chain. In the relevant market, service, Kodak and the ISOs are direct competitors; their relationship is horizontal. The interbrand competition at issue here is competition over the provision of service." Id. at 2084-5 n. 18.
Finally, the court recognized the concept of lock-in:
If the cost of switching is high, consumers who already have purchased the equipment, and are thus "locked-in" will tolerate some level of service prices before changing equipment brands. Under this scenario, a seller profitably could maintain supracompetitive prices in the aftermarket if the switching costs were high relative to the increase in service prices, and the number of locked-in customers were high relative to the number of new purchasers.
Id. at 2087. None of these important issues is discussed in the Guidelines.
- Use of bromides and superficial examples rather than the necessary in-depth factual, and technical analysis
In Kodak, the Supreme Court directed lower courts to engage in factual analysis rather than to dodge these important factual questions with insupportable presumptions:
Legal presumptions that rest on formalistic distinctions rather than actual market realities are generally disfavored in antitrust law. This Court has preferred to resolve antitrust claims on a case-by-case basis, focusing on the "particular facts disclosed by the record."
Kodak, 112 S. Ct. at 2082 (citations omitted). By not engaging in such factual analysis, the Guidelines do not provide the requisite guidance. Rather than using bare-bones examples, the Guidelines would have been more useful if they had focused on complex factual patterns of actual cases like Kodak.
- Innovation Markets May Be Inherently Undefinable.
The importance of relevant market definition outlined above illustrates one of the most glaring problems with the Guidelines. The so-called "innovation markets" to which the Guidelines refer may be incapable of being defined for antitrust purposes. As illustrated by the Kodak case, courts can differ widely on market definition even when there are actual products involved.
The difficulty in defining a market where products do not yet exist is even more apparent because it is not possible to apply the duPont standard of reasonable interchangeability to products that do not yet and may never exist. By definition, such "products" are not reasonably interchangeable with anything. The Guidelines simply give no guidance on this novel subject.
V.Three Recent Cases Illustrate the Deficiencies of the Guidelines: Allen-Myland, Microsoft, and Grumman.
In part because the Department of Justice has, both in the past and in the current Guidelines, failed to offer real-world examples of what constitutes an antitrust violation, courts routinely decide cases in diametrically opposed ways. For example, the Third Circuit recently issued an opinion, the central issue of which was market definition. The court of appeals and the district court had virtually opposite views on the definition of the proper market. If the Antitrust Division had chosen to issue Guidelines based on facts of real cases, the courts might not have reached such different conclusions.
Courts also have reached opposing conclusions regarding market power. The First Circuit in the Grumman case recently issued an opinion that creates an entirely new presumption regarding market power. This decision is contrary to Kodak, and also conflicts with a Ninth Circuit case. Again, the Guidelines could offer litigants and courts insight into how these cases should be handled, but for some reason have chosen not to do so. The result has been, and will continue to be, differing conclusions in the courts.
A.Allen-Myland illustrates the outcome-determinative nature of market definition.
A recent case decided by the Third Circuit illustrates the importance and complexity of market definition in high technology cases. Allen-Myland, Inc. v. International Business Machines Corporation, ___ F.3d ___, 1994 WL 420285 (3d Cir. 1994). The plaintiff in that case sued IBM for antitrust violations, claiming that "IBM had tied its [computer] upgrade installation service to the parts needed to perform the upgrades." Slip op. at 2. Market definition was the central issue in the case.
The district court's market definition and that of the court of appeals are in sharp contrast. The district court defined the relevant market to include large-scale mainframes, upgrades to large-scale mainframes, leased and smaller capacity computers, peripheral products and software, "box swaps," and upgrades using customer-provided parts. Slip op. at 4. The court of appeals found that the appropriate relevant market was limited to large scale mainframes.
The appellate court went through each of the district court's errors and, using computer-literate language, explained why the district court had erred. For example, in rejecting the district court's claim that leased computers should be included in the relevant market, the court of appeals stated: "Technology and price/performance ratios have been advancing so rapidly in the computer industry that used machines cannot be re-leased indefinitely. Accordingly, a powerful manufacturer like IBM was in a position to maximize its profits by carefully controlling the number of mainframes that would later appear on the used leasing market." Slip op. at 5.
The Allen-Myland court also recognized the concept of "lock in," as did the Supreme Court in Kodak, in rejecting the inclusion of smaller computers in the relevant market.
If it is prohibitively expensive to switch to a smaller capacity computer before the normal end of an application system's life cycle, then IBM, at least for those locked-in customers, would not face any realistic competition from smaller machines and would thus possess market power as if they did not exist.
Slip op. at 7.
Significantly, the court of appeals also vacated the district court's holding that no separate market for installations of upgrades existed. The analysis performed by the court of appeals indicated that such a market could, in fact, exist.
Had the lawyers in the Allen-Myland case had access to the Guidelines, such access would have provided no useful guidance on the most important issue in their case (and many other high-tech cases), market definition. The Guidelines should have used examples like the Allen-Myland case rather than the much more superficial examples which appear in the draft.
B. Microsoft illustrates the critical nature of interfaces.
The Department of Justice filed a Complaint against Microsoft in July of this year for alleged anticompetitive conduct. United States v. Microsoft, Civ. No. 94-1564 (D. D.C., July 15, 1994). Microsoft for many years has been the premier manufacture of operating system software. After perfecting its operating system software, Microsoft began work on applications software to run with the operating system software.
Naturally, the two software programs must interface with one another in order to operate the computer. Microsoft's applications software competitors alleged that it was extremely difficult, if not impossible, for other companies to compete with Microsoft in selling applications software that will run with Microsoft's operating system. The alleged reason for this difficulty is that Microsoft gives advanced information about its operating system software to those responsible for its applications software. This information allegedly gives Microsoft a competitive edge in applications software.
Microsoft's competitors proposed splitting up the applications software division and the operating software division of Microsoft so that the sharing of information would stop and other companies would have a fairer chance to compete with Microsoft. The settlement, however, does not go that far. "Almost 10 years later, PC manufacturers, consumers and software developers are even more tightly bound to Microsoft operating system technologies. The ties that bind are not contractual, they are technical, which is why the Justice Department settlement will be ineffective."[5]
As a solution to this interface problem, the author of the above statement and others suggested that Microsoft be required to disclose to the public its operating system interface specifications so that applications software companies would be able to manufacture software that would run with Microsoft's operating system. He also proposed that an oversight committee be established to ensure that the specifications were disclosed, and he would require all government agencies to purchase computer systems that meet those open standards.
Although the Guidelines mention the Microsoft case, they do not discuss it in any detail that would assist companies and lawyers advising them on how to deal with similar situations regarding interfaces. These situations are at the center of antitrust/intellectual property issues in high-technology markets.
C.Grumman illustrates the importance of factual analysis rather than presumptions.
One of the first appellate courts to rule on Kodak has ignored the Supreme Court's direction to use facts rather than presumptions. See Data General Corp. v. Grumman Systems Support Corporation, ____ F.3d ___, 1994-2 Trade Cases ¶ 70,716 (Sept. 14, 1994). In that case, Data General sued Grumman ("GSSC") for infringing Data General's copyrighted hardware diagnostics. GSSC filed a counterclaim for antitrust violations, claiming that Data General was using its power over diagnostic software to increase its monopoly of the market for maintenance of Data General computers.
In deciding against GSSC, the First Circuit chose to create an entirely new presumption regarding copyrights and market power.
[W]e hold that while exclusionary conduct can include a monopolist's unilateral refusal to license a copyright, an author's desire to exclude others from use of its copyrighted work is a presumptively valid business justification for any immediate harm to consumers.
Slip op. at 102.
Further confusing matters, the First Circuit's treatment of presumptions is contrary to the way the Ninth Circuit and the Supreme court have handled presumptions. In Digidyne v. Data General, 734 F.2d 1336 (9th Cir. 1984), cert. denied, 473 U.S. 908 (1985), the Ninth Circuit held that a copyrighted product "created a presumption of economic power sufficient to render the tying arrangement illegal per se. The burden to rebut the presumption shifted to [the defendant.]"
More importantly, under established Supreme Court precedent, a patent or copyright creates a presumption of market power. See Jefferson Parish Hospital v. Hyde, 466 U.S. 2, 16 (1984) ("... if the government has granted the seller a patent or similar monopoly over a product, it is fair to presume that the inability to buy the product elsewhere gives the seller market power.") ; U.S. v. Loews, 345 U.S. 38, 45 (1962) ("The requisite economic effect is presumed when the tying product is patented or copyrighted.")
Instead of giving simplistic examples of less than half a column in length, the Antitrust Division could have done a service by giving longer explications of actual cases like the Digidyne case. Without the particular factual analysis mandated by the Supreme Court in Kodak, the Guidelines are not particularly helpful. The generalizations of the Guidelines ignore the technical issues that define high-technology markets.
V. Conclusion
Because of the above deficiencies, the Guidelines are unlikely to receive much attention from the bench and the private bar. The Guidelines do give some insight to the theoretical mode of analysis in which the Antitrust Division engages before bringing an action, but it is beginning to appear that the current administration's fascination with industrial policy may inhibit it in bringing actions against successful U.S. companies in the high technology sector. It is unlikely that the Antitrust Division will find a larger target than Microsoft, and in that case it appears that the Division decided to settle for a meaningless result now rather than to devote the resources necessary for a thorough investigation of the numerous complaints by many major high-technology companies. Similarly, the lack of energy and resources put into the Guidelines make them unhelpful as a tool with which to advise clients.
ENDNOTES
1. Ronald S. Katz and Janet A. Hart are antitrust and intellectual property litigators with the San Francisco office of Coudert Brothers, which represented Image Technical Services and GSSC in cases described in this article.
2. Department of Justice Draft Intellectual Property Guidelines, Trade Regulation Reports, ¶ 50,141 (August 8, 1994) ; 59 Fed. Reg. 41339 (August 11, 1994). The Guidelines were issued in draft form to allow for 60 days of public comment.
3. The Guidelines also state the following: "As is true throughout these Guidelines, the factors listed are those that guide the Department's internal analysis in exercising its prosecutorial discretion. They are not intended to circumscribe how the Department will conduct the litigation of cases that it decides to bring...." Guidelines at ¶ 49,047 n. 15.
4. The Guidelines are careful to note that the new provision "is not intended to discourage parties falling outside the safety zone from adopting restrictions in their license arrangements that are reasonably necessary to achieve an efficiency-producing integration of economic activity."
5. M. Morris, supra, C-7 col.2.