United States: The Convergence Of The U.S. And European Laws Concerning Compulsory Licensing And Licensing Restrictions...Part II

II. Compulsory Disclosure and Licensing of IP in the U.S.

Although the essential facilities doctrine originated in non-IP cases, the U.S. courts have recently applied the doctrine (either expressly or impliedly) as a basis of compulsory licensing of IP.

In the absence of any purpose to create or maintain a monopoly, the Sherman Act does not restrict the long-recognized right of a company to freely decide with whom it will deal.[10] Employing the essential facilities doctrine, however, courts have curtailed the power of firms to freely refuse to deal with their competition. Under this doctrine, exclusion by a monopolist is unlawful where dealing is indispensable to effective competition. Such a refusal may be unlawful because a monopolist's control of an essential facility can allow it to leverage its monopoly power from one market into another.[11]

In the early part of this century, the Supreme Court formulated and applied the essential facilities doctrine to prevent one company from owning every railroad track leading into the city of St. Louis.[12] More recently, MCI relied on the doctrine to compel AT&T to give MCI access to its telephone lines.[13] The elements of an essential facilities claim are:

(1) control of the essential facility by a monopolist;

(2) the competitor's inability practically or reasonably to duplicate the essential facility;

(3) denial of the use of the facility to a competitor; and

(4) the feasibility of providing the facility.[14]

To be essential, a facility need not be indispensable. However, its denial must cause severe hardship to potential competitors. This criterion is not satisfied where access to the facility is simply more economical than alternative facilities.[15]

Although there have been many court decisions rejecting the application of the essential facilities doctrine to force IP licensing,[16] recent cases in the U.S. hold that IP may become an essential facility and may be required to be licensed to prevent or remedy antitrust violations.

1. The Kodak Cases: Compulsory IP Licensing as a Remedy for Antitrust Violations

In 1987, independent service organizations ("ISOs") filed an action against Kodak, seeking damages and injunctive relief for violations of the Sherman Act. The ISOs claimed that Kodak unlawfully tied the sale of service for Kodak printers and copiers to the sale of parts in violation of Section 1 of the Sherman Act, and monopolized or attempted to monopolize the sale of services for Kodak machines in violation of Section 2 of the Sherman Act.[17]

The district court granted Kodak's motion for summary judgment, which the Ninth Circuit reversed. Kodak appealed to the Supreme Court, which affirmed the denial of summary judgment. The Supreme Court held that issues of material fact remained, precluding summary judgment on both the § 1 and § 2 claims, and remanded it for trial to the district court.[18] One particular footnote in the Supreme Court's decision became the basis of future actions against Kodak and other companies, including Intel. There, the Supreme Court stated:

The Court has held many times that power gained through some natural and legal advantage such as a patent, copyright, or business acumen can give rise to liability if a seller exploits his dominant position in one market to expand his empire into the next.[19]

After remand to the district court and before closing arguments, the ISOs withdrew their § 1 tying and conspiracy claims. The remaining § 2 attempted monopolization and monopolization claims were submitted to the jury. A unanimous verdict awarded damages to the ISOs totaling $71.8 million after trebling.[20]

The Ninth Circuit affirmed, holding that the statutory intellectual property rights of a monopolist may not justify its refusal to deal, through licensing or otherwise making available its intellectual property, if a jury finds the refusal to deal to be pretextual.[21] However, the court rejected the district court's requirement that the parts be provided to ISOs at reasonable prices.[22]

The Ninth Circuit held that it was harmless error for the district court to fail to give any weight to Kodak's intellectual property rights (including 220 U.S. patents covering 65 parts for its high volume photocopiers and micrographic equipment and copyrights covering Kodak's diagnostic software and service software) in the jury instructions regarding business justification for Kodak's refusal to deal, because there was sufficient evidence to support a finding that a refusal to deal was a pretext for a willful extension of monopoly power.[23]

The court did state, however, that "'while exclusionary conduct can include a monopolist's unilateral refusal to license a [patent] or copyright,' or to sell its patented or copyrighted work, a monopolist's 'desire to exclude others from its [protected] work is a presumptively valid business justification for any immediate harm to consumers.'"[24] The court also admitted that it could "find no reported case in which a court has imposed antitrust liability for a unilateral refusal to sell or license a patent or copyright."[25] Nevertheless, the court, relying mostly on the Supreme Court's Kodak decision, ruled that the presumption of valid business justification can be rebutted by sufficient evidence that the alleged justification is pretextual, i.e., that the refusal to deal was not truly based on a desire to protect its intellectual property rights.[26]

Of course, the facts in Kodak involved much more than a mere unilateral refusal to deal; Kodak also secured agreements from its contractual OEMs not to sell parts to ISOs. These actions severely limited the ISOs' ability to compete in the service market for Kodak machines, because, naturally, competition in the service market requires that service providers have access to all parts.

In its defense, Kodak argued that its subjective motivation should be irrelevant and that a desire to beat the competition does not prove pretext or hostility to competitors. The court responded that when evidence suggests that the proffered business justification played no part in the decision to act, pretext could be shown. In support of that finding, the court noted that Kodak's parts manager testified that patents "did not cross [his] mind" when Kodak began its restrictive parts policy and also that the case involved many more non-patented parts or products (thousands) than ones that were patented (only 65).[27]

2. The 1997 Xerox Case: Can It Be Reconciled with Kodak?

In In Re: Independent Service Organizations Antitrust Litigation v. Xerox, 989 F. Supp. 1131, 1997 U.S. Dist. LEXIS 20996 (D. Kan. 1997) ("Xerox"), a Kansas federal court refused to apply the essential facilities doctrine to dictate the price Xerox was charging ISOs for its patented replacement parts. Although the Xerox court's decision is very critical of the Ninth Circuit Kodak case, the two cases can be reconciled.

The issue before the court in Xerox was much narrower than the issue decided in the Kodak case. While in the Kodak case the court had to decide whether Kodak should be required to provide ISOs with its patented and copyrighted replacement parts to remedy its antitrust violations, the Xerox court had to decide only whether it could dictate the pricing by Xerox of its IP covered replacement parts. Although the Xerox court in dicta liberally ventured into the question of compulsory licensing of IP in general, its holding was merely that on the facts of the case the court could not dictate prices Xerox charged ISOs for its replacement parts. That holding, therefore, was in agreement with the Kodak case which reversed the lower court's preliminary injunction requiring that Kodak sell its replacement parts to ISOs at reasonable prices.

Like Kodak, Xerox restricted ISOs' access to its product parts, including parts to its printers and copiers. CSU, one of the ISOs, claimed that Xerox's actions violated Section 2 of the Sherman Act by leveraging its monopoly to encompass the service market for its products. In 1994, Xerox settled an antitrust lawsuit brought by a class of ISOs in the United States District Court for the Eastern District of Texas (the "R&D Litigation"). CSU opted out of the R&D settlement on the same day that it filed the Xerox case. Pursuant to the R&D settlement, Xerox agreed to suspend its restrictive parts policy for a period of six and one-half years. The settlement also compelled Xerox to license diagnostic software, an essential component for service, for four and one-half years.[28]

CSU claimed in the Xerox case that after the R&D settlement, Xerox intensified its efforts to use price as a weapon to defeat ISO competition in the service market.[29] Xerox asserted several defenses. It contended that CSU did not suffer antitrust injury, because its alleged injury was attributable to Xerox's lawful refusal to sell patented parts and copyrighted software. Xerox also claimed that CSU could not assert a patent or copyright misuse defense to Xerox's infringement counterclaims based on Xerox's refusal to deal.[30]

The court found Xerox's pricing policy lawful. The court, in dicta, also addressed the issue of compulsory licensing in general. It stated that where a patent or copyright has been lawfully acquired, subsequent conduct permissible under the patent or copyright laws cannot give rise to liability under the antitrust laws.[31] Therefore, to the extent Xerox gained its monopoly power in any market by unilaterally refusing to license its patents, such conduct was, in this court's view, permissible. Xerox's legal right to charge the ISOs such royalties as it deemed appropriate arose from its intellectual property, not from an unlawful leveraging of its monopoly power in the parts market.[32]

The court stated that conduct which is expressly allowed by the patent laws cannot be illegal under antitrust laws.[33] The court relied on subsection 271(d)(4) of The Patent Reform Act of 1988, which provides:

No patent owner otherwise entitled to relief for infringement or contributory infringement shall be denied relief or deemed guilty of misuse or illegal extension of the patent right by reason of having . . . refused to license or use any right of the patent.[34]

The court stated that section 271(d)(4) should be interpreted to apply to antitrust claims. The court pointed out that the language "illegal extension of the patent right" closely tracks the antitrust law prohibition against an unlawful extension of market power beyond the limits of the patent monopoly.[35] The court also stated that the legislative history of section 271(d)(4) further supported the conclusion that the provision applies to antitrust claims -- Representative Kastenmeier had stated that "codification of the 'refusal to use or license' as not constituting patent misuse is consistent with the current case law and makes sense as a matter of public policy."[36]

The Xerox court stated that it makes little sense to preclude an infringer from asserting a misuse defense based on a patent holder's refusal to deal while simultaneously allowing the infringer to recover treble damages under the antitrust laws for the same conduct.[37] The court further pointed out that several courts have applied section 271 to both antitrust and patent misuse defense claims.[38]

The Xerox court found Aspen Skiing distinguishable for two reasons. First, Aspen Skiing did not involve intellectual property rights. The court stated that a unilateral refusal to license a patent should not be treated "like any other refusal to deal by a monopolist."[39] The court concluded that the general law regarding refusals to deal by a monopolist cannot logically be imported to the patent and copyright areas without seriously undermining the objectives of the intellectual property laws.[40] Moreover, the court found, CSU had not shown that analogous facts existed in its case, i.e., that competition in the service market required some cooperation among competitors or that consumers were harmed by Xerox's policies.[41] Interestingly, the latter ground on which the court distinguished Aspen Skiing showed that the Xerox court might consider compulsory licensing if it was necessary to preserve competition or to prevent harm to consumers.

The court rejected CSU's claim that there was antitrust liability because Xerox had an intent to monopolize. According to the Xerox court "[t]he standard articulated by the Ninth Circuit in Kodak [i.e., whether assertion of IP rights is merely a pretext,] makes it very difficult for a jury, a judge, or even the patent holder, to distinguish between a permissible refusal to deal (based on a desire to profit from and protect patent rights) and an impermissible refusal to deal (apparently based on a desire to obtain a competitive advantage by excluding competitors)."[42] It is because "[a] patent holder does not refuse to share its invention because the property is patented, rather, the refusal is based on a desire to obtain a competitive advantage." Therefore, "[t]o classify the desire to obtain a competitive advantage over competitors as pretext is to read the right to exclude out of the patent statute."[43]

The court pointed out that in essence, the difference between protected and unprotected conduct under Kodak is based on whether the company engaged in the formalistic ritual of documenting that the patents were the true reason of the refusal to deal.[44]

Finally, CSU argued that a refusal to license a copyright cannot be treated in the same manner as a refusal to license a patent because the Copyright Act does not contain a similar statutory provision to section 271(d)(4) of the Patent Act.[45] CSU maintained that section 271(d)(4) was a principal basis of the court's holding regarding patents.[46] The court found that although section 271(d)(4) lends additional support and is consistent with its ruling on patents, the court would reach the same result in the absence of section 271(d)(4).[47]

Although the Xerox court was very critical of the Ninth Circuit's Kodak decision, the holding of the two cases are not in conflict. In the Kodak case, the court found an antitrust violation and applied compulsory licensing as a remedy. It is somewhat unfortunate, however, that the Kodak court ventured into the confusing and bound to be highly controversial issue of "pretextual" assertion of IP rights. It appears that the court did not need to do that. Simply, when an antitrust violation is found, IP licensing is one acceptable remedy, among others, to restore a competitive market structure. Whether assertion of IP rights is "pretextual" or not should not matter after an antitrust violation has been shown on the merits.

The Kansas district court seems to have regarded the Kodak decision as more sweeping than it really was. Contrary to the Kansas court's fears (perhaps justified by the Ninth Circuit's analysis of the issue of "pretextual" assertion of IP rights), the Ninth Circuit did not hold that any "pretextual" assertion of IP rights will give rise to antitrust analysis, but rather that compulsory licensing of IP may be used as a remedy after a violation of Section 2 of the Sherman Act has been found. This is not a radical departure from well established antitrust remedies, which have often required divestiture of private property to remedy an antitrust violation and restore competition in a relevant market.

To understand the two decisions and reconcile them, it is necessary to keep in mind their different procedural postures. In the Kodak case, after the Supreme Court decision sustaining the complaint and after a full trial, the court found an antitrust violation and then required, as a remedy, that Kodak sell its replacement parts to ISOs. In Xerox, on the other hand, there had been a prior settlement of an antitrust action in a Texas district court which required Xerox to provide ISOs with needed replacement parts coupled with some IP licensing. The narrow issue before the Kansas district court was whether Xerox's pricing of the patented replacement parts was in itself an antitrust violation. The court held that it was not and (like the Kodak court rejected lower court's injunction forcing Kodak to sell its parts at "reasonable prices") it refused to interfere with Xerox's pricing of the replacement parts.

3. The Microsoft Case

The U.S. Justice Department and 20 states filed in May of this year an action against Microsoft Corporation ("Microsoft"), alleging violations of federal and numerous state antitrust statutes. The complaints against Microsoft allege that Microsoft:

"(1) unreasonably restrained competition by tying its Internet browser to Windows 98; (2) unreasonably restrained competition by entering into exclusive dealing arrangements with various Internet providers; (3) unreasonably restrained competition by imposing boot and start-up screen restrictions on original equipment manufacturers ('OEMs'); (4) illegally maintained a monopoly in its operating system software through various exclusionary and predatory practices, including but not limited to, the tying and exclusive dealing arrangements; and (5) attempted to monopolize the market for Internet browsers."[48]

The states also brought a separate claim of monopoly leveraging (arguing that Microsoft used its operating system monopoly to obtain a competitive advantage in the browser market) and pendent claims alleging violations of their respective antitrust laws.[49]

On September 14 the federal court for the District of Columbia denied Microsoft's motion for summary judgment on all claims except one -- the states' claim of monopoly leveraging.[50] Interesting for the purposes of this article is the court's analysis of the claim that Microsoft unlawfully restrained competition by imposing boot and start-up screen restrictions on OEMs.

When Windows 95 was first released, a number of OEMs who preinstalled Windows 95 customized the content and configurations of the computers' boot-up and first screens for commercial reasons. Some OEMs altered the arrangement, number and content of icons and folders which accessed Internet Access Providers ("IAPs"), Internet browsers and other software through the Windows 95 desktop. These OEMs entered into deals with IAPs and Internet Content Providers ("ICPs") that earned revenue for OEMs and garnered customers for the IAPs and ICPs. Indeed, in an internal e-mail, Bill Gates expressed his concern that OEMs were bundling non-Microsoft browsers and displaying them more prominently than the Microsoft browser.[51]

Allegedly to address this concern, Microsoft, as a condition of licensing Windows 98, began prohibiting OEMs from adding to the sequence of screens every user sees in the boot-up sequence and from modifying the first screen displayed to the user at the conclusion of the boot-up sequence. OEMs could not remove folders or icons from the Windows desktop, create icons or folders larger than those placed by Microsoft on the desktop, and could not alter the boot-up sequence by, for instance, presenting an OEM-created screen that would highlight a choice of Internet browsers or the OEM's own Internet offerings. Plaintiffs contended that the boot and start-up screen restrictions constituted an unreasonable restraint of trade in violation of Section 1 of the Sherman Act.[52]

Microsoft argued that its OEM license agreements merely highlighted and expressly stated the rights that Microsoft already enjoys under federal copyright law. Consequently, it contended, the agreements could not be challenged under the antitrust laws.[53] Microsoft argued that because it had no obligation to vend or license its copyrighted software, its actions were not unlawful.[54]

The court rejected Microsoft's arguments -- at the summary judgment stage -- because "copyright law does not give Microsoft blanket authority to license (or refuse to license) its intellectual property as it sees fit. A copyright does not give its holder immunity from laws of general applicability, including the antitrust laws."[55] Further, relying on the Supreme Court's Kodak decision, the court stated that "[c]opyright holders are restricted in their ability to extend their control to other markets."[56]

The court also rejected Microsoft's argument that its actions were protected under a copyright holder's "moral right of integrity." The "moral right" doctrine, which applies where a copyrighted work was significantly changed or "mutilated," but was promoted under its original name, was first recognized in Gilliam v. ABC.[57] The Microsoft court first pointed out that the "moral right" doctrine had been strongly criticized and many courts have refused to endorse it.[58] The court then stated:

Moreover, whatever policy justifications that may exist for a moral right of integrity in works of art are substantially weaker when the work at issue is a computer program, whose value lies in its functionality, not its artistry. The Copyright Act itself expressly allows owners of a copy of a computer program to "adapt" it in certain circumstances without the copyright owner's permission. Although Microsoft undoubtedly enjoys some "right against mutilation" in its software, there are significant factual questions in dispute on this issue, chief among them the extent of copyright protection in the specific portions of the software plaintiffs seek to modify.[59]

The court rejected Microsoft's argument that there was no liability because there was no market foreclosure since OEMs are free to configure the computers so that on all subsequent occasions (after the initial boot) they will boot directly into an alternative "shell" (such as Netscape Communicator). The court stated that several OEMs (and Microsoft executives) had expressed their belief that, while the licenses technically allow OEMs to engineer a mechanism for "user-initiated" action after the first boot-up is complete to alter the Microsoft-required screens, these factors are of negligible value to OEMs because plaintiffs contended that it was both costly and time-consuming for OEMs to develop such a mechanism.[60] Denying Microsoft's motion for summary judgment, the court stated that numerous factual issues remained as to "the extent of copyright protection in the specific portions of software plaintiffs [sought] to modify and whether Microsoft abused its copyright for anti-competitive purposes."[61] Moreover, the court stated that factual questions remained concerning "the legitimacy of Microsoft's claimed business justifications and Microsoft's claims that the restrictions do not foreclose competitors' opportunities."[62]

4. The Intel Case

Intel has provided its customers with technical information before the official commercial release of its new microprocessor products. This has made it possible for computer makers to have computers based on new Intel microprocessors ready for sale at or near the time of the official commercial release of the microprocessors. Although Intel considers the advance technical information to be proprietary, it makes the information widely available subject to nondisclosure agreements. This is part of the mutually beneficial relationship between Intel and its customers. Intel benefits because its customers -- computer systems manufacturers -- design and build new computer products that use the new Intel microprocessors. The customers, on the other hand, benefit by being able to introduce computer products with the latest and most advanced microprocessor technology on a timely basis.

On June 8, 1998, the FTC filed a complaint against Intel Corporation, the world's largest manufacturer of microprocessors, charging that Intel used its monopoly power to cement its dominance over the microprocessor market.[63] The FTC alleged that Intel illegally used its market power when it denied three of its customers continuing access to technical information necessary to develop computer systems based on Intel microprocessors, and took other steps to punish them for refusing to license key patents on Intel's terms. These three companies -- Digital Equipment Corporation, Intergraph Corporation and Compaq Computer Corporation -- hold important patents on microprocessor and related technologies. When they sought to enforce those patents against Intel or other computer companies who buy Intel products, Intel cut off the necessary technical information and threatened to cut off the supply of microprocessors.

The FTC alleges:

In each instance, Intel's conduct had a significant adverse effect on the ability of the targeted customers to develop and bring to market in a timely manner computer systems based on Intel microprocessors, and would have posed a more significant long-term threat to the business of those customers if they had not agreed to license their technologies to Intel . . . . Intel's coercive tactics to force customers to license away their [IP] rights diminishes the incentives of any firm dependent on Intel to develop microprocessor-related technologies. . . . Intel's conduct entrenches its monopoly power in the current generation of general-purpose microprocessors and reduces competition to develop new microprocessor technology and future generations of microprocessor products.[64]

Intel has conceded withdrawing the information and chip samples from the companies, but it has denied that it has a monopoly in the microprocessor market and has disputed the FTC's contention that its prototypes and advanced product information are "essential facilities" that customers must have access to in order to stay in business.[65]

Intel's response to charges that it violated antitrust law relies heavily on the view that it has a right to protect its intellectual property regardless of whether it has a monopoly. With regard to the issue of monopoly, while conceding it has the dominant market share, the company says that it does not have a monopoly in any relevant market.[66] Moreover, Intel has asserted that it has an "absolute right to refuse to license or share its intellectual property, including confidential information."[67]

In one portion of its filing, Intel argues that, even if it had monopoly power, it has the right to defend itself against lawsuits brought by its customers:

Intel's efforts to protect its intellectual property rights through the judicial system, including discussions and negotiations in anticipation of litigation or in attempts to avoid litigation, are protected by the First Amendment to the Untied States Constitution and the Noerr-Pennington Doctrine.[68]

The Noerr-Pennington doctrine gets one part of its name from the case Eastern RR Presidents Conference v. Noerr Motor Freight, Inc.,[69]where the Supreme Court rejected antitrust liability stemming from an aggressive lobbying campaign by railroads to persuade states to adopt legislation that would severely limit competition from truckers. The Court subsequently expanded the holding of Noerr to include activities aimed at the executive and judicial branches of government.[70]

Intel's general counsel, F. Thomas Dunlap, has said that FTC's action is "based upon a mistaken interpretation of the law and the facts."[71] Mr. Dunlap has also stated that "Intel intends to work through the administrative complaint process and, if necessary, appeal to a federal court."[72] Dunlap has stated that Intel had shared its intellectual property and early product samples for years with key customers, a practice that is beneficial to both parties.[73] Mr. Dunlap further added that "[a]t the same time, for more than 10 years, Intel has taken unprecedented steps to ensure that all of our activities and policies are in full compliance with existing law."[74] Dunlap has said that the FTC's decision signals "they want to change the very laws upon which we've based our policies."[75] According to Intel, the FTC's action is "an attempt to assert a new legal theory under antitrust law. . . [and] [a]lthough the key legal requirement of an antitrust claim is harm to competition, the FTC is unable to show harm to competition in any market."[76]

Mr. William Kovacik, who had worked at the FTC's Bureau of Competition and now teaches antitrust law at George Mason University Law School, agrees with Intel that the FTC is attempting to create new law.[77] He has said that, as stated by the FTC, the competitive harm is between Intel and its customers, who are only "potential competitors."[78] The FTC alleges that Intel prevents the customers from becoming competitors by "taking everything useful they have, vacuuming in all the information they have on microprocessors (through patent licensing agreements) so Intel can include that information in all its activities and, therefore, will always be ahead."[79] According to Mr. Kovacik, the FTC's chances of prevailing "are at best one in five."[80]

The FTC alleges in the complaint against Intel that the relevant market in which Intel allegedly has monopoly power is the market for general-purpose microprocessors.[81] At the same time, however, the FTC acknowledges that the customers -- Compaq Computer Corp., Digital Equipment Corp. and Intergraph Corp.[82]-- are each at least $1.8 billion away from even beginning to compete with Intel in that market.[83] The key question in this case will therefore be whether the customers are in fact potential competitors: could they enter into competition with Intel in a commercially viable way, and would they do so, absent Intel's licensing practices, within a year?

Intel, based in Santa Clara, California, reported revenues of approximately $25 billion and net income of approximately $6.9 billion for the fiscal year ended December 27, 1997.[84] The company designs, manufactures and sells a variety of semiconductor products, including a line of microprocessor products that are generally known, marketed and sold under the trade names Pentium, Pentium with MMX, Pentium Pro, and Pentium II.

a. Preliminary Injunction Against Intel in the Intergraph Case

In its action against Intel, the FTC is likely to rely on the Supreme Court's Aspen Skiing decision,[85] the two Kodak cases (the Supreme Court decision of 1992 and the Ninth Circuit decision of 1997), and Intergraph Corp. v. Intel Corp..[86] In Intergraph, the Federal District Court for the Northern District of Alabama granted a preliminary injunction requiring Intel to supply Intergraph with all technical and proprietary product information and product samples no later than Intel supplies the same information to similarly situated Intergraph competitors. Intergraph is an OEM which develops, manufactures and markets a series of high-end computer workstations. The court concluded that Intel's central processing units ("CPU") and the related technical support were essential facilities and even went so far as to hold -- in the context of a commercial agreement between two sophisticated parties -- that Intel's termination provisions in IP disclosure agreements were unconscionable.

In or about 1987, Intergraph purchased the Advanced Processor Division of Fairchild Semiconductor. As part of this acquisition, Intergraph acquired all rights to "Clipper," a "RISC"[87]-- based microprocessor that Intergraph was dependent upon for its workstation products. In 1997, as a prerequisite for Intergraph to participate in product development programs, Intel demanded that Intergraph sign new non-disclosure agreements which included broad license grants of Intergraph technology to Intel, including the technology covered by the Clipper patents. When Intergraph refused to sign the new non-disclosure agreements, Intel withdrew technical support to Intergraph relying on unilateral termination clauses in its existing non-disclosure agreements as contractual tools to refuse to provide technical information.

The court defined two relevant markets -- high performance microprocessors and Intel CPUs -- and found that Intel had willfully acquired monopoly power in each market. The court found the first element of monopolization under Section 2 of the Sherman Act -- monopoly power in the relevant market -- based on its findings that: (i) Intel's CPUs are not interchangeable with other CPUs; (ii) Intel's CPUs are superior in performance compared to CPUs of other companies; (iii) Intel has an enormous brand recognition; (iv) once customers are technologically and financially licked-in to an Intel CPU, they cannot feasibly switch to an alternative; (v) Intel has a huge installed base of customers locked in to Intel's network; and (vi) Intel has deliberately designed its related CPU architecture to prevent other CPUs from being compatible.[88]

The court ignored, however, the second element of a monopolization claim: whether Intel's monopoly was achieved by superior skill, foresight, industry, inventiveness or historic accident.[89] Nor did it analyze whether Intergraph was truly an actual or potential competitor of Intel. Instead, the court found that Intel's future product specifications and technical information were essential facilities.[90] The court concluded that "Intel's refusal to supply advanced CPUs and essential technical information to Intergraph likely violated Section 2 of Sherman Act, because they are not available from alternative sources and cannot be feasibly duplicated, and because competitors cannot effectively compete in the relevant markets without access to them.[91] Like the Ninth Circuit in Kodak, the Northern District of Alabama assessed Intel's motives in deciding whether Intel could enforce its IP rights. Citing Kodak, the court found that "Intel has no legitimate intellectual property basis with which it can refuse to supply Intel microprocessors and technical information to Intergraph, especially since Intel has been doing so for the last four years on a mutually beneficial basis."[92]

Rather than assessing subjective intent, the court would have done well to analyze whether, and if so to what extent, Intel's conduct has an objective effect of reducing competition in any relevant market in which both Intergraph and Intel are competitors.

On June 8, 1998, Intel appealed the preliminary injunction order to the U.S. Court of Appeals for the Federal Circuit. On appeal, Intel has denied that it has a monopoly in any relevant market and has argued that its behavior cannot possibly be construed as an attempt to leverage one monopoly into another. Intel's appellate brief notes that "the record does not suggest that Intergraph makes CPUs or that Intel makes workstations." Intel argues that the court, in granting the injunction, has failed to properly identify either the upstream market -- presumably some subset of CPUs -- or a downstream market into which it is trying to muscle a second monopoly. Intel contends that the two companies are not competitors in graphics subsystems, as the court suggests they are, and it points out that Intergraph's 10-K (filed after the lawsuit) does not list Intel as one of its competitors. The preliminary injunction, Intel says, is based on the district court's "misplaced desire to protect Intergraph rather than competition generally." According to Intel, the district court's legal conclusions "suggest a quest to find any legal basis to alleviate the perceived short-term harm to Intergraph's employees resulting from its management's decision that Intergraph is likely better off in the long run by suing Intel for billions of dollars."

But fundamentally, Intel argues that once the patent dispute broke out, it did what any company has a right to do: refuse to provide its own trade secrets and other confidential information to a litigation adversary. In blocking the sale of the test equipment to Intergraph, for example, Intel says it was exercising a right which it clearly has: "the right to terminate the NDAs at any time for any reason, without cause, upon notice."

Intergraph argues that Intel's appeal concentrates on only two claims -- monopoly leveraging and attempted monopolization -- and ignores remaining charges of illegal maintenance of monopoly, denial of access to essential facilities, unlawful refusal to deal, misuse of monopoly power and coercive reciprocity. By law, Intergraph said, it needed to establish likelihood of success on only one of these grounds to obtain the preliminary injunction. Intergraph has also claimed that the court's findings were substantiated by the FTC's June 8 antitrust lawsuit against Intel.

In the case of the merger of Ciba-Geigy and Sandoz, each of which maintained a gene therapy lab, the FTC required a spin off of one party's patents, so that independent research would continue after the merger. Sandoz agreed to license some gene therapy technology to Rhone-Pulenc Rorer, Inc., so that the firm could compete against the merged companies.[93] The EU also does not hesitate to call for conditions or substantial alterations to deals before giving the go-ahead. In Europe, Ciba-Geigy and Sandoz had to agree to license active substances for flea-killing compounds, over which the EU concluded they held dominant control.

American Home Products Corp.'s acquisition of Solvay, S.A.'s animal health business, was allowed by the FTC only after Solvay's North American rights to three types of vaccines were divested to another company.[94]

Similarly, the FTC cleared the acquisition of Immuno International AG by Baxter International Inc., manufacturers of a wide variety of biologic products derived from human blood plasma, to create the largest manufacturer of plasma products in the world.[95] To remedy the alleged anti-competitive consequences, the consent order required Immuno to license its sealant product Tisseel[96] to an approved licensee, Haemacure Corporation.