The Antitrust and IP Interface in the US and EU

By Richard Eccles, Romain Ferla


According to the US courts, both competition law and intellectual property (IP) law "are aimed at encouraging innovation, industry and competition" (Atari Games v Nintendo, 897 F.2d at 1576 [14 USPQ2d 1034] (Fed Cir 1990)). Similarly, the European Commission considers that there is no "inherent conflict between intellectual property rights and the Community competition rules" (Notice providing guidelines on the application of Article 81 of the EC Treaty to technology transfer agreements (OJ 2004 C101/02) (Guidelines on Technology Transfer Agreements)).

However, in practice, there has always been considerable tension between the two bodies of law as IP law provides for government granted temporary monopolies, while competition law prevents monopolisation.

Courts and policy makers need to find the optimal balance between the two bodies of law, so that innovation can be encouraged in the short term, with anti-competitive behaviour kept to a minimum, thereby safeguarding incentives for future inventions. Competition authorities and the courts on both sides of the Atlantic have recently devoted much attention and effort to striking the appropriate balance within their own jurisdictions. This chapter describes recent examples of these efforts. The task of promoting convergence across jurisdictions also deserves attention. Progress has been made through bilateral and multilateral bodies, but much work remains to be done. In the light of these developments, this chapter:

  • Examines the current legal approaches of the US to the antitrust and IP interface.
  • Looks at recent trends, issues and developments in the US.
  • Examines the current legal approaches of the EU to the antitrust and IP interface.
  • Looks at important developments in the EU's approach to the compulsory licensing of IP rights.
  • Compares the approaches of the two jurisdictions.
  • Discusses anticipated future developments in the US and EU.
  • Identifies opportunities for further convergence.


The principal anti-trust laws applicable to IP licences in the US are sections 1 and 2 of the Sherman Act (15 USC 1-2 (2004)). These laws prohibit:

  • Agreements that unreasonably restrain trade.
  • Conduct that may create or maintain a monopoly.

Over the last 20 to 30 years, there have been great changes in the attitudes of the Federal Trade Commission (FTC) and Department of Justice (DoJ) (referred to collectively as the Agencies) in relation to the role of IP in the competitive process. These range from systematic suspicion of IP licensing in the 1970s to a very lenient approach in the 1980s, to the more recent view that transactions involving IP will generally - but not always - promote, rather than restrain, competition.

Anti-trust guidelines for the licensing of IP

In 1995, the Agencies issued guidelines for the licensing of intellectual property (IP Guidelines) to help clarify their enforcement position. The main features of these IP Guidelines are described below.

Rule of reason and per se illegal restraint. In most cases, restraints in IP licensing arrangements are to be evaluated under the rule of reason (that is, by comparing a restraint's pro- and anti-competitive effects) (IP Guidelines).

However, certain categories of restraints are "so plainly anticompetitive that [they] should be treated as unlawful per se without an elaborate inquiry into [their] likely competitive effects" (section 3.4, IP Guidelines). These per se illegal restraints include:

  • Naked price-fixing.
  • Output restraints.
  • Market division among horizontal competitors.
  • Certain group boycotts.
  • Minimum resale price maintenance.

The IP Guidelines also establish safe harbours in which conduct is considered lawful, in the absence of extraordinary circumstances. The safe harbours differ according to whether the relevant market is a goods market, a technology market or an innovation market:

  • Goods market. A proposed licensing arrangement falls within the safe harbour if both:
    o it is not facially anti-competitive;
    o the licensor and its licensees collectively account for no more than 20% of each relevant goods market affected by the restraint.
  • Technology market. A proposed licensing arrangement falls within the safe harbour if both:
    o it is not facially anti-competitive;
    o there are at least four independently controlled substitute technologies.
  • Innovation market. A proposed licensing arrangement falls within the safe harbour if both:
    o it is not facially anti-competitive;
    o there are at least four independently controlled entities, in addition to the parties, that have the incentive, assets and characteristics necessary to engage in research and development (R&D) that is a close substitute for the activities of licensee and licensor.

Treatment of specific licensing practices. The IP Guidelines provide the Agencies' with the following general views on specific types of licensing arrangements:

Resale price maintenance. The Agencies treat restrictions that set the price at which a licensee must contract as per se unlawful (section 5.2, IP Guidelines).

  • Tying arrangement. The Agencies are likely to challenge a tie-in when all the following conditions are satisfied (section 5.3, IP Guidelines):
    o the licensor has market power in the tying product;
    o the arrangement adversely affects competition in the relevant market for the tied product;
    o any anti-competitive effects are not outweighed by efficiency justifications.
  • Exclusive dealing. The Agencies apply the rule of reason to arrangements that preclude the licensee from using competing technologies and focus on the extent to which the arrangement (section 5.4, IP Guidelines):
    o promotes efficient use of the licensor's technology;
    o adversely affects competition among competing technologies.
  • Cross-licensing and pooling arrangements. These arrangements are tested under the rule of reason (section 5.5, IP Guidelines). Pro-competitive considerations include the reduction of transaction costs and the integration of complementary technologies. Anti-competitive considerations include:
    o whether the arrangement will reduce competition between actual or potential competitors;
    o whether the arrangement may adversely affect incentives to engage in R&D (for example, cross-licensing of future inventions may discourage R&D efforts);
    o whether the pooling arrangement is open or closed; in the latter case the arrangement can be challenged if the pool participants collectively possess market power and the excluded firms cannot effectively compete without a licence.
  • Grant-back arrangements. These arrangements will be evaluated under the rule of reason, focusing on a number of factors, including:
    o whether the provision will reduce the licensee's incentive to improve the licensed technology;
    o whether the licensor has market power in the relevant technology or innovation market;
    o whether the grant-back requirement is exclusive.


Settlements of patent disputes, particularly in the pharmaceuticals Industry

In recent years, settlements of patent disputes in the pharmaceutical industry arising between pioneer drug manufacturers and potential generic competitors have been under considerable scrutiny by the Agencies. This general trend has been reinforced by the Medicare Prescription Drug, Improvement and Modernization Act 2003, which requires all drug companies, including brand-name companies and generic makers, to file certain agreements with the FTC and DoJ. Under this Act, filing requirements began on 7 January 2004 and drug companies must file all brand-generic settlement agreements with the anti-trust regulators within ten days of their execution.

Schering-Plough. Suspicion about brand-generic patent settlement agreements have recently been softened by the Eleventh Circuit decision in Schering-Plough (Schering-Plough Corp v FTC, 402 F.3d 1056 (11th Cir. 8 March 2005). In this case, Schering-Plough manufactured and marketed a pioneer drug on which it had a patent (743 patent). Schering filed patent infringement suits against two companies, Upsher and ESI, which manufactured generic versions of the drug. However, before trial in either suit, Schering reached a separate settlement with each company. In each case, Schering agreed that the generics could enter the market in advance of the expiration of the 743 patent - nearly six years ahead in the case of Usher and three years in the case of ESI - and then paid them for exclusive licences on other products and, in ESI's case, for its lawyers' fees. The FTC challenged these settlement agreements and concluded that Schering's payments to the generic makers were not "legitimate consideration" for licences, but that they were in fact part of an arrangement to delay entry dates, and that the agreements harmed competition and consumers as a result.

However, on appeal, the Eleventh Circuit found the licence payments from Schering to Upsher and ESI to be legitimate and bona fide, and not simply shams to delay generic entry. There was no indication that Upsher and ESI's products could have entered the market before the expiration date of the patent, and the settlement agreements provided for market entry before the expiration of the patent. Interestingly, the Eleventh Circuit held that patent settlements within the scope of the patent at issue should generally be favoured since they may enhance efficiency and spur innovation.

Terazosin II. All patent settlement agreements may not receive the same favourable treatment by the courts as in Schering-Plough. For example, in Terazosin II, a settlement agreement whereby an accused patent infringer agreed to stay out of the market pending appeal in exchange for payments was held per se illegal under section 1 of the Sherman Act (Re Terazosin Hydrochloride Antitrust Litigation (S.D. Fla. 5 January 2005 (Terazosin II)).

This case followed a patent infringement litigation initiated by Abbott against a generic maker, Geneva. As the patent district court had originally found the patent to be invalid, the Southern District Court of Florida held that the parties' agreement to delay generic entry pending appeal was illegally extending the patent's protections since there was no determination of whether Abbott was likely to succeed on the merits of any appeal. The court further found that the agreement did not maintain the status quo pending appeal, because the agreement added a large payment from the patent holder. A "true status quo", the court noted, would have had Geneva agreeing not to market its product without payments from Abbott.

In summary, the court found that the appellate-stay provision was not counterbalanced by any otherwise unobtainable pro-competitive benefits. Therefore, it concluded, the provision was a "naked, not an ancillary, restraint of trade, which is properly condemned under per se analysis".

Unilateral refusals of IP rights holder to license its IP

The traditional view of US courts that the unconditional and unilateral refusal by an IP owner to license its IP rights is justified was recently reinforced by declarations made by representatives of the Agencies in the course of 2004.

Commenting on the Supreme Court decision in Trinko (Verizon Communications Inc v Law Offices of Curtis v Trinko LLP (2004 WL 51011 (2004))), Assistant Attorney General R Hewitt Pate stated that Trinko "clarified that there is no basis in US antitrust law for a stand-alone essential facilities doctrine [of liability]" and "expressed profound skepticism that the antitrust laws were intended to create a duty by one competitor to assist its competitors by assuring them access to its tangible or intellectual property" (for the full text, see

Deputy Assistant Attorney General for Antitrust Policy Makan Delrahim, also citing Trinko, stated that while "compulsory licensing as a merger remedy is a well-established tool … nonmerger compulsory licensing imposed by an agency or the courts should be a rare beast" due to the difficulty of administering the remedy and the possibility of reducing innovation incentives (for the full text, see

Despite the reluctance of the Agencies and the courts to grant compulsory licences on IP rights, some claimants still denounce the unilateral refusal to license patents or copyrights, which constituted the core of a few interesting cases over the last months.

Telecom Technical Services. In Telecom Technical Services Inc v Rolm Co (388 F.3d 820 (11th Cir. 2004)), defendant Siemens manufactured private branch exchanges (PBXs) (that is, systems which direct telephone calls through a network of extensions). Siemens held IP rights over some of the hardware and all of the software used in its PBXs. Siemens licensed its PBX system and sold PBX servicing, but did not sell its software or parts to third parties for resale. An independent repairer hired by a customer for the maintenance of a Siemens PBX could only obtain parts from Siemens by providing a letter of agency by the customer authorising the purchase of spare parts on the customer's behalf. The claimants, mostly independent repairers specialising in PBX servicing, alleged that Siemens "created a monopoly in the market for servicing Siemens's PBX". The district court dismissed the claimant's claims on the ground that Siemens's refusal "to sell patented parts or copyrighted software" could not violate antitrust laws. This rather extreme view was not endorsed on appeal by the Eleventh Circuit. However, the Eleventh Circuit did confirm the district court's judgment but, instead of holding that an anti-trust claim could not be based on a unilateral refusal to license IP rights, it retained that there was no evidence that PBX users were actually harmed, since, despite Siemens's behaviour, they could still have their PBXs serviced by independent repairers.

Monsanto Co. Monsanto Co v McFarling (363 F. 3d 1336, 1342 Fed. Cir. 2004) was another interesting case of unilateral refusal to license IP rights in 2004. Monsanto sued for patent infringement after McFarling, a farmer, breached a biotechnology licence by replanting seeds from a crop grown from Monsanto patented soybeans. McFarling alleged that Monsanto tied a patented product (the original seed) to an unpatented "God-made" second generation soybean seed, prohibiting seed-saving of a type that was allegedly normal among farmers. The Federal Court rejected McFarling's arguments, reasoning that Monsanto had only used the legitimate powers of its patent: "At its simplest, McFarling effectively argues in different words that he should be granted a compulsory license … We decline to hold that Monsanto's raw exercise of its right to exclude from the patented invention by itself is a "tying" arrangement that exceeds the scope of the patent grant."

Misrepresentations to a government body in order to delay market entry

Under certain circumstances, misrepresentations in relation toCross-border October 2005, the FTC, Akzo Nobel NV, and its subsidiary Organon USA Inc, settled a case involving misrepresentations to the FDA to delay generic entry. Charges against Organon involved "fraudulent misrepresentation" to the FDA about the claims of a patent listed on the FDA's Orange Book, so as to delay by about eight months the introduction of generic competition to its antidepressant, Remeron. Under the settlement, Organon agreed to pay tens of millions of dollars in damages and to comply with strong injunctive terms barring future anti-competitive conduct (

Unilateral patent ambush is not limited to misrepresentation cases. It can also involve failure to disclose IP related to the standards being developed by a non-governmental standardsetting organisation (SSO). One well-known recent example concerns Rambus Inc, which develops technologies used in computer memory devices. Rambus participated in the work of an SSO, the Joint Electron Device Engineering Council (JEDEC), without revealing to JEDEC or its members that it possessed a patent and pending patent applications that involved technologies ultimately adopted in JEDEC standards. Rambus subsequently sued memory device manufacturers for patent infringement. Although the FTC complaint against Rabus was initially dismissed by the Administrative Law Judge (Rambus Inc, FTC, Dkt No 9302 (Initial Decision), see, an appeal by FTC staff to the full Commission is pending, and in May 2005, Rambus acknowledged the FTC was reopening the case to include new evidence against the chip manufacturer.

Exclusion from standard-setting processes

SSOs can give rise to another kind of anti-competitive behaviour, which consists in excluding some market players from the standard-setting process so as to prevent these operators from input or access to the standard during their development process. This kind of abuse was recently highlighted in a publicised complaint (filed with the District Court of Delaware) by the microchip manufacturer, AMD, against the market leader, Intel ( In this complaint, AMD alleges that Intel is excluding it from the Advanced DRAM Technology (ADT) Consortium, an SSO developing standards for new generations of computer memory chips. According to the complaint, Intel structured the ADT Consortium with multiple tiers of membership, each with different levels of access to information, and consigned AMD to the lowest tier, therefore ensuring that it would receive access to the memory standard only on its completion, but not during its development. The complaint also alleges that Intel tried to influence unduly standard-setting processes before another SSO (JEDEC) so as to disadvantage AMD by forcing it to arrange costly modifications of its products.

Patents and presumption of market power

To establish that a tying arrangement constitutes a per se violation of section 1 of the Sherman Act, an anti-trust claimant must prove that the defendant has "appreciable market power" in the tying product market (Eastman Kodak v Image Technical Services Inc. (504 US 451, 462 (1992)). In January 2005, a Federal Circuit Panel recognised a special exception for patented products, holding that a patent creates a presumption that its owner has market power in a relevant market consisting of the patented product (Independent Ink v Illinois Tool Works Inc (Fed Cir. 25 January 2005)).

The case is currently pending before the US Supreme Court. The FTC has filed an amicus curiae brief supporting Illinois Tool Works (the patent holder) and submitting that there is no economic basis for inferring market power from the mere fact that the defendant holds a patent. This position is consistent with the IP Guidelines.

Government initiatives on the proper balance of competition and patent law and policy

Defining a better balance of competition and patent law is still an ongoing process.

In October 2003, the FTC issued the first of two reports on how to promote innovation by finding the proper balance between competition and patent law and policy. This report contains conclusions and recommendations addressing the patent system (report available at The second report which should make similar recommendations for anti-trust law has not yet been issued.

Following the FTC's 2003 Report, a series of hearings were organised in 2005 in order to initiate a discussion among all stakeholders in the patent system on the reform proposals made by the FTC. Similar meetings were also held by the National Academies' STEP (NAS) Board (which released a report in April 2004) and by the American Intellectual Property Law Association, in the hope of eventually achieving a consensus on the content of needed reforms.

In October 2004, the DoJ released a Report on Intellectual Property. Although this report was not issued by the Antitrust
Division (but by the DoJ's Task Force on Intellectual Property) it recommended three interesting anti-trust initiatives:

  • Supporting the rights of IP owners to determine independently whether or not to license their technology.
  • Encouraging the use of the DoJ's business review procedure for guidance on anti-trust concerns relating to industry standards for the prevention of IP theft.
  • Promoting international co-operation on the application of anti-trust laws to IP rights.


As recognised by the European Commission, there is no "inherent conflict between intellectual property rights and the Community competition rules" as "both bodies of law share the same basic objective of promoting consumer welfare and an efficient allocation of resources" (section 7, Guidelines on Technology Transfer Agreements). EC competition authorities recognise that the monopoly situation generated by the holding of IP rights does not hinder competition as, in the long term, it contributes to an innovative and competitive economy.

In particular, the European Court of Justice (ECJ) found aCross-border compromise between IP law and competition law by drawing a distinction between the "existence" of IP rights and their "exercise". While EC competition law never interferes with the "existence" of IP rights, it may affect how these IP rights are exercised. Indeed the exercise of IP rights and, in particular, the conditions of their transfer, is subject to Articles 81(1) and 82 of the EC Treaty:

  • Article 81(1) prohibits agreements which have as their object or effect the restriction of competition.
  • Article 82 prohibits abuses of dominance.

Most licensing agreements do not fall into the scope of Article 81(1), as they usually do not restrict competition and create procompetitive efficiencies by disseminating technology and promoting innovation.

In this respect, in April 2004, the Commission issued Regulation (EC) No. 772/2004 on the application of Article 81(3) of the EC Treaty to categories of technology transfer agreements (Technology Transfer Block Exemption Regulation) which
provides a "safe harbour" for companies willing to ensure that their licensing agreements do not infringe Article 81. Unlike the former block exemption regime (Regulation (EC) No 240/96 on the application of Article 85(3) of the EC Treaty to certain categories of technology transfer agreements), the Technology Transfer Block Exemption Regulation provides for an effects-based approach similar to that of the IP Guidelines in the US: whether the Technology Transfer Block Exemption Regulation applies to an agreement will now depend on the market shares of the parties. The relevant thresholds are:

  • 20% (combined market share) of the relevant market where the parties are competitors or potential competitors.
  • 30% (individual market share) of the relevant market where they do not compete.

This approach is fundamentally different from the one endorsed under the former block exemption. The former block exemption was drafted along the distinction between "white-listed clauses" and "black-listed clauses". However, even though it was prescriptive, it still offered the parties to an agreement the benefit of security and confidence, as they could know for sure that the block exemption would apply provided the agreement was drafted accordingly.

The new Technology Transfer Block Exemption Regulation, with its reliance on market shares, creates uncertainty. However, it also creates a greater flexibility, as the former "white list" of permitted clauses is abolished and the "black list" of prohibited clauses, known as "hard-core restrictions", has been reduced in scope and adapted to the nature of the relationship between the parties (different hard-core restrictions apply depending on whether the parties are competing or non-competing undertakings). With respect to Article 82 of the EC Treaty, the European Commission (and the competition authorities of member states) considers that the overall contribution of IP rights to innovation does not entitle IP owners to refuse to license their rights in all circumstances.

In several decisions, the ECJ made it clear that the owners of IP rights do not enjoy a complete immunity in the exclusive exploitation of their rights under the EC Treaty. Even more, under "exceptional circumstances" and where a particular exercise of IP rights by their owners conflicts with Article 82 of the EC Treaty, EC competition authorities reserve the right to order a compulsory licence.


On 29 April 2004, the ECJ delivered its judgment in IMS (IMS Health GmbH & Co OHG v NDC Health GmbH & Co KG, 29 April 2004 (Case C-418/01)). This case is of central importance as it signifies the ECJ's current opinion on the issue of the application of the essential facilities doctrine in a context where the facility at stake is protected by IP rights.

In its judgment, the ECJ repeated the assessment contained in the Oscar Bronner judgment (Oscar Bronner GmbH & Co KG v Mediaprint Zeitungs, 26 November 1998 (Case C-7/97)), according to which the application of the essential facilities doctrine requires a demonstration that the replication of a similar facility by a company of comparable size would not be economically viable. The ECJ added that whether the participation by customers in the development of the facility constituted an additional barrier to the creation of a competing facility was a factor that should be taken into account by the national court when making its assessment of whether the copyright was essential. Finally, regarding the refusal to issue a licence, the ECJ held that for a refusal to be regarded as abusive it must:

  • Prevent the emergence of a new product or service for which there is a potential demand.
  • Be without objective justification.
  • Be capable of eliminating all competition on the relevant market.

Unfortunately, the concept of a new product was not elaborated on and remains for the national court to define.

That very recent position of the ECJ must also be read in the light of another very recent case, handled by the Commission - the 2004 Microsoft decision (Microsoft/W2000 (Case COMP/ 37.792) (23/03/04)) - which also specifically concerned, among other matters, the possible abusive use of IP rights. Among several abuses, Microsoft was condemned for failing to release key specifications to competitors to enable their computer systems to interact properly with Windows software. This conclusion was clearly based on a refusal to supply IP information. In substance, the Commission held that even though Microsoft's interoperability protocols may be protected by IP rights, a refusal by a dominant undertaking to provide such information to its competitors "may, under exceptional circumstances, be contrary to the general public good by constituting an abuse of a dominant position with harmful effects on innovation and on consumers" (paragraph 711, Microsoft/W2000).

Although the European Commission denied any conflict betweenCross-border the ECJs' judgment in the IMS case and its prohibition decision of 24 March 2004 against Microsoft (, it seems likely that the Commission may have a hard time trying to reconcile its decision against the software giant with the conditions set out by the ECJ in IMS for compulsory licensing of IP.

Microsoft filed an appeal with the Court of First Instance (CFI) against the Commission's decision. Microsoft also filed a request to suspend the remedies ordered against it. While the latter was dismissed by the CFI on 22 December 2004, the appeal on the merits is still pending and should be heard in the coming months. In its request for a suspension of the remedies, Microsoft submitted that, by requiring it to grant licences to its competitors, the Commission's decision forces it to disclose certain information in breach of its IP rights, which would cause serious and irreparable harm to its activities.

The President of the CFI acknowledged that requiring an undertaking to issue licences affecting its IP rights is a substantial
breach of the exclusive prerogatives which the undertaking derives from those rights. However, he also held that it was a
necessary breach under the principle established in IMS, since the examination carried out by the Community judicature
consists specifically in weighing up the protection conferred by an IP right on its holder against the requirements of free competition set out in the EC Treaty (paragraphs 248 and 249, Order of the President of the CFI of 22 December 2004).

Therefore, such a breach of IP rights does not constitute, in itself, a serious and irreparable harm justifying the suspension of remedies. As Microsoft did not establish any other irreparable harm, the President of the CFI refused to grant interim measures. Even though its request for suspension was rejected by the CFI, the result of Microsoft's appeal on the merits is far from being certain. Indeed, when hearing the appeal on the merits, the CFI will be faced with a very difficult choice to make, as many believe that a restrictive decision in such a high-profile case would undermine the incentives to innovate.


As shown by the developments above, case law regarding the interface between IP and competition law is dense and evolving gradually. In addition, the variety of IP rights and the difference in market power that they can grant should also not be underestimated; indeed, the market power held by the owner of a patented good which has become an industry standard could hardly be compared to the one held on one of many goods on a given market. To put it simply, holding all IP rights on a standard interface or software certainly has more leverage potential than holding a patent on a medicine which is about to fall in the public domain. Therefore, any attempt to succinctly summarise case law, particularly on a cross-border US and EU basis, would necessarily be simplistic.

However, a few important analogies and differences on the IP and anti-trust interface clearly do stand out between the two systems and can therefore be highlighted.

Similarities between the US and the EU

The main important similarities between the US and the EU on the IP-anti-trust interface can be summarised as follows:

  • Neither jurisdiction presumes that IP rights automatically confer market power.
  • Licensing agreements are analysed under an economic effects-based approach and are generally considered as procompetitive unless they either:
    o contain hard-core restrictions (price-fixing, output restraints, market division or customer allocation);
    o involve parties with combined market shares reaching specified levels.

With the recent Astra Zeneca decision (Generics/Astra Zeneca (Case COMP/37.507) (15/06/2005)) decision provided by the European Commission, there is now clear and solid case law on both sides of the Atlantic to challenge abuses of governmental procedures in relation to IP rights. In particular, the misuse of IP rights systems and procedures aimed at unduly perpetuating IP protection and, thereby, at blocking or delaying potential competitors' from entering the market.

Differences between the US and the EU

The main differences between the US and the EU regarding the anti-trust-IP interface can be summarised as follows:

  • While unilateral patent ambush has given rise to several landmark cases in the US over the past few years, European competition authorities have not yet been able to challenge successfully this kind of behaviour. This probably arises from the fact that EU competition law, contrary to its US counterpart, does not prohibit the creation of dominance per se but the abuse of a dominant position. Therefore, in the EU, unilateral patent ambush can only be caught when implemented by a company which is dominant in the first place. However, the Commission is said to be currently investigating allegations of patent ambush by Rambus, in the memory chip sector (see Financial Times, 23 June 2005). These allegations are the same as the ones pursued by the FTC.
  • Territorial restrictions in licensing arrangements are rarely characterised as unlawful in the US. In the EU, however, territorial restrictions are more likely to violate competition laws, probably because market integration remains a priority as the internal market is still under construction.
  • While in the US, unilateral refusals to deal are almost never treated as anti-trust violations, EU competition authorities have analysed refusals to license IP rights as abuses of dominance in three relatively recent cases.



In the US, in the light of the past and recent evolutions in case law, it seems likely that the Agencies will continue to apply their pro-IP approach. Indeed, recent speeches by representatives of the Agencies indicate that they will be increasingly reluctant to force IP rights holders to grant licences to their competitors and that they will tend to limit their intervention in the anti-trust and IP interface to blatant cases of monopolisation that directly harm consumers, such as misrepresentations to the patent office to delay market entry or unilateral patent ambush. As shown by the recent proposals of the DoJ's Task Force on Intellectual Property (which follows an earlier Report issued by the FTC in October 2003), the Agencies will also probably continue to promote the reform of IP law at the "upstream" level, in order to remove the factors that encourage patent offices to lower their guard and approve applications carelessly.


That same trend of addressing the issues at the upstream level also seems to emerge in EC competition law. The Microsoft case clearly illustrated the intrinsic difficulties involved in granting compulsory licences on IP rights which are held by dominant companies. The European Commission may therefore also be tempted to treat the potential competitive issues at a more "upstream" stage, in particular through merger control. In this respect, the recent Contentguard case (IP/05/295) could represent a clear sign by the Commission that it is not ready to let super-dominant companies create new monopolies through the acquisition of strategic IP rights and technologies. In that particular case, Microsoft and Time Warner withdrew their notification regarding their proposed acquisition of joint control on ContentGuard. Further to that notification, the Commission had concerns that Microsoft may attempt to block competitors' access to ContentGuard's valuable Digital Rights Management patents and, consequently, launched an in-depth investigation. The Commission held that, following a substantial change in ContentGuard's governing rules, and the entry of Thomson as a key shareholder, Microsoft would no longer have the ability to impose in ContentGuard a licensing policy that would put its rivals in the digital rights management market at a competitive disadvantage. Notwithstanding the above, the coming trends of the EU approach to the IP and competition law interface will also largely be shaped by the coming and much awaited judgment of the CFI in the Microsoft case. Considering the economic importance of the undertakings involved and the publicity that this judgment will certainly be given, it could set the tone of European policy with respect to the anti-trust and IP interface in the coming years. The issue of the possible qualification of the exercise of IP rights as an abuse of a dominant position is also being examined in the ongoing reform conducted by the European Commission on the application of Article 82 of the EC Treaty. This reform is a great opportunity for the Commission to clarify or adjust its approach to abusive use of IP rights and it is expected to make specific proposals on this issue.

This Article 82 review should also be read in the light of the report titled "An economic approach to Article 82", which was
issued in July 2004 by the Economic Advisory Group on Competition Policy (EAGCP), an independent group of experts commissioned by the Chief Economist of the Directorate General for Competition (DG Comp). This report was specifically commissioned to provide the Commission with an economic approach to Article 82 and, consequently, to give an opinion on the reform of the Commission's policy on the abuse of dominant position. Regarding the specific issue of compulsory licensing of IP rights, the report endorses a prudent approach, similar to the US Supreme Court's decision in Trinko. It argues that even if a refusal to deal does harm consumers in the short-run, it may be socially beneficial in the long-run, pointing to the need to preserve the incentive to innovate.

Although the EAGCP's report does not the present the European Commission's position on essential facilities and compulsory licensing, it illustrates what seems to be a growing trend in Europe to stress the importance of IP rights and, sometimes, its supremacy over competition law.

That growing trend also seems to be endorsed by DG Comp. For example, the European Competition Commissioner, Mrs Neelie Kroes, made clear that promoting innovation was a top priority on her agenda ( In addition, DG Comp released a consultation document on state aid for innovation ( which, among other things, identifies the "problems affecting innovation in Europe" (pages 18 to 19). Among those problems, the Commission stresses the "unsatisfactory IP protection" and the "unattractive risk/reward ratios for investing in radically innovative products". In light of the above, it appears that in their search of the appropriate balance between
IP and competition law, DG Comp officials currently do not seem to favour "compulsory licensing", suggesting that it will be used as a regulatory tool only in exceptional circumstances.

At the member state level, a recent decision of the French Supreme Court (Cour de Cassation) reinforces this trend. This decision, in NMPP v MLP (12 July 2005), rejected the qualification of essential facility for a software, although this qualification had been retained prima facie by both the Competition Council and the Paris Court of Appeal (see French Supreme Court decision in essential facilities case at

However, in June 2005, the Autorità Garante della Concorrenza e del Mercato, the Italian competition authority, gave hope to compulsory licensing partisans by ordering Merck (Decision of 15 June 2004, Case A364, Merck-Principi Attivi, IAA Bulletin No. 23/2005) to license the production of imipenem cilastatina (an active ingredient in an antibiotic called Tienam, used for the treatment of particularly serous infections, most often contracted in hospitals) for export sales. Although Merck had a patent for Tienam in Italy, the patent had expired in all other European countries. The decision follows an investigation opened in February 2005, following Merck's alleged refusal to grant a licence for the Italian production and export of the active ingredient for the manufacture of generic drugs in other European countries. Cross-border NMPP and Merck show that a decentralised enforcement of EC competition law can lead to very different outcomes from one member state to the other, especially in complex IP and anti-trust cases. Therefore, there is a clear need for the Commission and the ECJ to set clear guidelines on the application of Article 82 of the EC treaty to IP rights holders. However, the new Technology Transfer Block Exemption Regulation was praised in the US as a great sign of transatlantic convergence (see the recent speech by Deputy Assistant Attorney General of the Antitrust Division of the Department of Justice, Mr Makan Delrahim at


The interface between anti-trust and IP law is probably one of the areas where important differences remain between the US and the EU. In this respect, the Commission's decision in the Microsoft case attracted much criticism from US practitioners and enforcers: some US Congressmen have even claimed that the European anti-trust actions against Microsoft have violated the anti-trust agreement between the US and the EU (European Competition Law Review, June 2005, page 309).

In the current environment, US and EU co-operation and harmonisation efforts usually focus on merger control and international cartels, rather than on the IP and competition law interface. For example, the European Commission and the Agencies are working together to promote convergence among the world's anti-trust enforcers, notably within the International Competition Network (ICN), which held its fourth annual conference in June 2005. However, until now, the ICN has mostly concentrated its efforts on merger control, issuing recommendations to improve competition agencies' merger review processes. The IP and anti-trust interface is not yet on the agenda.

Finally, it is worth noting that on 18 May the European Commission adopted a Communication entitled, "A stronger EU-US Partnership and a more Open Market for the 21st century". It contains a wide range of practical policy proposals for a joint EUUS strategy to boost economic integration, and to strengthen the broader framework of EU-US relations. In particular, the Communication acknowledges that the EU and US economies have become ever more intertwined and therefore calls for better and reinforced regulatory co-operation in the areas of anti-trust policy. This should constitute an appropriate basis for further transatlantic co-operation and harmonisation of anti-trust policies, including in relation to the IP and anti-trust interface.

This chapter was first published in PLC Cross-border Competition Handbook 2005/6 Volume 1 and is reproduced with the permission of the publisher, Practical Law Company. For further information or to obtain copies please contact jennifer.magnan@practical, or visit