common market law review - wydzia‚ prawa i administracji uw

EU Patent 63Common Market Law Review 47: 63–115, 2010.© 2009 Kluwer Law International. Printed in the United Kingdom.

THE EU PATENT: CUI BONO ET QUO VADIT?

THOMAS JAEGER*

1. Introduction

Decades of debate over the creation of a sui generis patent right for the internal market have gone by without result. In recent years, discussions have shifted away from the substantive patent right and focused more on institutional and enforcement issues. In spite of numerous proposals produced by the Commission and various Council presidencies over the years, an initial agreement among Member States in the Council could be reached only very recently, in December 2009. Although that initial agreement signals just the start of the lengthy leg-islative process under the co-decision procedure, the outcome of which is all but certain, and although it does not extend to some important details of the system, prominently the language regime, fees and the rules of procedure for the patent court, the agreement is celebrated as a political breakthrough. Is such euphoria justified? On the basis of the reasons for the prolonged stalemate over the EU patent, this article explores the project’s chances of success fol-lowing the initial Council agreement, highlights necessary changes and tries to display what the patent system’s eventual shape at the end of the legislative process might be.

2. 52 years of discussion: Where do we stand today?

The idea to create a European patent system is as old as the Rome Treaties: it can be traced back to 1957. 1 Numerous proposals for a single patent were launched over the years, with the major texts dating from 1975 (first proposal for a patent convention), 2 1989 (second proposal for a convention), 3 2000 (first

* Dr.iur., LL.M. The author is researcher at the Max Planck Institute for Intellectual Property, Competition and Tax Law in Munich. Gratitude is owed to Mr Christoph Heinrich for his work on the manuscript. The usual disclaimer applies.

1. Cf. already Beier, “Stand und Aussichten der europäischen Rechtsvereinheitlichung auf dem Gebiete des gewerblichen Rechtsschutzes”, (1969) GRUR Int., 145, at 146 et seq.

2. Cf. Convention for the European patent for the common market (First Community Patent Convention), O.J. 1976, L 17/1.

3. Cf. Agreement relating to Community patents (Second Community Patent Convention), O.J. 1989, L 401/1.

64 Jaeger CML Rev. 2010

proposal for a patent regulation), 4 2004 (second proposal for a regulation) 5 and 2007 6 (Commission strategy paper). 7 Unlike the European Communities founded by those Treaties, however, the (EEC, EC and, now the Lisbon Treaty is adopted,) EU patent never came to see the light of day. The points of disagreement that stalled the patent were never really about the patent right as such, i.e. about the core features of its design. Rather, sim-ple protectionism among Member States halted the plans in the early days.8 Intra-European protectionism has meanwhile disappeared in the face of increas-ing economic globalization and yielded to widespread acceptance among the better part of Member States,9 private stakeholders and academics, 10 that the single patent is crucial for securing European competitiveness on the global level.11 Accordingly, a drive for fresh patent proposals was undertaken by the Commission and various Council presidencies after the turn of the millennium. That wave of fresh initiatives has, however, long held up by disagreement over formal issues: language requirements, the administrative set-up and revenue and cost allocation, the relationship to the European patent system and national patent rights, and, importantly, the court enforcement structure. 12 All of these issues will be discussed and assessed below.

4. Cf. Proposal for a Council Regulation on the Community patent (Regulation 2000), O.J. 2000, C 337E/278.

5. Cf. Proposal for a Council Regulation on the Community patent (Regulation 2004), Council Document No. 7119/04; at the time of print under discussion in the form of the latest revised draft of 29 Sept. 2009, Council Document No. 13706/09. Now part of the Council’s political agreement on an EU patent package, cf. Press Release of 4 Dec. 2009, IP/09/1880.

6. Cf. Communication Enhancing the patent system in Europe, COM(2007)165 Final; also Press Release of 3 Apr. 2007, IP/07/463, 2.

7. For a detailed overview of historic developments: see Ullrich, “National, European and Community patent protection: Time for reconsideration” in Ohly and Klippel (Eds.), Geistiges Eigentum und Gemeinfreiheit (Tübingen, 2007), p. 61, at pp. 65 et seq.; Communication 2007, cited supra note 6, points 1 and 3.

8. Cf. Beier, op. cit. supra note 1, 147; Bossung, “The return of European patent law to the European Union”, (1996) IIC, 287, at 288 et seq.

9. Cf. in particular Conclusions of the Presidency of the Brussels European Council, 14 and 15 Dec. 2006, para 29.

10. Cf., for many, the recent summary by Schneider, “Patents in Europe and their court – Is there light at the end of the tunnel?” in Waldeck, Pyrmont, Adelman, Brauneis, Drexl and Nack (Eds.), Patents and Technological Progress in a Globalized World, Liber Amicorum for Joseph Straus (Berlin, 2009), p. 633, at pp. 634 et seq.; also Ullrich, “Patent protection in Europe: Inte-grating Europe into the Community or the Community into Europe?”, (2002) ELJ, 433, at 484.

11. This motive can be traced back to the very roots of the Community patent plans, cf. already Beier, op. cit. supra note 1, 147 et seq.

12. Cf. Communication 2007, cited supra note 6, 3; also Leith, “Revision of the EPC, the Community Patent Regulation and European technical judges”, (2001) EIPR, 250, at 250; Hölder, “The Community Patent – Break-through or setback?”, (2004) EIPR, 43, at 46; Bonita-tibus, “The Community Patent System Proposal and patent infringement proceedings: An eye towards greater harmonization in European intellectual property law”, (2001) Pace L. Rev., 202,

EU Patent 65

It was only in December 2009 that an initial political agreement on the estab-lishment of a patent system for the EU could be reached.13 Yet, that agreement excludes some of the sensitive points just mentioned: common ground was not found regarding translation arrangements. A subsequent, separate act is still required here, without which the EU patent Regulation cannot enter into force. Furthermore, for renewal fees and regarding the cooperation between patent offices, only a common understanding was drafted, and also the rules of pro-cedure for the patent judiciary are still in the drafting stage. In terms of legis-lative procedure, the path from the initial political agreement in the council to the EU patent is still a long one: the standard legislative procedure under Arti-cle 294 TFEU (the former co-decision procedure under Art. 251 EC) will have to be run through and the European Parliament will have the next say there. Also, the ECJ was asked for a legal opinion on the compatibility of the new patent court with EU law. If significant changes are required, the debate may be prolonged further. Finally, as will be discussed below, the patent judiciary will have to be set up by way of a mixed agreement involving the EU and all participating EU and EPO Member States, and will hardly become operative before at least a significant number of those parties have completed their national ratification processes. Outside the framework of EU law, a form of patent cooperation was set up under the European Patent Convention (EPC) in 1973. 14 Administered by the European Patent Office (EPO) in Munich (the Munich system), the EPC uni-fied the substantive rules and procedures for obtaining a patent and for com-plaints. The Munich system is, however, fundamentally different in character from what an EU right would be: under the Munich system, no single European

at 205 et seq.; Luginbühl, “Streitregelungsübereinkommen vs. Gemeinschaftspatent?”, (2004) GRUR Int., 357, at 38 et seq.; Landfermann, “Stand der Diskussionen um ein europäisches Pat-entgerichtssystem – Ausblick”, (2001) VPP-Rundbrief, 113, at 113 et seq.; Landfermann, “Die Entwicklung der europäischen Gerichtsbarkeit in Patentsachen”, (2003) Mitt PatAnw, 341, at 343 et seq.; Schade, “Das Streitregelungssystem zum Gemeinschaftspatent nach dem Ver-ordnungs-Vorschlag der Kommission”, (2000) GRUR, 827, at 830 et seq.; Tilmann, “Patent-schutzsystem in Europa”, (1998) GRUR, 325, at 325 et seq.; Tilmann, “Neue Überlegungen im Patentrecht”, (2006) GRUR, 824, at 824 et seq.; Tilmann, “Gemeinschaftspatent mit einem zen-tralen Gericht”, (2003) GRUR Int., 381, at 381.

13. Cf. Press Release of 4 Dec. 2009, supra note 5.14. Cf. Convention on the Grant of European Patents of 5 Oct. 1973 (EPC), applicable ver-

sion of 2000, available at <www.epo.org/patents/law/legal-texts/epc.html>.Cf. Ullrich, “Die gemeinschaftsrechtliche Entwicklung des Rechts des Geistigen Eigentums” in Behrens (Ed.), Stand und Perspektiven des Schutzes des Geistigen Eigentums in Europa (Baden-Baden, 2004), p. 9, passim; Ullrich, op. cit. supra note 10, 433, at 440 et seq.; Koch and Stauder, Vereinbarung über Gemeinschaftspatente, 2nd ed. (Köln, 1997), pp. 1 et seq.; Bossung, “Rückführung des Europäischen Patentrechts in die Europäische Union”, (1995) GRUR Int., 923, at 924; Liedl, Vorschläge zum Gemeinschaftspatent und zur Streitregelung Europäischer Patente (Hamburg, 2007), p. 310.


patent right is available. The EPO grants the bundle of national patents applied for, but post-grant administration and enforcement of those patents remains nationally fragmented. For this reason also, the Munich and EU systems are not mutually exclusive.15 This remaining national fragmentation of adminis-tration and enforcement of the patent rights in particular is a prominent short-coming of the Munich patent system, and, together with administrative, processing and translation costs up to nine times as high as in the US or Japan,16 renders obtaining patent protection in a significant number of European States relatively unattractive. In 2003, a European Patent Litigation Agreement (EPLA) 17 was drafted to improve at least the enforcement aspect for the Munich system by transferring enforcement to a single European Pat-ent Judiciary (EPJ). The improvement of the functioning and attractiveness of the Munich sys-tem envisaged by the EPLA might have entrenched the division of patent sys-tems in Europe, but, luckily for the EU patent, EPLA never saw the light of day. As is discussed in more detail below,18 the Commission rightly 19 and successfully 20 disputed the competence of EU Member States to participate in the EPLA on their own, in view of pre-existing EU legislation in the fields of jurisdiction and recognition of judgments 21 and enforcement of intellectual property rights 22 requiring the EU’s assent to EPLA.

15. Already the earliest proposal envisaged the EU patent to be integrated into the Munich system insofar as the EPO would grant EU patents as well as national patents if applied for, cf. Art. 1(1) First Community Patent Convention, cited supra note 2.

16. Cf. Communication 2007, cited supra note 6, 2 et seq.17. Cf. Draft Agreement on the establishment of a European patent litigation system of 2003

(EPLA), in the form of the Revised Draft of 2005, available at <www.epo.org/patents/law/ legislative-initiatives/epla.html>

18. See infra section 6.6.1.19. Equally Pagenberg, “Industry, legal profession and patent judges press for adoption of

the European Patent Litigation Agreement (EPLA)”, (2006) IIC, 46, at 49; Liedl, op. cit. supra note 14, p. 311; Telg, Die Neuordnung der europäischen Patentgerichtsbarkeit – Entwicklungen und Perspektiven der Streitregelung auf dem Gebiet des europäischen Patentrechts (Hamburg, 2005), pp. 131 et seq.; Tilmann (2006), op. cit. supra note 12, 825; Laubinger, Die internation-ale Zuständigkeit der Gerichte für Patentstreitsachen in Europa (Hamburg, 2005), p. 136; Hye-Knudsen, Marken-, Patent- und Urheberrechtsverletzungen im europäischen Internation-alen Zivilprozessrecht (Tübingen, 2005), p. 158; Bukow, Verletzungsklagen aus gewerblichen Schutzrechten (Hamburg, 2003), p. 21; similarly Landfermann (2003), op. cit. supra note 12, 348; Wannemacher, Die Außenkompetenz der EG im Bereich des internationalen Zivilverfahren-srechts (Frankfurt am Main, 2003), pp. 8 et seq.

20. Cf. also Recommendation to authorize the Commission to open negotiations for the adoption of an Agreement creating a Unified Patent Litigation System, SEC(2009)330 final, 3.

21. Cf. Council Regulation (EC) No. 44/2001 of 22 Dec. 2000 on jurisdiction and the recog-nition and enforcement of judgments in civil and commercial matters, O.J. 2001, L 12/1.

22. Cf. Directive 2004/48/EC of the European Parliament and of the Council of 29 April 2004 on the enforcement of intellectual property rights (Enforcement Directive), O.J. 2004, L 157/45.

EU Patent 67

Although for this reason, EPLA is off the agenda today, it demonstrated that the Munich system is plagued by an urgent need for reform. At the same time, the Commission and some Council presidencies are inclined to use the EU’s foothold gained in litigation for Munich patents as bargaining power to push for consent to the creation of an EU patent. In stark contrast to a 2003 proposal for the litigation of EU patents,23 which was essentially rooted in the existing EU court system relying on the CFI (and panels) and the ECJ, the post-EPLA litigation proposals of 200724 and 2009 25 advocate unified litigation for Munich patents and future EU patents before a common court. These models, includ-ing the latest basis for the Council’s 2009 political agreement, are discussed below. EPLA has thus led to an important, in fact: unprecedented, change in to the debate over the EU patent, in that two groups of formerly independent factors have now become intertwined by virtue of the Commission’s bargaining power: 26 Firstly, the Munich system will not see reform without a compromise being reached in parallel over the future of the EU patent. The future of the two systems must therefore be discussed together. Secondly, because litiga-tion is the key aspect of reform for the Munich system, the creation of the sub-stantive EU patent right cannot go ahead without parallel agreement on a, separate or common, litigation model. Litigation has thus become a key aspect for the EU patent also. In sum, the discussions over the future of the patent system or systems in Europe have, by virtue of EPLA, become significantly more complicated in terms of territorial (EU and EPC territories) and material (substantive right and litigation) scope. No single one of those aspects can be answered without at least a prejudgment as to the answers to all other aspects and simple solutions are no longer viable. At the same time, this expansion of the import of discussions in territorial and material terms also means that the future shape of patents and patent enforcement in Europe can be addressed through comprehensive solutions. In turn, today’s deal is all or nothing: either we take an EU patent plus a reconciliation of the EPC and EU systems plus a unified patent court or we don’t get anything at all.

23. Cf. Proposal for a Council Decision conferring jurisdiction on the Court of Justice in dis-putes relating to the Community patent, COM(2003)827 final; Proposal for a Council Decision establishing the Community Patent Court and concerning appeals before the Court of First Instance, COM(2003)828 final.

24. Cf. Communication 2007, cited supra note 6, point 2.25. Cf. Draft Agreement on the European and Community Patents Court (Agreement 2009)

and Draft Statute (Draft Statute ECPC), both Council Document No. 5072/09; also Recommen-dation 2009, cited supra note 20. Now part of the Council’s political agreement, cf. Press Release supra note 5.

26. Cf. also Jaeger, Hilty, Drexl and Ullrich, “Comments of the Max Planck Institute for Intellectual Property, Competition and Tax Law on the 2009 Commission Proposal for the Estab-lishment of a Unified European Patent Judiciary”, (2009) IIC (forthcoming), para 3.


This article explains the developments in relation to the EU patent and pat-ent enforcement so far and sets out to explore the way ahead. It will examine what a comprehensive patent system may look like and how we might proceed to implement it. The starting point here is re-visiting the question of whether an EU patent is needed at all or whether we should instead focus on a reform or expansion of the Munich system. If an EU patent entails certain benefits vis-à-vis Munich patents, what key features must be ensured for the EU pat-ent to realize those benefits? On that basis, against the background of the mod-els discussed so far and their flaws, the enforcement structure for the EU patent and their compatibility with the EU legal order will be examined. Finally, the very next steps ahead for the EU patent are sketched out in broad strokes.

3. EU patent: Cui bono?

We do not need an EU patent unless it yields clear advantages as compared to patents handed out under the Munich system. Based on the EU patent propos-als launched so far, this section compares the characteristics of the EU patent and of Munich patents and assesses what potential benefits an EU patent may hold for the internal market.

3.1. Uniformity of rights and claims

The most obvious attribute of the EU patent as compared to the EPC system is the provision of a unitary right with unitary effects and limitations throughout the EU territory,27 similar28 to the existing uniform rights for EU t rademarks, designs and (albeit with a lesser degree of sophistication) plant varieties.29 The EPO currently hands out territorially limited, national patent rights only. In addition, the conditions and limitations for patent protection and patent claims are largely harmonized via the EPC, but especially in relation to claims (i.e. the scope of the right upon enforcement), there is no uniformity.30 Ancil-lary questions, such as the conditions for a civil law transfer of the right and

27. Cf. already Art. 2(2) First Community Patent Convention, cited supra note 2; Art. 2(2) Second Community Patent Convention, cited supra note 3.

28. Cf. also Ullrich, op. cit. supra note 10, 472.29. Cf. Arts. 4 et seq., 16 and 95 et seq. Council Regulation (EC) No. 207/2009 of 26 Feb.

2009 on the Community trade mark (CTMR), O.J. 2009, L 78/1; Arts. 4 et seq., 27 and 80 et seq. Council Regulation (EC) No. 6/2002 of 12 Dec. 2001 on Community designs (CDR), O.J. 2002, L 3/1; Arts. 5 et seq., 22 and 94 et seq. Council Regulation (EC) No. 2100/94 of 27 July 1994 on Community plant variety rights (PVR), O.J. 1994, L 227/1.

30. Cf. also Ullrich, op. cit. supra note 10, 455 et seq.

EU Patent 69

certain characteristics of the right in rem, the bearing of risk associated with the right or its subjection to insolvency and execution, are largely left to the civil law applicable in each State of protection under the EPC.31

This may mean that in multi-State cases, national courts will have to apply cumulatively the laws of several territories of protection. For licensing con-tracts and other contracts involving patents, the Rome I Regulation has recently brought some uniformity in terms of the applicable law even in the absence of a choice of law by the parties in the contract, by stipulating applicability of the law of the country where the party required to effect the characteristic per-formance of the contract has his habitual residence.32 For licensing contracts, this will typically (although not always)33 mean that the law at the place of establishment of the licensor is applicable.34 In the ancillary contractual mat-ters not covered by the Rome I Regulation, however,35 the national courts still need to apply the leges contractus of all relevant jurisdictions. Likewise in infringement cases as well as for claims relating to culpa in contrahendo, unjust enrichment and agency of necessity, the Rome II Regulation stipulates application of the lex loci protectionis principle for national patent rights.36 A court seized with a multi-State infringement action is thus bound to apply the provisions of as many Member States laws as there are national patents involved. This may complicate procedures so considerably that the accumu-lated application of foreign substantive law by courts from other systems has been criticized as a de facto denial of justice.37 By contrast, for infringement of sui generis EU IP rights, the Rome II Regulation requires application only of the law of the one State where the infringement was committed (lex loci delicti).38 The EU patent would therefore create considerably less complexity

31. Cf. in particular Art. 74 EPC, cited supra note 14.32. Cf. Art. 4 (2) Regulation (EC) No. 593/2008 of the European Parliament and of the

Council of 17 June 2008 on the law applicable to contractual obligations (Rome I), O.J. 2008, L 177/6.

33. Cf. Opinion of the European Economic and Social Committee on the Proposal for a Regulation of the European Parliament and of the Council on the law applicable to contractual obligations, O.J. 2006, C 318/56.

34. Cf. (in relation to copyright contracts) Schack, Urheber- und Urheberverlagsrecht, 4th ed. (Frankfurt, 2007), paras. 1142 et seq.

35. Cf. in particular Art. 1(2) Rome I, cited supra note 32.36. Cf. Art. 8(1) and 13 Regulation (EC) No. 864/2007 of the European Parliament and of the

Council of 11 July 2007 on the law applicable to non-contractual obligations (Rome II), O.J. 2007, L 199/40.

37. Cf. Heiss and Loacker, “Die Vergemeinschaftung des Kollisionsrechts der außervertrag-lichen Schuldverhältnisse durch Rom II”, (2007) JBl, 613, at 635, and the references cited there.; equally for other IP rights than patents Schack, “Internationale Urheber-, Marken- und Wett-bewerbsrechtsverletzungen im Internet”, (2000) MMR, 59, at 64; Tilmann (2003), op. cit. supra note 12, 386 et seq.

38. Cf. Art. 8(2) Rome II, cited supra note 36.


for infringement litigation, regardless of whether national courts or a special-ized judiciary is made be competent to hear such cases.39

The Commission considers this fragmentation of patent rights under the cur-rent Munich system – as national rights under the EPC – as being a major obstacle for the effectiveness of patent protection in Europe today.40 Fragmen-tation of rights increases uncertainty as to the scope and limitations of the pat-ent right in the various jurisdictions and accordingly limits its economic value, but it also increases procedural and management costs with the collateral neg-ative effect that smaller Member States are disadvantaged in the transfer of technologies in an effort to keep registration and management costs low. The EU patent would remedy this lack of uniformity not just in relation to patent claims,41 but also for ancillary questions.42 Furthermore, it is stipulated that a subsidiary application of national law to questions left open is limited to one Member State (typically the court at the domicile of the registrant at the time of registration).43 The EU patent will therefore facilitate litigation con-cerning licensing contracts in particular by establishing a highly complete sys-tem of protection.44 Accordingly, the uniform character of the EU patent right is generally seen as a major improvement over the status quo of protection under the EPC.45

39. In particular, the ECJ’s jurisprudence on an expansive interpretation of the Brussels Regulation’s similar forum delicti rule (cf. e.g. Case C-68/93, Shevill, [1995] ECR I-415, para 20), is not yet relevant for the interpretation of the Rome II Regulation.

40. Cf. Green Paper on the Community patent and the patent system in Europe (Green Paper 1997), COM(1997)314 final, 7 et seq.; similarly Ullrich, op. cit. supra note 10, 436 et seq., Boni-tatibus, op. cit. supra note 12, 226 et seq. and 230 et seq.; Krieger, “Das Luxemburger Überein-kommen über das Gemeinschaftspatent Herausforderung und Verpflichtung”, (1987) GRUR Int., 729, at 733 et seq.; Krieger, “Wann kommt das europäische Gemeinschaftspatent? – Zwei Brüder als Kämpfer für den Schutz des geistigen Eigentums in Deutschland, in Europa und in der Welt”, (1998) GRUR, 256, at 260; Hölder, op. cit. supra note 12, 45.

41. Cf. e.g. the definition of infringements and exceptions in Arts. 7, 8 and 19 Regulation 2000, cited supra note 4, and in Arts. 7, 8 and 19 Regulation 2004, cited supra note 5; and the corresponding the delineation of claims in Arts. 31 et seq., 42 et seq. and 46 Regulation 2000, cited supra note 4, and in Arts. 31 et seq., 42 et seq. and 46 Regulation 2004, cited supra note 5.

42. Cf. Arts. 40 et seq. First Community Patent Convention, cited supra note 2; Arts. 39 et seq. Second Community Patent Convention, cited supra note 3; Arts. 14 et seq. Regulation 2000, cited supra note 4; Arts. 14 et seq. Regulation 2004, cited supra note 5.

43. Cf. Art. 38 Second Community Patent Convention, cited supra note 3; Art. 14(1) Regu-lation 2000, cited supra note 4; Art. 14(1) Regulation 2004, cited supra note 5; still slightly dif-ferent Art. 38 First Community Patent Convention, cited supra note 2.

44. Doubtful, however, Tilmann (2003), op. cit. supra note 12, 387 et seq.45. See also the references cited supra note 40.

EU Patent 71

3.2. Uniform compulsory licences

Hand in hand with the uniform character of the EU patent right comes the pos-sibility to obtain EU-wide compulsory licences and a harmonization of the respective conditions. This is, again, an important advancement as compared to the Munich system.46 At present, under Article 31 TRIPS and the punctual rule for biological material in the EPC Implementing Regulation,47 the condi-tions for compulsory licences for Munich patents are divergent in the EPC States.48

The EU patent49 would provide for compulsory licensing rules essentially modelled on Article 31 TRIPS,50 but going significantly beyond that provision in terms of legal quality (uniform standard, not an option) and substance. In particular, compulsory licences for an EU patent may be granted in court51 for lack or insufficiency of exploitation of the patent within three years, in times of crisis, or in other situations of extreme urgency, but also and very impor-tantly, similarly to EU jurisprudence on IP essential facilities,52 i f the patent right hinders an important technical advance of considerable economic significance because another patent or plant variety right cannot be used with-out the first patent.53 These compulsory licensing provisions for the EU pat-ent are also significantly more detailed and broader than what is currently laid down for the EU plant variety right.54

Uniform conditions for compulsory licences will increase legal certainty and are favourable to innovation and competition, particularly in neighbour-ing products and markets. In other words, the leveraging power derived from restrictive licensing policies of dominant proprietors of EU patents will be lim-ited compared to the existing regime. What is more, this effect is significantly reinforced by the flanking establishment of a uniform court, as discussed imme-

46. More critical, however, Ullrich, op. cit. supra note 10, 480 et seq.47. Cf. Rule 33 Implementing Regulations to the Convention on the Grant of European Pat-

ents of 5 Oct. 1973, applicable version of 7 Dec. 2006, available at the link cited supra note 14.48. Cf. Ullrich, op. cit. supra note 14, p. 22.49. Cf. Arts. 46 et seq. First Community Patent Convention, cited supra note 2; Arts. 45 et

seq. Second Community Patent Convention, cited supra note 3; Arts. 21 et seq. Regulation 2000, cited supra note 4; Arts. 21 et seq. Regulation 2004, cited supra note 5.

50. Cf. Ullrich, op. cit. supra note 10, 472 et seq.51. An earlier version unwisely accorded that power to the Commission, cf. Art. 21(1) Reg-

ulation 2000, cited supra note 4, apparently inspired by the current rules for EU plant variety rights.

52. Cf. particularly Joined Cases C-241 & 242/91 P, RTE and ITP (“Magill”), [1995] ECR I-743, paras. 48 et seq.; Case C-418/01, IMS Health, [2004] ECR I-5039, para 52; Case T-201/04, Microsoft, [2007] II-3601, paras. 331 et seq.

53. Cf. Art. 21 Regulation 2004, cited supra note 5.54. Cf. Art. 29 PVR, cited supra note 29.


diately below, which renders the presently nationally fragmented procedure for obtaining compulsory licences for several Member States quicker, easier and less costly. These features of the EU patent are favourable from the points of view of society, national economics and competing SMEs and start-ups, but they may also imply that the EU patent becomes less attractive for larger firms. Such firms typically boast a financial and infrastructural background that easily per-mits them to obtain and manage patent registrations in national regimes. These players may try to lower the risk of losing a key patent – and thus a dominant position – to a competitor by continuing to opt for national patents, not EU patents. By not registering an EU patent, the bar for compulsory licences for key patents can be kept at the present high level.55

This may be one good reason why many authors reject such an optionality principle, i.e. the (currently allowed)56 coexistence of EU and national pat-ents.57 N onetheless, such practices would probably not diminish the EU pat-ent’s success among the large numbers of financially less robust firms, to whom costs and management burdens do matter, and of firms with strategically less crucial patents.

3.3. Follow-on effects

A unitary EU patent right can also be expected to entail some positive follow-on effects.58 This concerns, for example, the long-standing59 plans for EU legislation to align national patent laws, which may have a better chance of success against the background of agreement on an EU patent. Even in the absence of harmonizing legislation, however, an indirect harmonizing influ-ence on the substance of national patent laws may flow from the EU patent.60

55. Cf. also Arts. 3 and 15(1)(f) and (2) Agreement 2009, cited supra note 25.56. Cf. Rec. 4 Regulation 2004, cited supra note 5.57. Cf. Jaeger, Hilty, Drexl and Ullrich, op. cit. supra note 26, para 9; Willems, “The EPC:

The Emperor’s Phantom Clothes? A blueprint instead of a Green Paper”, (1998) IPQ, 1, at 5 et seq.; Beier, op. cit. supra note 1, 150; Bossung, “Nachtrag zu ‘Unionspatent statt Gemeinschaft-spatent’”, (2002) GRUR Int., 575, at 576; Bossung, op. cit. supra note 14, 934 et seq.; similarly Krieger (1998), op. cit. supra note 40, 259.

58. Similarly Krieger (1987), op. cit. supra note 40, 732 and 734.59. Cf. Resolution on Art. 6 First Community Patent Convention on the Adaptation of

national patent law, reprinted in Haertel, Gemeinschaftspatentübereinkommen (Köln, 1977), p. 82; Resolution to the Second Community Patent Convention on the Adaptation of national patent law, reprinted in Koch and Stauder, op. cit. supra note 14, p. 99.

60. Cf. already Krieger (1987), op. cit. supra note 40, 732 and 734.

EU Patent 73

Another hope61 a ssociated with the establishment of a single unitary EU pat-ent right is an ancillary effect towards the further removal of obstacles to the free movement of goods and capital as well as to the freedom of establishment, as could already be observed in relation to the creation of the EU trademark and design rights.62 For patents, improvements in the enjoyment of the funda-mental market freedoms may result from the uniform rules on exhaustion,63 on licensing contracts64 or on compulsory licensing,65 to name just some exam-ples. These rules promise to be more effective and more transparent than the ECJ’s jurisprudence on Article 30 EC (now Art. 36 TFEU) in particular,66 under which restrictions justified by the specific character of patent protection may persist.67

3.4. Result

An EU patent right shows several key improvements over the status quo of fragmented patent rights under the Munich system. The foremost intrinsic effect associated with the creation of a unitary right is its fully unitary character in terms of conditions and scope. This can be expected to increase legal cer-tainty and facilitate the exploitation of the patent right for rightholders. Uniformity can, however, also be considered advantageous for potential third-party users of EU patents, where uniformity in the conditions for compulsory licences facilitates access to technologies and lowers access costs. Finally, the intrinsic effects will likely be accompanied by some follow-on effects in terms of a harmonization of national patent laws, either through EU legislation or via indirect influence, and a removal of remaining restrictions in the exercise of the fundamental market freedoms.

61. Cf. Bonitatibus, op. cit. supra note 12, 231 et seq.; Di Cataldo, “From the European Pat-ent to a Community Patent”, 8 Col. J. Eur. L. (2002), 19, at 30 et seq.; Krieger (1987), op. cit. supra note 40, 732; Krieger, (1998) GRUR, op. cit. supra note 40, 260; less optimistic is Ullrich, op. cit. supra note 10, 48 et seq.; Ullrich, “Gewerblicher Rechtsschutz und Urheberrecht im Gemeinsamen Markt” in Immenga and Mestmäcker (Eds.), Kommentar zum europäischen Kartellrecht, 4th ed. (München, 2007), para 96.

62. Cf. Di Cataldo, op. cit. supra note 61, 30 et seq.63. Cf. Art. 10 Regulation 2004, cited supra note 5.64. Ibid., Art. 19.65. Ibid., Art. 21. 66. Cf. Case 15/74, Centrafarm v. Sterling Drug, [1974] ECR 1147, para 9; Case 187/80,

Merck v. Stephar, [1981] ECR 2063, paras. 4 et seq.; Case 19/84, Pharmon, [1985] ECR 2281, paras. 22 et seq.; Joined Cases C-267 & 268/95, Merck v. Primecrown, [1996] ECR I-6285, paras. 30 et seq.; Case C-316/95, Generics, [1997] ECR I-3929, paras. 19 et seq.

67. For a different view: Ullrich, op. cit. supra note 61, para 96.


4. Reinforcement of positive effects through unified litigation

The benefits of the unitary character of the EU patent right are reinforced by the establishment of a uniform litigation system. Some form of court enforce-ment has always formed part of the proposals for an EEC or EC patent, but not all of them foresaw litigation before a single court. To date, none of the existing EU intellectual property rights can rely on a common court for infringement claims: infringement litigation is decentralized and remains the responsibility of the Member States’ courts. Although this would therefore be a novelty in the EU law system of courts and court actions, all proposals for an EU patent except the very first (which was similar to today’s EU trademarks and designs litigation)68 foresaw some form of centralization of infringement litigation – albeit, as will be shown below, widely divergent. In addition, the latest design proposals, on which the Council’s 2009 political agreement is also based, have coupled the centralization of litigation with an integration of the EU and Munich patent systems.69

4.1. Economic effects of existing litigation insufficiencies

The benefits of unitary litigation have long been propagated by the Commission70 and are avidly debated and mostly supported in literature.71 T he single most convincing argument for the establishment of a unitary patent court system, however, comes from a current economic study undertaken for the Commission. That study analyses for the first time the actual costs of the present fragmentation

68. Cf. Arts. 63 and 69 First Community Patent Convention, cited supra note 2.69. Cf. Communication 2007, cited supra note 6, 10 et seq.; Art. 1 Recommendation 2009,

cited supra note 20.70. Cf., e.g, Green Paper 1997, cited supra note 40, 7 et seq.71. Cf. e.g. Leonard, “Letters”, (2001) EIPR, 264, at 264; Krieger (1987), op. cit. supra note

40, 733 et seq.; Krieger (1998), op. cit. supra note 40, 260; Bonitatibus, op. cit. supra note 12, 228 et seq.; Cannon, “Achieving the benefits of a centralized Community patent system at min-imal cost”, 35 Case Western Reserve J. of Int. L. (2004), 415, at 434; Fabry, “Die Harmonis-ierung der europäischen Patentrechtsprechung – Notwendiges Übel oder üble Notwendigkeit?”, (2008) GRUR, 7, at 7 et seq.; Ullrich, op. cit. supra note 10, 436 et seq.; Di Cataldo, op. cit. supra note 61, 21; a differentiated view is by Jaeger, “Ein Königreich für ein Schwert? – Kohärenz und Inkohärenz im Rechtsrahmen der Verfolgung von Immaterialgüterrechten in der Europäischen Union” in Hilty, Jaeger and Kitz (Eds.), Geistiges Eigentum – Herausforderung Durchsetzung (Berlin, 2008), p. 155, at p. 219; again differentiated is Jenkins, “Countdown to the Community Patent?”, (2004) Patent World, 26, at 27; Luginbühl, op. cit. supra note 12, 385 et seq.; Schäfers, “Anmerkungen zu einem gemeinschaftsrechtlichen Gemeinschaftspatent”, (1999) GRUR, 820, at 824; critical Tilmann (2003), op. cit. supra note 12, 385; opposed Pagen-berg, “Another year of debates on patent jurisdiction in Europe and no end in sight?”, (2007) IIC, 805, at 820 et seq.; Jaeger et al., op. cit. supra note 26, paras. 12 et seq.

EU Patent 75

in litigation72 t o private parties and national economies and suggests that a unitary court system doing away with these deficiencies is likely to entail sav-ings for private parties between 148 and 289 million Euros.73

The study identifies essentially four groups of major deficiencies of the present litigation system:74 1) Duplication of litigation in multiple jurisdic-tions, 2) divergent outcomes without a possibility of consolidation, 3) a lack of incentives for parties to consolidate divergences through settlements espe-cially in high-profit sectors (like pharmaceuticals) and because of the possi-bility for strategic litigation and forum shopping and 4) insufficiencies in the procedural framework for patent litigation under the Brussels I Regulation,75 e specially regarding the unavailability of cross-border injunctions76 and its openness to delay strategies and torpedo motions.77

The study suggests that the creation of a unified court would lead to improve-ments along all these points, with the degree of improvement depending on the cost implications of the details of design.78 Assessed in the study was the design according to the latest 2009 proposal.79 The unitary structure would

72. For a legal analysis of the entailing deficits, cf. particularly Ullrich, op. cit. supra note 10, 436 et seq.; Jaeger in Hilty, Jaeger and Kitz, op. cit. supra note 71, p. 157; Artelsmair, “A com-prehensive patent system needed for Europe” in Kur, Luginbühl, Waage (Eds.), “... und sie bewegt sich doch!” Patent Law on the Move, Liber Amicorum for Gert Kolle and Dieter Stauder (Köln, 2005), p. 5, at pp. 5 et seq.; Kur, “A Farewell to Cross-Border Injunctions? The ECJ Deci-sions GAT v. LuK and Roche Nederland v. Primus and Goldenberg”, (2006) IIC, 844; Liedl, op. cit. supra note 14, pp. 76 et seq.; Luginbühl and Stauder, “Der Europäische Gerichtshof setzt den grenzüberschreitenden Entscheidungen in Patentsachen ein vorläufiges Ende”, (2006) sic!, 876; Kur and Metzger et al., “Exclusive jurisdiction and cross border IP (patent) infringement suggestions for amendment of the Brussels I regulation”, (2007) IER, 1; Hess-Blumer, “Cross-border Litigation – und sie lebt doch!”, (2006) sic!, 882; Rößler, “The Court of Jurisdiction for Joint Parties in International Patent Disputes”, (2007) IIC, 380; Schauwecker, “Europäische Union/EFTA – Revidiertes Lugano-Übereinkommen kodifiziert umstrittene EuGH-Recht-sprechung zur internationalen Zuständigkeit bei grenzüberschreitenden Patentverletzungskla-gen”, (2007) GRUR Int., 552; Knaak, “Internationale Zuständigkeiten und Möglichkeiten des forum shopping in Gemeinschaftsmarkensachen – Auswirkungen der EuGH-Urteile Roche Nie-derlande und GAT/LUK auf das Gemeinschaftsmarkenrecht”, (2007) GRUR Int., 386; Lange, “Der internationale Gerichtsstand der Streitgenossenschaft im Kennzeichenrecht im Lichte der “Roche/Primus”-Entscheidung des EuGH”, (2007) GRUR, 107; Hellestadius and Meier-Ewert, “Jurisdiction and Choice of Law in Intellectual Property Matters – Perspectives for the Future (EU and World-Wide)”, (2005) IIC, 313, at 314 et seq.; Tilmann and von Falck, “EU-Patentrech-tsharmonisierung II: Forum-Shopping und Torpedo”, (2000) GRUR, 579.

73. Cf. Harhoff, “Economic cost-benefit analysis of a unified and integrated European Patent Litigation System”, 26 Feb. 2009, available at <ec.europa.eu/internal_market/indprop/patent/index_en.htm>, 5.

74. Ibid., 14 et seq.; also Communication 2007, cited supra note 6, 6.75. Cf. supra note 21.76. Cf., for many, Kur (2006), op. cit. supra note 72.77. Cf., for many, Leitzen, “Comeback des ‘Torpedo’?”, (2004) GRUR Int., 1010.78. Cf. Harhoff, op. cit. supra note 73, 18 and 21 et seq. 79. Cf. supra note 25.


obviously end the duplication of proceedings and the various problems flow-ing thereof, i.e. divergent outcomes,80 non-consolidation and the need for cross-border injunctions.81 Importantly also, the unitary court structure also puts an end to the current principle of separation of revocation (or invalidation) and infringement proceedings under the Brussels I Regulation,82 thus forestalling delay strategies and torpedo motions and reducing the attractiveness of forum shopping. These features, plus the low litigation costs and other features (such as an arbitration mechanism, a decentralized structure etc.) will particularly profi t SMEs and start-ups.83

4.2. Enhanced access to compulsory licences

The end to duplicated litigation is particularly relevant for actions to obtain compulsory licences. The unitary court would have the power to grant com-pulsory licences for EU patents effective for the entire internal market. In the present fragmented system, these claims must be litigated separately in every jurisdiction: as a measure of public authority and public interest, applications for compulsory licences are not covered by the jurisdictional rules of the Brussels I Regulation,84 to the effect that there is no international jurisdiction for compulsory licences to foreign patents. The same is true for claims based on Article 102 TFEU (ex Art. 82 EC) to obtain a compulsory licence to access an essential IP facility.85 These claims presuppose that the restriction of access is to be deemed an abuse of a dominant position, which will only exceptionally86

80. For the benefits of a consolidated interpretation of the law, cf. also Ullrich, op. cit. supra note 10, 480 et seq.; Di Cataldo, op. cit. supra note 61, 21; Krieger (1998), op. cit. supra note 40, 260; Bonitatibus, op. cit. supra note 12, 228 et seq.

81. Cf. Harhoff, op. cit. supra note 73, 22 and 42 et seq. 82. Cf. Art. 22(4) Brussels I Regulation, cited supra note 21.83. Cf. Harhoff, op. cit. supra note 73, 50; similarly (benefits of low cost and/or faster pro-

cedures) Bonitatibus, op. cit. supra note 12, 231; Ullrich, op. cit. supra note 10, 440 et seq. and 469; Hölder, op. cit. supra note 12, 43; Cannon, op. cit. supra note 71, 434; Krieger (1998), op. cit. supra note 40, 260; Krieger (1987), op. cit. supra note 40, 733 et seq.; Leonard, op. cit. supra note 71, 264; doubtful Tilmann (2003), op. cit. supra note 12, 385.

84. Cf. Kropholler, Europäisches Zivilprozeßrecht, 8th ed. (Frankfurt am Main, 2005), Art. 22 EuGVVO, para 49; Rogge and Grabinski in Benkard (Ed.), Patentgesetz, 10th ed. (Mün-chen, 2006), § 139 PatG, para 101d; Geimer and Schütze, Europäisches Zivilverfahrensrecht, 2nd ed. (München, 2004), Art. 22 EuGVVO, para 235; Hye-Knudsen, op. cit. supra note 19, p. 25; Stauder, “Die ausschließliche internationale Zuständigkeit in Patentstreitsachen nach dem Brüsseler Übereinkommen – Art. 16 Nr.4 EuGVÜ, Art. 19 EuGVÜ”, (1985) IPrax, 76, at 78.

85. For details cf., for many, Höppner, “Missbräuchliche Verhinderung ‘neuer’ Produkte durch Immaterialgüterrechte – Zur Anwendung von Art. 82 EG auf Lizenzverweigerungen”, (2005) GRUR Int., 457.

86. Cf. Case C-418/01, IMS Health, [2004] ECR I-5039, para 35, and the case law cited there and supra in note 52.

EU Patent 77

be the case and is difficult to substantiate.87 What is more, however, the extra-territorial jurisdiction for infringements of competition law accorded to national competition authorities and courts by Regulation 1/200388 hinges upon the rules of the Brussels I Regulation.89 Since Regulation 1/2003 contains no inde-pendent basis for international jurisdiction,90 claims for compulsory licences must also be brought individually in each State of protection when based on Article 102 TFEU (ex 82 EC). This severe gap of the present EU legal order and the EPC system would be remedied by the unitary patent jurisdiction, as regards EU patents already by way of the explicit compulsory licensing provi-sion envisaged, as regards European patents as the unitary court will close a gap in the Brussels I Regulation.

4.3. Changes in litigation patterns

Beyond those immediate benefits of effectiveness of the legal process, the aforementioned economic study also estimates that the unified court will change the overall demand structure for litigation. It is assumed that the envis-aged low litigation costs (at around the current German, French or Dutch levels) will initially induce parties to choose litigation over settlement. Once a body of reliable jurisprudence has developed, the inverse effect will show, i.e. par-ties will choose to settle rather than to go to court because the outcome of liti-gation is predictable.91 This means that jurisprudence will also allow proprietors to better assess the strength of their rights in the internal market as a whole

87. Cf. Drexl, Hilty and Kur, “Designschutz für Ersatzteile – Der Kommissionsvorschlag zur Einführung einer Reparaturklausel”, (2005) GRUR Int., 449, at 453 et seq.; Conde Gallego, “Die Anwendung des kartellrechtlichen Missbrauchsverbots auf ‘unerlässliche’ Immaterialgüter-rechte im Lichte der IMS Health- und Standard-Spundfass-Urteile”, (2006) GRUR Int., 16, at 28; Heinemann, “Gefährdung von Rechten des geistigen Eigentums durch Kartellrecht? – Der Fall ‘Microsoft’ und die Rechtsprechung des EuGH”, (2006) GRUR, 705, at 713.

88. Cf. Art. 5 and (but debatable) 6 Council Regulation (EC) No. 1/2003 of 16 Dec. 2002 on the implementation of the rules on competition laid down in Arts. 81 and 82 of the Treaty, O.J. 2003, L 1/1.

89. In addition, under Art. 5 Regulation 1/2003 (powers of national competition authorities), the Brussels I Regulation is only applicable where the national competition authority is a court, e.g. in Austria, cf. Thyri, Kartellrechtsvollzug in Österreich (Vienna, 2007), para 64. For public authorities that are not courts, there is generally no international jurisdiction under Regulation 1/2003. For an example of application of the Brussels I Regulation under Art. 5 Regulation 1/2003, cf. OGH (Austrian Supreme Court), 16 July 2008, 16Ok3/08, Bayerische Staatsforsten, point 3.

90. Cf. also Bauer in Hirsch, Montag and Säcker (Eds.), Münchener Kommentar zum Euro-päischen und Deutschen Wettbewerbsrecht – Band 2, GWB (Munich, 2008), Art. 5 VO 1/2003, para 32, and Art. 6 VO 1/2003, para 8; Dalheimer in Grabitz and Hilf (Eds.), Das Recht der Europäischen Union, 37th ed. (München, 2008), Art. 5 VO 1/2003, paras. 24 et seq.

91. Cf. Harhoff, op. cit. supra note 73, 22 and 43 et seq.


and thereby increase the value of those rights and facilitate their exploitation.92 Associated with this effect are a predicted increasing use of patents in Member States which currently have low patent numbers and an increasing use of liti-gation in Member States with cost levels above those of the unified court.93 Also, an important welfare effect is expected to derive from a lowering of cost and procedural hurdles for revocation proceedings (including, in particular, no requirement to exhaust opposition at the EPO first)94 under the assessed pro-posal.95 More and faster96 revocation activity means that fewer erroneously granted patents (so-called trivial patents) impede competition. Also, cheaper revocation suggests that it will become more difficult for bad faith proprietors to extort licence fees from other parties based on weak or questionable patent rights (so-called trolling).97

4.4. Result

To summarize these points, it may be said that the creation of a single special-ized patent judiciary would remedy the major existing jurisdictional inefficien-cies of the EPC system and remove the need for the establishment of a separate court for that system, as proposed by the EPLA. Importantly also, litigation for compulsory licences, where there is currently no international jurisdiction as such claims are not covered by the jurisdictional rules of the Brussels I Regulation,98 would become significantly easier.99

These expectations for systemic improvement apply subject to the general reservation already stated in earlier studies, that the actual effects of unitary jurisprudence on the patent system and patent quality eventually depend on the case law developed by that court: Especially if the unitary court takes a lax stance regarding the scrutiny of patent quality and the inventive step, such jurisprudence may be more detrimental from a welfare point of view than the upkeep of a fragmented system with at least a few very strict courts.100

92. Ibid., 45 et seq.; similarly Krieger (1987), op. cit. supra note 40, 733 et seq.; Fabry, op. cit. supra note 71, 7 et seq.; Bonitatibus, op. cit. supra note 12, 228 et seq.; Cannon, op. cit. supra note 71, 434; Leonard, op. cit. supra note 71, 264.

93. Cf. Harhoff, op. cit. supra note 73, 22 and 43 et seq.94. Cf. Art. 15a(7) Agreement 2009, cited supra note 25.95. Ibid., 44. 96. Doubtful Tilmann (2003), op. cit. supra note 12, 385.97. Cf. Harhoff, op. cit. supra note 73, 44 et seq.98. Council Regulation, cited supra note 21.99. Cf. in particular the negative economic effects of duplicated litgation as analysed by

Harhoff, op. cit. supra note 73, 40 et seq.100. Cf. Wissenschaftlicher Beirat beim Bundesministerium für Wirtschaft und Technologie,

“Patentschutz und Innovation – Gutachten Nr. 1/07 (2007)”, available at <www.bmwi.de/

EU Patent 79

5. The status quo of disc ussions

5.1. Shift of the debate towards enforcement

The above assessment of the substantive and formal features of the EU patent clearly indicates that it, and flanking enforcement structures, would be benefi-cial and desirable for the internal market. Still, in spite of the Council’s 2009 political agreement on an EU patent package, the patent and patent court still have a long way to go. In procedural terms, as was already highlighted, a lengthy legislative process is still ahead and the EU law compatibility also has to undergo preliminary checking. In terms of substance, translation require-ments and the details of revenue and cost allocation in particular still divide EU Member States. In spite of these persisting points of disagreement, a deal on the EU patent may never have been quite as close as today. The debate over EPLA has cre-ated awareness among stakeholders about the urgency of the need for reform and reinvigorated the debate and the willingness to accept compromise. EPLA has also shifted the focus away from the substantive provisions of the EU pat-ent to the design of the enforcement system and to the question of an integra-tion of the EU and Munich patent systems. The substantive features of the EU patent are largely undisputed and the proposal for the substantive rules of the EU patent has not seen major changes since 2004. The initial political agreement on the EU patent reached by the Council in 2009 is accordingly essentially based on the 2004 proposal and its advancements from the consul-tations and discussions since then. Accordingly, the enforcement model forms the central focus of the challenges ahead. As will be shown in this section, points of division here are, among many others, particularly the degree of cen-tralization, representation before the court and the qualification of the bench. This section undertakes an assessment of the enforcement model now agreed by the Council in 2009 and examines possible alternatives to that model. As a basis for this assessment, however, a brief look at what models have already been discussed and rejected in the past is warranted.

5.2. Unde venit: Older models

The initial proposal for litigation of EU patents of 1975 was quite simple and in fact resembles the existing litigation model for EU trademarks and designs.

BMWi/Navigation/Service/publikationen.html> 21; similarly also IBM Business Consulting Services, “The role of national patent Offices, the European patent office as well as Japanese and US patent offices in promoting the patent system”, 14. Feb. 2003, available at <ec.europa.eu/internal_market/indprop/patent/index_en.htm>, 5 et seq.


In a nutshell, that model foresaw separate litigation for revocation and infringe-ment proceedings. Revocation was to be dealt with within the Office by boards of appeal, whose decisions in turn were subject to appeal with the ECJ.101 Infringement lay in the competence of the national courts, with (unlike for trademarks and designs today) counter-actions for invalidity inadmis-sible before those courts.102 Homogeneity in jurisprudence among national courts was to be ensured via an involvement of the EPO in national infringe-ment proceedings as amicus curiae103 and through preliminary references to the ECJ.104 All litigation under this proposal was separate from litigation for European patents. This feature remained unchanged up until 2007, when the plan to integrate the two systems was born in response to the EPLA.105

The idea of a specialized court for patents was introduced by the subsequent proposal from 1989. The idea of that Community Patent Appeals Court (COPAC) was essentially to exercise some supervision over the jurisprudence of national courts for infringement proceedings106 and to adjudicate appeals a gainst EPO decisions.107 In turn, national courts were given competence to deal with questions of invalidity for the whole Community territory (as is the case with trademarks and designs today). The COPAC’s involvement in national procedures was via an interim procedure for questions of invalidity raised before a national court of second instance and via preliminary refer-ences for other questions. The COPAC was in turn bound to refer important questions of Community law under any of these procedures to the ECJ for a preliminary ruling.108

The late 1990s brought a shift away from the focus on an international agree-ment, in the hope of faster success in the case that the court system was estab-lished within the EU legal order.109 While the initial planning for that phase had revolved around a system combining existing national and EU courts,110

101. Cf. Art. 62 et seq. First Community Patent Convention, cited supra note 2.102. Ibid., Art. 69. 103. Ibid., Art. 78.104. Ibid., Art. 73. 105. Cf. Communication 2007, cited supra note 6, 10 et seq.106. Arts. 22 and 30 of the Protocol on the Settlement of Litigation Concerning the Infringe-

ment and Validity of Community Patents (Protocol 1989) to the Second Community Patent Con-vention, cited supra note 3.

107. Cf. Art. 28 Protocol 1989, cited supra note 106. 108. Cf. Art. 2(2) of the preambular provisions of the Second Community Patent Conven-

tion, cited supra note 3. 109. Cf. already Green Paper 1997, cited supra note 40, 6 et seq.110. Ibid., 13 et seq.

EU Patent 81

the proposal published in the year 2000 again returned to the idea of establish-ing a specialized Community Intellectual Property Court (CIPC). Compared to the COPAC, the CIPC brought largely uniform and fully cen-tralized jurisprudence for revocation and infringement proceedings through-out the EU territory.111 National courts would retain competence only for residuary questions, such as, for example, litigation over the right to the pat-ent or transfer and licensing questions.112 The 2000 proposal also introduced an important change in the scope of jurisdiction that still persists today: under that proposal, the CIPC is not competent for appeals against decisions of the EPO.113 Instead, the Office’s internal opposition and appeal procedures apply to the EU patent. This option was chosen “with a view to retaining for as long as possible the unified treatment of simultaneous application for a Commu-nity patent and a European patent”,114 but it has two major downsides. First, the exclusion of appeals from the competence of the CIPC clearly reduces the uniformity of patent practice and creates the need for compensatory consoli-dation of the two strands of practice in some other form. Second, and much more importantly, insofar as the EPO applies EU law when deciding on a pat-ent application, its acts must also be subject to review by an EU court. This provision may therefore jeopardize the autonomy of the EU legal order which was enshrined in Article 220 EC, now replaced by Article 19 TEU; this prob-lem is discussed in more detail below. The proposal of 2004 finally integrated this structure into the framework of the legal bases for specialized EU courts that were created under the Treaty of Nice: The CIPC (re-named Community Patent Court, CPC) took the form of a specialized, centralized tribunal under the roof of the CFI, in accordance with Article 225a EC (now 257 TFEU).115 Its competence was unchanged and related to infringement and invalidity alike,116 whereas national courts retained resid-ual competences only.117 Appeals against decisions of the CPC went to the CFI. Onward appeal to the ECJ was excluded, but the ECJ could decide to review exceptional cases posing a serious risk to the unity or consistency of EU law.118 Again, there was no competence to review decisions of the EPO.

111. Cf. Arts. 30 et seq. Regulation 2000, cited supra note 4.112. Ibid., Art. 46 et seq.113. Ibid., Arts. 15 et seq. 114. Ibid., Art. 15.115. Cf. Art. 30 Regulation 2004, cited supra note 5.116. Ibid., Arts. 31 et seq. 117. Ibid., Art. 46 Regulation.118. Cf. Art. 225(2) EC (now 256 TFEU); Art. 62 ECJ Statute.


5.3. Status quo: The latest model

In spite of the vast variety of designs, none of the preceding proposals found a majority among Member States.119 Especially the strong element of central-ization contained in the 2000 and 2004 proposals encountered heavy opposi-tion.120 There were also concerns that the CFI lacked expertise to deal with highly specialized patent cases and that, accordingly, the quality of adjudica-tion would suffer.121 In addition, the EPLA proposals of 2003 and 2005 high-lighted the need for a solution to litigation inefficiencies also under the EPC system. The EPLA put pressure on the Commission to broaden its horizon and to devise a litigation model that would also respond to the needs of the EPC system and thus avoid an entrenchment of institutional parallelism for patents in Europe. Accordingly, the Commission in 2007 made a radical policy change and announced it would alter its design proposal in three fundamental respects: decentralization, integration of the EPC system and, accordingly, return to an international law basis.122

A proposal advanced by the Czech Presidency of the first half of 2009 lays out a litigation system designed along those lines. This proposal has now been adopted by the Council under the December 2009 political agreement. It envisages a specialized European and Community Patents Court (ECPC), set up on the basis of a mixed agreement between the EU, the EPC States and the Member States, with jurisdiction for actions for infringement and revocation related to European and EU patents.123 National courts retain residual compe-tences only.124 There is no need to exhaust opposition procedures at the EPO for bringing a revocation action.125 For EU patents, the ECPC is also compe-tent for litigation relating to compulsory licensing.126

The ECPC comprises two instances, with the first instance decentralized in the sense that it consists of a central division and several local or regional

119. For an overview of the variety of discussions since the 2004 proposal cf. e.g. Pagenberg, op. cit. supra note 71, 806 et seq.

120. Cf. already Working Paper of the Belgian Presidency, 28 Oct. 2001, Council Document No. 13408/02; also Tilmann (2003), op. cit. supra note 12, 384; Schäfers, op. cit. supra note 71, 826; Schneider, op. cit. supra note 10, p. 641.

121. Cf. Regulation 2000, cited supra note 4, 14; Commission working document on the planned Community patent jurisdiction, COM(2002)480 final, 3 et seq. and 11 et seq.; Jacob, “The Perfect Patent Court” in Kur, Luginbühl and Waage, op. cit. supra note 72, p. 313, at pp. 318 et seq.

122. Cf. Communication 2007, cited supra note 6, 8 et seq.123. Cf. Art. 1 and 15(1) Agreement 2009, cited supra note 25. 124. Ibid., Art. 15(2).125. Ibid., Art. 15a(7). 126. Ibid., Art. 15(1).

EU Patent 83

divisions in the various Member States.127 In stark contrast to the 2004 pro-posal, the integration of the ECPC into the EU legal order is only very lim-ited: the only linkage ensuring control of an EU court is the obligation of the second instance to request a preliminary ruling from the ECJ over questions of interpretation of EU law.128

As was already the case with all more recent proposals, no ECPC compe-tence is envisaged as regards a review of EPO decisions. Appeals will be dealt with by EPO’s own boards of appeal. The earlier drafts for the Community patent had still envisaged an appeals line for EPO decisions to the respective patent court. The 2000 draft patent regulation, however, switched to using EPO’s boards of appeal “with a view to retaining for as long as possible the unified treatment of simultaneous application for a Community patent and a European patent. It also avoids burdening the centralised Community court with a proliferation of appeals lodged during the examination procedure and before the Community patent has been granted.”129

6. Quo vadit: Towards a workable litigation model

The litigation model now agreed by the Councnil takes an over-arching and multi-faceted approach to balancing the diverging interests of Member States and stakeholders and therefore represents the most workable compromise so far. Furthermore, that model is refreshingly innovative compared to the previ-ous models in several aspects. For once, importantly, it provides a solid basis for a de facto convergence of patent systems, patent protection and patent enforcement in Europe. In addition, regarding substance, it is a modern proposal in the sense that it incorporates a number of recent legal and political advance-ments and, therefore, represents the state of the art in IP enforce ment. Nonetheless, this latest model too is not fully workable yet and requires some limited adaptations. An assessment of the workability of the model at the basis of the 2009 political agreement and necessary adaptations is under-taken in this section.

6.1. Convergence of patent systems

Unlike any of the earlier models, the agreed model provides for flexibility to accommodate both EU and Munich patents under one common roof of litiga-tion for the first time. This is not just a move simply to avoid a duplication of

127. Ibid., Arts. 4 et seq. 128. Ibid., Art. 48. 129. Regulation 2000, cited supra note 4, 15 et seq.


infrastructure, but a much needed move to put an end to contradictory assessments of patents and interpretations of patent rights by courts, thereby also promising to harmonize interpretative differences between the various national patents and an EU patent. In short, if the ECPC is put into practice, Europe will finally be on track toward a de facto unitary patent right. That right may be formally fragmented upon the choice of the patentee, for as long as European and EU patents coexist, but where the substance of the patent in enforcement is essentially the same in every jurisdiction, the patentee’s choice of patent territories and enforcement jurisdictions will be determined, as should be the case, by economic considerations and not by others. We have already observed a similar development in the parallelism between Community trade-marks130 and harmonized131 national trademark law, which both (in second instance or under the preliminary rulings procedure) fall within the interpreta-tive competence of the ECJ.132

6.2. Rights, remedies and procedures

In terms of substance, importantly, the model now agreed contains a set of provisions on the powers of the ECPC in infringement proceedings.133 They will be flanked by a detailed act laying out the rules of procedure for the ECPC,134 which is currently still in the drafting stage between the Commission, Council and expert judges. These are modelled on the 2004 IP Enforcement Directive,135 which for the first time pr ovided for a harmonization of core pro-cedural provisions, rights and remedies in IP infringement proceedings before the national courts. Accordingly, as is true for the Enforcement Directive,136 the powers envisaged for th e ECPC go far beyond the basic level established

130. Cf. CTMR, cited supra 29. 131. Cf. Directive 2008/95/EC of the European Parliament and of the Council of 22 Oct.

2008 to approximate the laws of the Member States relating to trade marks, O.J. 2008, L 299/25.132. For examples of interpretative parallels drawn between Community trademarks and

national trademarks cf. e.g. recently Case C-62/08, UDV North America, order of 19 Feb. 2009, nyr, para 42; Case C-320/07 P, Antarctica, judgment of 12 March 2009, nyr, para 43; Case 16/06 P, Les Éditions Albert René Sàrl, judgment of 18 Dec. 2008, nyr, para 46. Cf. also the harmoniz-ing effect described by Urlesberger, “‘Legitimate reasons’ for the proprietor of a trade mark reg-istered in the EU to oppose further dealings in the goods after they have been put on the market for the first time”, 36 CML Rev. (1999), 1195, at 1196 et seq.

133. Cf. Part III Chapter IV Agreement 2009, cited supra note 25.134. Cf. for the publicly available second draft Council Working Document No. 11813/09, at

the time of publication however already amended twice (fourth draft of 16 Oct. 2009).135. Cf. supra note 22.136. Cf., e.g., Franz, “Vorschläge für ein faires Verfahren”, (2005) ZUM, 802; Frey and

Rudolph, “EU-Richtlinie zur Durchsetzung der Rechte des geistigen Eigentums”, (2004) ZUM, 522, at 525; Jaeger in Hilty, Jaeger and Kitz, op. cit. supra note 71, pp. 200 et seq.; von Hartz,

EU Patent 85

in particular by the TRIPS Agreement. The ECPC will therefore dispose of a set of modern rules, which are tailor-made for patent infringement proceedings and may be deemed essentially equivalent to the procedural rules that national courts in the various Member States have at their disposal today.137 Compared to the procedural rules that were envisaged for the EPLA’s EPJ,138 the ECPC’s procedural power s show an essentially similar approach and scope, but are generally more detailed and in many instances more comprehensive. For exam-ple, the possibility to order the production of evidence, including banking, financial or commercial documents, upon due substantiation of the claim139 or the infringer’s responsibility for the costs for corrective measures140 are fore-seen in the Enforcement Directive but absent in EPLA. Already foreseen in the Enforcement Directive and contained in EPLA in less detail are, for exam-ple, the rules on preliminary measures for preserving evidence even before the commencement of the proceedings on the merit,141 or, importantly (as one of the core novelties of the Enforcement Directive),142 a far-reaching right of inf ormation about the involvement of third parties.143 In some respects, the ECPC’s powers even go beyond the provisions of the Enforcement Directive and of EPLA, for example as regards a rule for reversal of the burden of proof for process patents.144 In other respects, the TRIPS-plus enforcement powers accorded under the 2009 Proposal, the Enforcement Directive and under EPLA run largely parallel, e.g. as concerns the provisions on damages (where EPLA is even more somewhat generous in allowing damage claims against third par-ties tolerating an infringement),145 the possibility to take measures to preserve

“Beweissicherungsmöglichkeiten im Urheberrecht nach der Enforcement-Richtlinie im deut-schen Recht”, (2005) ZUM, 376, at 379 et seq.

137. Similarly already Drexl, Hilty and Kur, “Proposal for a Directive on Measures and Pro-cedures to Ensure the Enforcement of Intellectual Property Rights – A First Statement”, (2003) IIC, 530, at 534 et seq.

138. Cf. Part IV Chapter II EPLA, cited supra note 17; Chapter IV Draft Statute of the Euro-pean Patent Court (Draft Statute EPJ), available at <www.epo.org/patents/law/legislative-initia-tives/epla.html>

139. Cf. Art. 35 Agreement 2009, cited supra note 25; Art. 6 Enforcement Directive, cited supra note 22; less far-reaching Art. 54 EPLA, cited supra note 17.

140. Cf. Art. 38(3) Agreement 2009, cited supra note 25; Art. 10(2) Enforcement Directive, cited supra note 22; unclear under Art. 58 EPLA, cited supra note 17.

141. Cf. Art. 35a Agreement 2009, cited supra note 25; Art. 7 Enforcement Directive, cited supra note 22; Arts. 70, 71 and 74 EPLA, cited supra note 17.

142. Cf. also Kur, “The Enforcement Directive – Rough start, happy landing?”, (2004) IIC, 821, at 821 and 826.

143. Cf. Art. 39 Agreement 2009, cited supra note 25; Art. 8 Enforcement Directive, cited supra note 22; Art. 68 EPLA, cited supra note 17.

144. Cf. Art. 34 Agreement 2009, cited supra note 25.145. Ibid., Art. 41; Art. 13 Enforcement Directive, cited supra note 22; Arts. 64 and

65 EPLA, cited supra note 17.


evidence and provisional and protective measures inaudita altera parte,146 the availability of freezing orders,147 the types of final corrective measures148 or court fees.149

Rightholders will clearly applaud the drive for high enforcement efficiency in the ECPC’s powers. From the point of view of their competitors or new mar-ket entrants, however, excessively ambitious enforcement powers for the ECPC create a risk of impediment of legitimate economic activity by over-enforce-ment and spurious litigation.150 In this regard, the Enforce ment Directive has rightly been criticized for its rightholder-biased approach and for a lack of clear limits to the extent of enforcement rights and remedies and of a corre-sponding expansion of the rights of defence.151 Although it already establishes “a high … level of protection”,152 that Directive yields to rights and sanctions153 in national law which are even “more favourable to rightholders”.154 On top of failing to bring about a significant harmonization of the rights of defence in IP enforcement, the Directive thus even seems to authorize the curtailment of any of its few defendant rights.155 In sum, the Enforcement Directive does not warrant or spell out the principles of procedural fairness and equality

146. Cf. Art. 35(4) Agreement 2009, cited supra note 25; Art. 9(4) Enforcement Directive, cited supra note 22; Art. 70(3) EPLA, cited supra note 17.

147. Cf. Art. 35b Agreement 2009, cited supra note 25; Art. 9(2) Enforcement Directive, cited supra note 22; Art. 73 EPLA, cited supra note 17.

148. Cf. Arts. 37a and 38 Agreement 2009, cited supra note 25; Arts. 10 et seq. Enforcement Directive, cited supra note 22; Arts. 60 et seq. EPLA, cited supra note 17.

149. Cf. Art. 42 Agreement 2009, cited supra note 25; Art. 14 Enforcement Directive, cited supra note 22; Art. 58 EPLA, cited supra note 17.

150. On this risk also Harhoff, op. cit. supra note 73, 45 et seq.; Kur, op. cit. supra note 141, 824 et seq.; Heath, “Wrongful patent enforcement – Threats and post-infringement invalidity in comparative perspective”, (2008) IIC, 307.

151. Cf. Franz, op. cit. supra note 135, 803; Jaeger in Hilty, Jaeger and Kitz, op. cit. supra note 71, pp. 202 et seq.; Kur, op. cit. supra note 141, 824 et seq.; Spindler, “Der Auskunfts-anspruch gegen Verletzer und Dritte im Urheberrecht nach neuem Recht”, (2008) ZUM, 640, at 648; Peukert and Kur, “Stellungnahme des Max-Planck-Instituts für Geistiges Eigentum, Wett-bewerbs- und Steuerrecht zur Umsetzung der Richtlinie 2004/48/EG zur Durchsetzung der Rechte des geistigen Eigentums in deutsches Recht”, (2006) GRUR Int., 292, at 293; Drexl, Hilty and Kur, op. cit. supra note 136, 530, at 530 et seq. Cf. also the critical statement on a con-text of a planned Directive on criminal sanctions for IP infringements, Hilty, Kur and Peukert, “Statement of the Max Planck Institute for Intellectual Property, Competition and Tax Law on the Proposal for a Directive of the European Parliament and of the Council on Criminal Mea-sures Aimed at Ensuring the Enforcement of Intellectual Property Rights”, (2006) IIC, 970, at 973 et seq.

152. Recital 10 Enforcement Directive, cited supra note 22.153. Ibid., Art.154. Ibid., Art. 2(1).155. Less pessimistic (intrinsic limitations in the Directive), however, Benhamou, “Compen-

sation of damages for infringements of intellectual property rights in France, under Directive 2004/48/EC and its transposition law – New notions?”, (2009) IIC, 125, at 127.

EU Patent 87

between the parties in IP enforcement, but instead relies on general principles of EU law, national laws and national courts to safeguard them. The fact that the ECPC’s powers are, as just mentioned, largely modelled on the provisions of the Enforcement Directive, may give rise to some concern that the latest model would bring about a similarly one-sided litigation system for righthold-ers and thereby entail essentially anti-competitive or anti-innovative effects. A closer look, however, reveals that these concerns are not justified, due to both the structure and competences of the ECPC, and ceilings and procedural safeguards woven into the 2009 Proposal. In terms of structure and competences, the ECPC is designed to provide for fast and low-cost revocation proceedings.156 Such proceedings provide a good defence line against “patent trolls”, help to cleanse the market from weak pat-ents and thus provide a tool to forestall spurious litigation based on weak pat-ents.157 Therefore, even if the ECPC’s procedural powers in infringement proceedings showed a bias in favour of rightholders, this advantage would at least in part be balanced out by the lowering of procedural thresholds for revo-cation, which favours competitors. The same is true as regards the lowering of thresholds for obtaining compulsory licences, illustrated earlier above, which again favour competitors only and work to re-balance possible advan-tages for rightholders in relation to rights and remedies for infringements. However, as was likewise discussed above, under the latest model these advan-tages in litigation over compulsory licences will only come into effect for EU patents,158 not European patents, which also limits the re-balancing effect of this feature. In terms of enforcement ceilings, the 2009 Proposal excludes a number of infringement claims, e.g. in relation to European patents claims against non-commercial acts, against certain experiments and testing, against infringing parts of foreign vessels, against uses authorized under the so-called farmer’s privilege or other acts specifically allowed by EU law (e.g. regarding computer programs or biotech inventions).159 While most of those ceilings are mere cod-ifications of limitations contained elsewhere, the exclusion of non- commercial infringements is, for example, an important constraint vis-à-vis the indiscrim-inate scope of the Enforcement Directive, which only exceptionally restricts the availability of claims to commercial infringements.160

Finally, the provisions for the ECPC also reveal some concern to strengthen the rights of defence and to better safeguard the interests of defendants as

156. Cf. Harhoff, op. cit. supra note 73, 18 et seq. and 45 et seq. 157. Ibid., 45 et seq. 158. Cf. Arts. 3 and 15(1)(f) and (2) Agreement 2009, cited supra note 25. 159. Ibid., Art. 14c. 160. Cf. Recital 14 and Arts. 6(2), 8(1) and 9(2) Enforcement Directive, cited supra note 22.


compared to the Enforcement Directive. One important example here is the provision on the protection of confidential information in proceedings before the ECPC, allowing in particular for a restriction of access to such informa-tion to certain persons,161 which to some extent re-balances the strong infor-mation rights162 against the defendant and third parties. The Enforcement Directive, by contrast, introduces the same strong right of information163 but fails to spell out minimum standards and tools for protecting confidentiality.164

To sum up on the procedural powers of the ECPC: enforcement in the ECPC system is designed to be just as attractive and efficient as enforcement by national courts or under EPLA, but the ECPC rules are, in particular, more detailed and comprehensive than EPLA, and more balanced and better cali-brated than the Enforcement Directive. Not least for the structural improve-ments and ceilings introduced, the ECPC may promote innovation more than any other IP enforcement system currently in place in Europe.

6.3. Sacrificing effectiveness for flexibility?

6.3.1. Points of compromiseThe model agreed by the Council tries to bridge several gaps, particularly between the Munich system and the EU litigation structure, between the need for uniformity and decentralization and user proximity, and between the cul-tural and legal differences of the various existing systems, e.g. as regards language requirements, the involvement of technical judges and the like. This bridge-building character accounts for two important characteristics of the Proposal: firstly, it is much more of a conglomeration of compromise than any of the previous proposals advanced and accordingly also boasts the best chances ever for its implementation. Secondly, it seeks to set up a litigation structure that is exceedingly flexible, to accommodate party preferences in respect of several of its key features. Parties may litigate European patents or EU patents165 and quarrel over the widest possible range of claims (infringe-ment, validity, compulsory licensing, damages etc.), all before the same court,166 they may or may not use arbitration,167 they may agree to bring the action before any regional court of their choice or before the central court,168 they may or may not have technical judges depending on the court they opt

161. Cf. Art. 40 Agreement 2009, cited supra note 25.162. Ibid., Art. 39.163. Cf. Art. 8(1) and (2) Enforcement Directive, cited supra note 22.164. Ibid., particularly Recital 20 and Arts. 6(2), 7(1) and 8(3)(e). 165. Cf. Art. 1 Agreement 2009, cited supra note 25.166. Ibid., Art. 15. 167. Ibid., Art. 17.168. Ibid., Art. 15a(6).

EU Patent 89

for,169 they may generally use any of the languages of the EU or of other con-tracting States or, alternatively, they may exclude certain procedural languages by opting for a specific regional division,170 they may or may not motion for the court to waive translation requirements or to provide interpretation,171 they may choose a lawyer or a patent attorney for their representation172 and so on and on. The list of flexible arrangements to cater to party needs and enhance the ECPC’s attractiveness is quite impressive. The compromise and flexibility characterizing the latest patent litigation model may raise concern over possible negative repercussions for the effec-tiveness of the ECPC. From the economic point of view, the ECPC was recently assessed as promising high efficiency in terms of immediate (litigation cost reductions) and follow-on (better assessment of patent value) positive eco-nomic effects for litigants and for the internal market (elimination of weak pat-ents and increased patenting activity).173 Apart from other elements, such as the arbitration procedure or the involvement of patent attorneys, the Propos-al’s flexibility in terms of decentralization was particularly highlighted.174 Eco-nomic analysis therefore clearly appraises the flexible approach chosen as cost-efficient. From the legal point of view, it was shown earlier above that the creation of a unitary right and of a unified litigation structure already have significant intrinsic positive effects as compared to the status quo. Pointedly stated, the repercussions of fragmentation and duplication of litigation under the current system are so bad that almost any type of unified litigation will bring some kind of improvement. The positive effect of a unified structure is only counteracted by follow-on flaws of design, such as, prominently, bad quality of the jurisprudence and case law175 or high entry thresholds for parties (e.g. costs, distance, languages etc.), which render the system hard to use and unattractive. The anticipated quality of the ECPC’s jurisprudence will be discussed separately below. Apart from the simple assertion that there will be some benefit for patent litigation already from the set-up of a unified system, there is the question whether there is room for improvement and a legal fine-tuning of the structure so as to render it as effective as possible. The elements probably deserving most attention in this regard are the inter-linkages of procedure between, firstly,

169. Ibid., Art. 6(2) and (5).170. Ibid., Art. 29.171. Ibid., Art. 31. 172. Ibid., Art. 28. 173. Cf. Harhoff, op. cit. supra note 73, 22 et seq., 36 et seq. and 52 et seq.174. Ibid., 50. 175. On this point in particular, cf. Wissenschaftlicher Beirat beim Bundesministerium für

Wirtschaft und Technologie, op. cit. supra note 100, 21 et seq.


the various decentralized courts and, secondly, between the ECPC and the EU courts. If designed badly, these links carry the potential of re-introducing prob-lems such as forum-shopping, procedural delays, duplication of litigation or argument and even contradictory outcomes into the unified system.

6.3.2. Decentralization and jurisdictionAs regards entry thresholds, the aforementioned flexibility features, most importantly the decentralized character and language arrangements, precisely serve the purpose of keeping them low and enhancing the attractiveness of the ECPC for litigants. Decentralization has long been a key concern in the public discussions surrounding EU patent litigation.176 However, there are only two conceivable ways to implement decentralization, either the enforcement of the EU patent by national courts under the (preliminary rulings-based) supervision of the ECJ, as is the model implemented for Community trademarks and designs, or the formation of local divisions of a supranational court. The enforcement of Community trademarks and designs by national courts works better than the enforcement of national IP rights, but it is still hampered by certain flaws resulting from the fragmentation of national laws and the prolif-eration and parallelism of competent courts.177 As already pointed out above, the unified litigation system avoids those problems from the outset and is therefore more effective than the system of enforcement in national courts. Against this background, the model of decentralization chosen by the 2009 Proposal enhances the effectiveness of that system. Finally, particular care was taken to provide a low-cost structure that will be cheaper than the EPLA model and comparable to the current cost-level in Germany.178 All in all, therefore, the benefits flowing from the ECPC’s unified structure are reinforced by sen-sitivity to party preferences and low entry thresholds, rather than reversed. It must be ensured in a decentralized system that the distribution of juris-diction among the courts is clear and leaves no room for duplication or torpe-doing, and that the professional quality and speed are similar for all of the courts so as to exclude forum-shopping within the system. The latter question of comparable high quality of jurisprudence is discussed separately in the next section. Regarding the distribution of jurisdiction under the 2009 Proposal,

176. Cf. Leonard, op. cit. supra note 71, 264; Tilmann (2003), op. cit. supra note 12, 383; Leith, op. cit. supra note 12, 250 et seq.; Brinkhof, “Die Schlichtung von Patentstreitigkeiten in Europa – Über Traum, Wirklichkeit und Vision”, (2001) GRUR, 600, at 602 et seq.; Nooteboom, “Die Arbeiten zur Schaffung einer Gemeinschaftspatentgerichtsbarkeit” in Keller, Plassmann and von Falck (Eds.), Liber Amicorum for Winfried Tilmann (Köln, 2003), p. 567, at pp. 575 et seq.; Hölder, op. cit. supra note 12, 45 et seq.; Schäfers, op. cit. supra note 71, 824, at 826.

177. For a summarized assessment, cf. Jaeger in Hilty, Jaeger and Kitz, op. cit. supra note 71, pp. 157 et seq.

178. Cf. Harhoff, op. cit. supra note 73, 21.

EU Patent 91

actions for revocation and concerning compulsory licences must be brought before the central court. Actions concerning infringements, injunctions, dam-ages and prior use can be brought before local divisions. Here, the 2009 Pro-posal establishes the principles actor sequitur forum rei (domicile of the defendant) and forum delicti (place of infringement) as parallel bases for juris-diction.179 There is, in particular, no common jurisdiction for actions against multiple defendants.180 The generous catalogue of jurisdictional rules of the Brussels I Regulation181 is therefore melted down to two places of jurisdiction. Furthermore, there is no priority rule (lis pendens)182 in the 2009 Proposal in favour of the court first seized for the same cause of action or related actions, e.g. in a multi-State infringement or based on the second of the two grounds of jurisdiction. In fact, there is not even a competence to stay proceedings on the division’s own motion or to hand the case to the central division. Given that the proliferation of jurisdictions under the Brussels I Regulation and the openness of the priority rule to abuse are major reasons for today’s procedural problems in patent litigation, this quest for simple rules and brev-ity is generally to be welcomed, but only as long as this contributes to increas-ing the effectiveness of the system. In fact, however, the 2009 Proposal may be doing too much of a good thing here. If the lack of a priority rule in partic-ular is not a lapse but intentional, the idea must have been that if multiple sei-zures occur, possible diverging outcomes will be consolidated upon appeal. This, however, would run counter to the aims of providing fast and cost-effec-tive litigation to the parties and may foster abusive multiple litigation by finan-cially potent litigants in a way similar the aforementioned “trolling”-problem. A simple and obvious solution to this problem would be to require parties in multiple seizures to agree on one place of jurisdiction within a given time or, in the absence of agreement, provide for a mandatory devolution of the case to the central court. As regards the bases for jurisdiction, first of all, the places of domicile and damage should be complemented by a common jurisdiction for multi-party proceedings. This is mandated, again, by the need for cost-effective and swift litigation and the avoidance of contradictory outcomes. If the intention of the non-inclusion of this forum into the 2009 Proposal was to protect defendants against litigation under unfamiliar systems (e.g. in divisions not operating in their language) and to forestall forum-shopping by plaintiffs, a fair solution might be to transfer such cases to the central division upon joint application by the defendants. This would put the defendants in control of the decision

179. Cf. Art. 15a(1) Agreement 2009, cited supra note 25.180. Cf., e.g., Art. 6(1) Brussels I Regulation, cited supra note 21.181. Brussels I Regulation, ibid. 182. Ibid., Art. 27.


whether to litigate before local divisions instead, and to take on the ensuing higher costs. Alternatively, a mandatory spider-in-the-web-approach could be envisaged, declaring the local division competent where the main defendant (the “spider”) is located.183 The two places of jurisdiction envisaged under the 2009 Proposal (domicile and damage) contain only minor ambiguities and these are, furthermore, similar to those existing and tolerated under the Brus-sels I Regulation today. Accordingly, only minor clarifications are called for here. In relation to the jurisdiction of domicile, lack of clarity results from the lack of declaration whether the domicile of an undertaking is to be determined according to the incorporation theory (place of incorporation) or the head office theory (location of head office or siège réel). Given the existing division in the legal traditions of the Member States, the ECJ has thus far shied away from adopting one of the two concepts for EU law and accepts both.184 So does the Brussels I Regulation.185 In the area of jurisdiction, the duplication of the domi-cile of an undertaking necessarily entails a duplication of litigation fora upon the choice of the plaintiff, who is then given the option to sue at the place of incorporation as well as at the location of the head office.186 A similar prolif-eration of options of the plaintiff to sue can be observed for the jurisdiction of the place of damage. The ECJ interprets Brussels I Regulation’s damages forum187 as covering both the place where the damage occurred and the place of the event giving rise to it.188 In patent cases therefore, the plaintiff has the option of going to court either at the place where the counterfeited goods were produced, where they were marketed or where there is a threat189 that they will be marketed. It will therefore be possible to take the defendant to court wher-ever he has a point of sale, an agent or customers. In some cases, this may mean that the defendant can actually be sued throughout the EU territory (and, in the case of the ECPC, even beyond). Here again therefore, limiting factors for extreme cases of proliferations of competence may be envisaged, such as an option for the local division to refer the case to the central division or, vice versa and probably more effective to forestall parallel action before several

183. Equally (for the similar problem under the Brussels I Regulation) already Kur and Metzger et al., op. cit. supra note 72, 6 et seq.

184. Cf., recently, Case C-210/06, Cartesio, judgment of 18 Dec. 2008, nyr, para 109; con-sistent jurisprudence.

185. Cf. Art. 60(1) Brussels I Regulation, cited supra note 21. 186. Cf. also Jaeger in Hilty, Jaeger and Kitz, op. cit. supra note 71, pp. 162 et seq. 187. Cf. Art. 5(3) Brussels I Regulation, cited supra note 21. 188. Cf., e.g., Case 21/76, Mines de potasse d’Alsace, [1976] ECR 1735, paras. 15 et seq.;

Case C-220/88, Dumez, [1990] ECR I-49, para 10; Shevill, cited supra note 69, para 20; Case C-364/93, Marinari, [1995] ECR I-2719, para 11; Case C-18/02, Danmarks Rederiforening, [2004] ECR I-1417, para 40; Case C-168/02, Kronhofer, [2004] ECR I-6009, para 16.

189. Cf. Art. 15a(1)(a) Agreement 2009, cited supra note 25.

EU Patent 93

divisions, the central court’s right to attract all related cases from local divi-sions for common decision.

6.3.3. Effectiveness of the review procedure for EU law Another point potentially reducing the system’s effectiveness is the link between the ECPC and the EU courts. As will be shown further below, the involvement of an EU court on questions of EU law is required under Article 19 TEU (which replaces Art. 220 EC) and can therefore not be done away with. However, as is known from national experience with the preliminary rulings procedure under Article 267 TFEU (ex 234 EC), involvement of the ECJ delays the procedure by roughly one-and-a-half years.190 The 2009 Proposal conse-quently seeks to limit the involvement of the ECJ to a minimum. The ECJ must be asked for preliminary reference over questions of Community law in cases pending before the ECPC’s appeals instance only.191 On points of facts and for cases settled in first instance, therefore, there is neither a requirement nor, in fact, a competence to request preliminary rulings. As regards requests for pre-liminary rulings in second instance, that obligation may seem quite extensive since any case of application of a future EU patent regulation would necessar-ily involve an interpretation of Community law and thus require preliminary references in all cases reaching second instance. In particular, there are no limiting qualifications in the 2009 Proposal, e.g. only for significant questions of EU law or the like. However, the ECJ imposed such limitations itself under its doctrines of acte eclairé192 and acte clair,193 relieving courts from t he duty to request a preliminary ruling if the same or a similar question has been answered by the ECJ before or in the very exceptional case that the answer is obvious.194 However, those limiting effects only set in as a body of ECJ case law on the EU patent regulation gradually develops. In the early stages of operation of the ECPC, therefore, every case involving the EU patent regula-tion before the appeals court is likely to require a preliminary ruling and thus delay the procedure. It is difficult to assess how such a delay might impact the attractiveness of the ECPC system in its crucial early stages of operation. In particular, some parties may find the delay plus the opportunity to win an extra

190. In 2008, the average duration of preliminary ruling proceedings was 16.8 months, as against 19.3 months in 2007 and 19.8 months in 2006. Cf. Annual Report of the European Court of Justice, partly available at <curia.europa.eu/jcms/jcms/Jo2_7000/> 94. There is, however, an accelerated procedure available for exceptional cases under Art. 104a ECJ Rules of Procedure.

191. Cf. Art. 48 Agreement 2009, cited supra note 25. 192. Fundamental: Joined Cases 28-30/62, Da Costa, [1963] ECR 63, 81; Case 283/81,

CILFIT, [1983] ECR 3415, para 14.193. CILFIT, cited supra note 191, para 15.194. Extensively Lenaerts and Arts, Procedural Law of the European Union, 2nd ed.

( London, 2006), paras. 2-048 et seq.


review instance a benefit. From the point of view of effectiveness, the interest in setting case law precedents that are legally correct (i.e. have been double-checked for EU law compatibility) clearly outweighs the negative impact of the delay. Such an effect of immediate certitude about the correctness and reli-ability of the case law of the court of appeal can, in particular, not be attained by laying down a mere competence of the ECJ to attract or reopen cases involv-ing a serious risk of the unity or consistency of EU law being affected (similarly to what is laid down for judicial panels in 256 TFEU (e2x Art. 225(3) EC). All in all therefore, although an even more limited involvement of the ECJ in ECPC appeals procedures may seem appealing at first glance, the rule for preliminary rulings in the 2009 Proposal lies in the interest of developing solid and reliable case law for EU patents and should not be changed.

6.3.4. ResultTo sum up on this point, the model now politically agreed by the Council includes political compromise and, even more so, a considerable amount of flexible procedural provisions in the interest of parties. Importantly however, these features do not generally come at the cost of hampering the ECPC’s effectiveness derived from the establishment of a unified litigation structure. While the overall structure under the ECPC proposal is therefore a clear improvement vis-à-vis the status quo, some details of the structural set-up should nonetheless be changed. In this regard, an important provision absent in the 2009 Proposal is a priority or devolution rule to avoid multiple parallel proceedings before the divisions of first instance. Without such a rule, proceed-ings before the divisions of first instance may be duplicated and thereby re-introduce some of the problems of the current system into the ECPC model. For the same reason, a rule for the joining of proceedings against multiple defendants before one division seems warranted for. That rule could, for exam-ple, follow a “spider-in-the-web” concept. Possible, albeit less pressing, are also clarifications to rule out a proliferation of jurisdictions from ambiguities in the rules of jurisdiction at domicile or for torts. Finally, undesired procedural effects hampering the effectiveness of the ECPC structure do not flow from the linking of the ECPC system with the ECJ either. Although the latest model contains a broad obligation to request preliminary rulings for questions of interpretation of EU law before the appeals instance, that obligation should not be diminished in the interest of developing solid case law for EU patents as fast as possible.

EU Patent 95

6.4. Quality of ECPC Jurisprudence

6.4.1. Initial concernsAnother main concern over the creation of a unified patent judiciary was the quality of jurisdiction.195 Criticism was directed against the EU courts196 as well as against national courts.197 As r egards the latter courts in fact, mistrust in the quality of foreign judgments (prejudice against the so-called “Palermo judge”) lies at the base of a number of application problems of the Brussels I Regulation in cross-border cases, e.g. the very fact that the territoriality prin-ciple for IP rights is still enshrined in that Regulation or the idea of anti-suit injunctions relating to foreign jurisdictions.198 Accordingly, the ECJ was repeat-edly called upon to ensure respect of the principle of mutual trust between the courts of different Member States,199 which underpins the Brussels I Regulation200 and forms a precondition for the harmonization of jurisdiction envisaged therein. Economic models,201 as well as a glance across the Atlanti c to the US experience202 with the “pro-patent” jurisprudence of the centralized US patent court “Court of Appeals for the Federal Circuit” (CAFC), both suggest that a strong preoccupation with the right amount of professional expe-rience in patent law is indeed imperative in the set-up of a unified litigation structure in order to forestall a degradation of patent standards through bad case law and ensuing anti-innovative effects. If the unified court applies lax

195. For an overview of discussions on jurisdictional quality, cf. Pagenberg, op. cit. supra note 71, 807, at 809, 811 and 818 et seq.

196. Cf. Regulation 2000, cited supra note 4, 14; also Sedemund-Treiber, “Strukturen einer europäischen Patentgerichtsbarkeit”, (1999) Mitt. Pat. Anw., 121, at 123.

197. Cf. Tilmann (2003), op. cit. supra note 12, 382; Willems, op. cit. supra note 57, 11; Schäfers, op. cit. supra note 71, 826; similarly Regulation 2000, cited supra note 4, 14.

198. Cf. also Jaeger in Hilty, Jaeger and Kitz, op. cit. supra note 71, p. 177.199. Cf. e.g. Case C-159/02, Turner, [2004] ECR I-3565, para 28; Case C-116/02, Gasser,

[2003] ECR I-14693, para 72; Case C-185/07, Allianz, judgment of 10 Feb. 2009, nyr, para 14; Case C-341/04, Eurofood IFSC, [2006] ECR I-3813, paras. 40 et seq.; Case C-195/08 PPU, Rinau, [2008] ECR I-5271, paras. 50 et seq.

200. Cf. Rec. 21 Brussels I Regulation, cited supra note 21. Cf most recently on anti-suit injunctions, Case C-185/07, Allianz Generali Assicurazioni Generali v. West Tankers, judgment of 10 Feb. 2009, nyr, annotated by Meidanis and Giannakoulias, 46 CML Rev., 1709–1724.

201. Cf. Hall, Graham, Harhoff and Mowery, “Prospects for Improving U.S. Patent Quality via Post-grant Opposition”, NBER Working Paper No. W9731 (2003), 3 et seq.; Wissenschaft-licher Beirat beim Bundesministerium für Wirtschaft und Technologie, op. cit. supra note 100, 21 et seq.

202. CAFC is short for the centralized US patent court “Court of Appeals for the Federal Circuit”. Cf. e.g. Hall et al., op. cit. supra note 200, pp. 1 et seq., 7 et seq. and 15; Wiss. Beirat beim Bundesministerium für Wirtschaft und Technologie, op. cit. supra note 100, p. 6; Lemley and Allison, “Empirical evidence on the validity of litigated patents”, 26 AIPLA Q.J. (1998), 26, at 26 et seq.


patent standards or develops a body of impractical case law, it is better for businesses and innovation to retain a fragmented system of litigation with at least a few very strict courts.203 Furthermore, as was mentioned before, ensur-ing high quality and standards within a patent litigation system is also crucial to forestall forum-shopping among the local divisions of a decentralized structure. As important as the problem of ensuring the quality of jurisprudence may there-fore be, it is also a problem that any new court system faces, be that the ECPC, the EPLA’s EPJ or any other. In other words, the only alternative to some degree of trial and error in the setting-up of a unified judiciary is to abandon the unification project altogether and to stick with enforcement in national courts. Nonetheless, with the right precautions taken, the risk that a unified court will hand down bad judgments can be rendered negligible. The agreed model answers to the preceding discussions on judgment qual-ity and incorporates the so far (also in comparison to the EPLA)204 most evolved elements to ensure professionalism of ECPC judges. Those elements are pri-marily the aforementioned flexible rule for the inclusion of technical judges,205 the authorization of dissenting opinions from the bench,206 the establishment of a professional training framework for judges “to improve and increase … patent litigation expertise”,207 the formation of multinational panels in local divisions208 and, finally and importantly, a sophisticated appointment proce-dure to the office of judge.209

6.4.2. Technical judgesA flexibility rule for technical judges in the latest proposal allows States with-out experience or trust in technical judges to refrain from having them. This is a workable compromise allowing both litigation traditions, with technical judges (such as in Germany or the UK) and without them (such as in France or the Netherlands),210 legal purism and interdisciplinarity, to coexist. On the

203. Cf. Wiss. Beirat beim Bundesministerium für Wirtschaft und Technologie, op. cit. supra note 100

204. Cf. particularly Art. 2 Draft Statute EPJ, cited supra note 137.205. Cf. Art. 6(2) and (5) Agreement 2009, cited supra note 25. 206. Ibid., Art. 51(2). 207. Ibid., Art. 14.208. Ibid., Art. 6(4). 209. Ibid., Art. 11; Art. 3 Draft Statute ECPC; both cited supra note 25.210. Cf. on the statistics of staffing the study by Mejer and van Pottelsberghe de la Pot-

terie, The Economic Implications of a Fragmented Patent System in Europe, CESifo Venice Summer Institute (2008), 8, available at <www.cesifo-group.de/portal/page/portal/ ifoContent/N/neucesifo/ CONFERENCES/ SC_CONF_2008/ Venice%20Summer%20Institute%202008/vsi08-Ghosal/vsi08_rr_Pottelsberghe.pdf>

EU Patent 97

appeals level, there will always be technical judges in a ratio of 2:3.211 This means that there will be a significant technical input, but in the end lawyers will have the final word by virtue of their majority of votes212 on the panel and because of the compulsory chairing213 of panels by a legal judge. This ensures that at the appeals level at latest, technical considerations will be sufficiently heeded without compromising the legal quality of the argument. The mixed composition of the appeals instance also means that if a case is referred to the ECJ for preliminary ruling on points of EU law, the ECJ will, insofar as those points of EU law concern or bear on patent law, have the arguments of the technical judges available and will be able to draw on them for its own judgment.214

6.4.3. Dissenting opinionsDissenting opinions (especially when based on technical considerations) may serve as a source of additional information and clarification for the ECJ. Arguments expressed in dissenting opinions may also be drawn upon by prac-titioners for the sake of argument in subsequent cases, forcing the bench to reconsider and to improve or abandon its previous position. Finally, the threat of a dissenting opinion being expressed in a case already influences the delib-eration process and fosters compromise and the elaboration of balanced argu-ments.215 In other words, dissenting opinions foster a culture of competition of arguments216 and reinforce discussions among the ben ch and with parties.

6.4.4. Ensuring a multinational composition of the benchThe agreed proposal foresees multinational panels at all instances.217 Particularly as regards the local divisions of first instance, the multinational composition of panels helps to crack open and gradually discontinue national peculiarities in jurisprudential tradition by bringing in an outside legal approach. This entails two quality-enhancing functions. On the one hand, the legal debate on the panel is invigorated and more arguments are considered by the bench. This should also reduce the number of necessary corrections of judgments upon appeal.

211. Cf. Art. 7(1) Agreement 2009, cited supra note 25.212. Ibid., Art. 51(1). 213. Ibid., Art. 7(2). 214. Cf. generally de la Serre and Sibony, „Expert evidence before the EC Courts“,

45 CML Rev., 941–985.215. Cf. Steiner, “Die Bedeutung der Dissenting Opinion für den rechtspolitischen Diskurs

– Bereits die Ankündigung kann die Kompromissfähigkeit des Senats beleben”, (2007) ZRP, 245, at 246.

216. Cf. in this Sense Nard and Duffy, “Rethinking Patent Law’s Uniformity Principle”, 101 Northwestern Univ. L. Rev. (2007), 1619, at 1623.

217. Cf. Art. 6(1) and 7(1) Agreement 2009, cited supra note 25.


Under the preliminary rulings procedure of Article 267 TFEU (ex 234 EC), by contrast, outside legal thinking is externally218 imposed upon national courts in a separate procedure and there is no active dialogue219 between legal tradi-tions. On the other hand, the multinational composition of local divisions exercises a harmonizing effect as to the quality and approaches taken by the courts of first instance which would otherwise only kick in on appeal and which forestalls a re-introduction of forum shopping into the unified litigation system. Both the pro-dialogue and the anti-forum-shopping effects are probably some-what reduced but certainly not marred by the fact that local panels are to be composed of domestic and outside judges in a 2:1 ratio and that the domestic judges will therefore have the final word on those panels.220

6.4.5. Appointment of judgesA crucial point in ensuring the quality of jurisprudence is the procedure for appointment to office. As regards the ECJ and the General Court,221 until recently no procedural safeguards existed against a selection of judges on political grounds instead of professional qualification.222 Sensitivity towards this problem among EU stakeholders has, however, led to recent improvements. Both the appointment procedure for the recently established EU Civil Service Tribunal223 and the reformed general appointment procedure for EU courts under the Treaty of Lisbon224 foresee the establishment of advisory p anels consisting of independent experts (retired EU judges, members of national supreme courts or other lawyers of recognized competence), which assess the suitability of proposed candidates and formulate appointment recommenda-tions. The appointment procedure for the ECPC will follow the same track and strive for objective evaluation through public advertising of vacancies and evaluation by the Advisory Committee for the ECPC.225 The Advisory Committee is to be composed of “patent judges and practitioners … with the

218. This applies notwithstanding obligations for national courts to take the EU interest into account on their own motion, e.g. by way of consistent interpretation. Cf Art. 4(3) TEU (which replaces Art. 10 EC); Case 14/83, von Colson, [1984] ECR 1981, para 26.

219. This applies notwithstanding the euphemistic description by the ECJ of the exchange of documents under the preliminary rulings procedure as a form of dialogue, cf. e.g. Cartesio, cited supra note 183, para 91; Case C-2/06, Kempter, [2008] ECR I-411, para 42.

220. Cf. Art. 51(1) Agreement 2009, cited supra note 25.221. Cf. Art. 223 EC.222. Cf. also Rasmussen, “Present and future European judicial problems after enlargement

and the Post-2005 ideological debate”, 44 CML Rev. (2007), 1661, at 1668 and 1676 et seq.223. Cf. Art. 3 Annex to Council Decision 2004/752/EC, Euratom, of 2 Nov. 2004 estab-

lishing the European Union Civil Service Tribunal, O.J. 2004, L 33/7 (now Annex to the ECJ Statute).

224. Cf. Art. 255 TFEU..225. Cf. Art. 11 Agreement 2009; Art. 3 Draft Statute ECPC; both cited supra note 25.

EU Patent 99

highest recognised competence”226 and will therefore guarantee a selection of candidates free of political influence and on professional merits alone. Since all judges for the ECPC, legal as well as technical, will have to undergo this selection procedure, jurisprudence in the unified litigation system will be above reproach in terms of professionalism.

6.5. Alternatives to the Model of the 2009 political agreement

It was shown that in the years of discussions preceding that agreement, almost any conceivable court set-up was discussed and rejected. This includes, in particular, differences in the scope of jurisdiction (especially regarding the separation or combination of revocation and infringement proceedings), vary-ing degrees of integration into the EU legal order (from international law to a direct Treaty base in Art. 225a EC (now 257 TFEU) and different models of centralization and decentralization (from national courts enforcement to a sin-gle centralized court) and of stronger or weaker harmonization of patent juris-prudence (from preliminary references to direct appeals). Nonetheless, the question arises as to whether any additional alternatives to the structure for the ECPC that is sketched out in the agreed 2009 Proposal exist or whether perhaps one of the previous models would have been more desirable. This is dealt with in the following.

6.5.1. Methodology for assessing alternativesEffectiveness of the litigation system must be the primary concern when dis-cussing structural design. Effectiveness means ensuring the inner functionality of the court and extends to factors such as the scope of competences and pro-ceedings, a clear delineation of jurisdiction and, of course, the quality of judg-ments, the avoidance of contradictory judgments and the availability of swift and unambiguous correction of contradictions or bad quality. By contrast, the accessibility of the system for parties, as important as that may be, is only a second-level concern compared to effectiveness: difficult access to a good court is still better than easy access to a bad court. Accessibility extends to factors such as decentralization, language arrangements and costs. On the third level, behind elements ensuring effectiveness and accessibility, come elements seeking to make the best possible use of available synergies in order to render the setting-up of the system quicker and less costly. Synergies arise from, e.g., a recourse to existing infrastructure and expertise. Ideally, all of these aims, effectiveness, accessibility and high synergies, would be combined for a litigation system. A closer look, however, reveals

226. Art. 57b(2) Agreement 2009, cited supra note 25.


quite quickly that these aims are often contradictory and thus a decision must be made whether to adopt one or the other of two possible but mutually incom-patible features. A simple example for this dilemma is the dichotomy between effectiveness and decentralization. In a highly centralized system, the risk of contradictory jurisprudence, procedural delay, forum-shopping etc. may be less than in a decentralized system. However, a decentralized system is more easily accessible. The above three-level test suggests that when a balance must be struck between incompatible objectives, priority should be given to ele-ments ensuring effectiveness over elements ensuring accessibility over ele-ments striving for synergies. In the aforementioned example therefore, centralization would generally prevail over decentralization. However, this result is not correct, because the initial presumption that centralization equals effectiveness is not correct: It was shown before that a centralized court can be ineffective if the quality of jurisprudence is bad.227 Furthermore, even within a centralized system, different strands of jurisprudence may emerge among chambers. Examples here are the CAFC’s so-called “panel dependency”,228 the back and forth between a progressive CFI and a more cautious ECJ (e.g. over issues like the opening-up of the criterion of individual concern in Art. 263 TFEU (ex 230 EC) or the recognition of the EU Charter of Fundamental Rights as binding) or the sometimes divergent decision-making practice of the boards of appeal at the EU’s trademarks and designs office OHIM. In other words, centralization of a court system does not per se enhance effectiveness, but only insofar as it effectively guarantees uniform jurisprudence. The latter function, however, may be safeguarded also by the establishment of a com-mon appeals instance. On the other hand, a decentralized second (and final) instance would be detrimental to effectiveness, since harmonization would then have to be achieved by other and less effective229 means, e.g. preliminary references from the second instance to the ECJ. As a result therefore, the above three-level test suggests to resolve the conflict between centralization and decentralization in favour of centralization at the appeals level (effectiveness over accessibility) but in favour of decentralization at the level of the first instance (where effectiveness does not bite and accessibility therefore pre-vails).

227. Cf. e.g. Hall et al., op. cit. supra note 200, pp. 1 et seq., 3 et seq., 7 et seq. and 15; Wiss. Beirat beim Bundesministerium für Wirtschaft und Technologie, op. cit. supra note 100, pp. 6 and 21 et seq.; Lemley and Allison, op. cit. supra note 201, 26 et seq.

228. Cf. Nard and Duffy, op. cit. supra note 215, 1627. 229. For the disadvantages of the preliminary reference procedure as compared to direct

actions and appeals for individual rights protection, cf. extensively Opinion of A.G. Jacobs in Case C-50/00 P, Unión de Pequeños Agricultores, [2002] ECR I-6677, paras. 61 et seq.

EU Patent 101

These considerations can be extended to all of the features of the latest model for a check as to whether the choices made are appropriate for providing a lit-igation system that is highly effective but yields to considerations of accessi-bility and synergies where possible. Whenever the answer to this question is negative, an adaptation is called for. This approach allows for a quick and reli-able check of the ECPC’s central features against possible alternatives.

6.5.2. An alternative balance between centralization and decentralization?It was just illustrated that the balance between centralization and accessibility struck by the 2009 Proposal is appropriate. The alternatives here are decen-tralization with recourse to existing national courts and preliminary references to a central judiciary (e.g. the ECJ or, in the future, a specialized panel of the CFI) or the creation of a fully centralized (EU law or international law based) judiciary. Both latter models have been under discussion and were rejected before, neither yields significant efficiency gains compared to the ECPC model and the benefit of better accessibility of the national courts model is outweighed by disadvantages on the effectiveness level (in particular attribution of juris-diction and harmonization of jurisprudence) and must therefore be disregarded under the above three-level test.

6.5.3. Alternatives to the combination of revocation and infringement proceedings

A combination of revocation and invalidity proceedings yields clear gains for effectiveness (in particular no break-up of related proceedings and no risk of irreconcilable judgments) and is, as such, welcomed by most authors.230 A model incorporating a separation of those proceedings is therefore no viable alternative. On the other side of the spectrum, it may be asked whether the combination of the two proceedings under the ECPC goes far enough, since the EPO’s own opposition and invalidation procedure remains separate from the ECPC sys-tem. The combination of revocation and infringement proceedings before the same court seeks to respond to the practical fact that most infringement claims face a counter-claim for invalidity raised in defence. If this counter-claim or counter-action for invalidity leads to the separation of proceedings, the

230. Cf., e.g, Krieger, “Nochmals: Die Aussetzung des Patentverletzungsprozesses”, (1996) GRUR, 941, at 942; Beier, “Die Rechtsbehelfe des Patentanmelders und seiner Wettbewerber im Vergleich – Eine rechtsvergleichende Untersuchung zur Chancengleichheit im Patentverfahren”, (1989) GRUR Int., 1, at 1 et seq.; Schäfers, op. cit. supra note 71, 825; Sedemund-Treiber, op. cit. supra note 195, 121 et seq.; Pakuscher, “Zur Zuständigkeit des Bundesgerichtshofs und des Bundespatentgerichts in Patentnichtigkeitsverfahren”, (1995) GRUR, 705, at 708; Land-fermann, op. cit. supra note 12, 347; different Willems, op. cit. supra note 57, 13 et seq.; balanced Tilmann (1998), op. cit. supra note 12, 326 et seq. and 333 et seq.


effectiveness of patent litigation may be significantly hampered.231 If, how-ever, the dispute does not involve infringement but only extends to the inva-lidity of the patent, as is the case with EPO opposition proceedings, there is no risk of separation of related procedures. In fact, taking opposition away from EPO would have the contrary effect of bringing about a separation in pat-ent practice, i.e. the EPO granting patents on the one and the ECPC revoking them on the other. If, however, the EPO retains competence for opposition, but EPO’s decisions may subsequently be corrected by the ECPC, this will exert direct influence on the conditions for granting patents. This is therefore a more effective solution than an excessive transfer of revocation competences to the ECPC. It has to be pointed out, however, that the ECPC under the 2009 Proposal does not have the aforementioned competence to review EPO deci-sions. This competence should therefore be created not only for reasons of effectiveness, as argued here, but, as will be shown further below, already as a requirement of general EU law.

6.5.4. Other featuresOther main features of the 2009 Proposal are flexibility and responsiveness to party preference in terms of the availability of technical judges (effectiveness, level one), multinational composition (effectiveness, level one), language requirements (accessibility, level two) and costs (accessibility, level two). These features were discussed before and assessed as enhancing the effective-ness or accessibility of the system respectively. These features are comple-mentary instead of incompatible with one another, so that there is no need to subject them to the three-level test. In relation to the reduction of places of jurisdiction of the local first level divisions to domicile and the place of dam-ages, it was likewise discussed above that this reduction generally enhances effectiveness (and therefore overrides the argument of easier access to court through more places of jurisdiction), but that this limitation probably goes too far. It was argued that a common jurisdiction for actions against multiple defen-dants would enhance the effectiveness (avoidance of contradictory outcomes) and accessibility (costs and duration as complementary factors) of proceedings. Equally, a priority rule is called for, to render the ECPC system effective (fore-stalling abuse) and easily accessibly (costs and duration as complementary factors).

6.5.5. ResultTo sum up on this point, a structured assessment of the ECPC model shows that generally, a good balance was struck between overriding requirements of

231. Different Willems, op. cit. supra note 57, 13 et seq.; balanced Tilmann (1998), op. cit. supra note 12, 326 et seq. and 333 et seq.

EU Patent 103

effectiveness and ancillary considerations of accessibility. There is limited need for adaptations from the points of view of effectiveness and accessibility (e.g. regarding review of EPO decisions and places of jurisdiction), but cer-tainly no need to create a fully-fledged alternative litigation model from scratch. The ECPC model is, in other words, workable and balanced. A down-side, however, is that this model allows for only very few synergies in terms of use of existing infrastructure and quick set-up, but since such synergies would also mean that the litigation structure (then based on national courts or existing EU courts) would be less effective, they have to be disregarded as enjoying the lowest priority in an assessment of conflicting aims.

6.6. Fitting infringement litigation into the EU legal order

The EU legal order sets a number of conditions for the setting-up of the ECPC, which will be discussed in this section. One fundamental issue here is the impact of ECPC jurisprudence on the EU legal order and, consequently, the delineation of parameters for interaction between the two legal systems in the international law instrument setting up the ECPC. Related to this is the ques-tion of what, if any, corresponding adaptations of competences of the EU courts are required in EU law. Preceding both discussions, however, the allocation of competences required for the setting-up of a patent litigation system between the EU and its Member States will briefly be recounted.

6.6.1. EPLA competences: An abandoned battlefieldThe Commission’s single most effective tool to bring the EPLA to a fall and to end the threat of a permanent entrenchment of the separation of European patent systems was the argument that the conclusion of the EPLA required the consent of the EU. According to the EU institutions, the Member States had already transferred certain key competences for patent litigation to the EU by the time the EPLA was drafted.232 Although most authors were essentially sup-portive of this position,233 there was initially some outcry of an alleged encroaching upon Member States’ rights.234 That opposition has, however, gradually ebbed away, as it is hardly to be disputed that the EU has in the years

232. Cf. Communication 2007, cited supra note 6, 12; Interim Legal Opinion of the Legal Service of the European Parliament, SJ-0844/06, para 67; Legal Opinion of the Legal Service of the Council, Council Document No. 8416/01, paras. 7 et seq.

233. See supra note 19. 234. Cf. Oser, “European Patent Litigation Agreement – Zulässigkeit und Zukunft einer

Streit regelung für Europa”, (2006) GRUR Int., 539, at 543 et seq.; Luginbühl, op. cit. supra note 12, 364 et seq.; Willems, “Awaiting the Community Patent – A suggestion for a flexible interim solution”, (2002) IIC, 561, at 568; Schneider, Die Patentgerichtsbarkeit in Europa – Status Quo und Reform (München, 2005), pp. 305 et seq. and 321 et seq.; Straus, “Der Beitrag Deutschlands


2001 and 2004 indeed exercised competences in the areas of jurisdiction and recognition and enforcement of judgments in civil and commercial matters235 and of procedures for the enforcement of intellectual property rights236 respec-tively. Internal market legislation is, in principle, a competence shared between the EU and Member States.237 Once the EU occupies a given field, how ever, that field becomes an exclusive competence of the Union238 and Member States laws are pre-empted.239 This pre-emptive effect also extends to the exclusion of international agreements in the field concerned.240 Insofar as the EPLA incorporated rules on jurisdiction,241 recognition of judgments242 and remedies,243 (at least) those provisions would therefore have had to be agreed to by the EU.

6.6.2. Competences for the conclusion of an ECPC agreementThe Commission envisages that the unified patent litigation system “should be created by the conclusion of an agreement in accordance with the procedure foreseen by Article 300 EC [now replaced by Art. 218 TFEU] involving the Community, its Member States and other Contracting States of the European Patent Convention”.244 This mixed agreement is necessary because none of the actors involved holds all competences necessary for its conclusion alone. It was just shown in the context of EPLA that EU participation would be required at least for the provisions relating to jurisdiction,245 recognition of judgments246

zur Entwicklung des internationalen gewerblichen Rechtsschutzes”, (2003) GRUR Int., 805, at 810 et seq.

235. Cf. Brussels I Regulation, cited supra note 21.236. Cf. Enforcement Directive, cited supra note 22. 237. Cf. Lenaerts and Van Nuffel, Constitutional Law of the European Union, 2nd ed.

( London, 2005), para 5-023.238. Cf. Case 22/70, Commission v. Council (AETR), [1971] ECR 263, para 39; Opinion

1/03, Lugano Convention, [2006] ECR I-1145, paras. 141 et seq.; Opinion 2/91, ILO Convention No. 170, [1993] ECR I-1061, para 9; Opinion 1/94, TRIPS Agreement, [1994] ECR I-5267, paras. 72 et seq.; Opinion 2/92, Third Revised Decision of the OECD on national treatment, [1995] ECR I-521, paras. 31 et seq.

239. Cf. for the US-law inspired concept of pre-emption in the context of EU law, Lenaerts and Van Nuffel, op. cit. supra note 236, para 5-023.

240. Cf. Opinion 1/03, cited supra note 237, paras. 123 and 151; Opinion 2/92, cited supra note 237, para 31; Opinion 2/91, cited supra note 237, para 8; Opinion 1/00, ECAA Agreement, [2002] ECR I-3493, paras. 5, 11 and 21; Opinion 1/91, EEA Agreement I, [1991] ECR I-6079, para 71.

241. Cf. particularly Arts. 38 and 39 EPLA, cited supra note 17. Equally Opinion 1/03, cited supra note 237, paras. 141 et seq.

242. Cf. particularly Art. 43 EPLA, cited supra note 17. Equally Opinion 1/03, cited supra note 237, para 173.

243. Cf. particularly Arts. 60 to 75 EPLA, cited supra note 17. 244. Press Release of 24 March 2009, IP/09/460.245. Cf. particularly Art. 15 Agreement 2009, cited supra note 25.246. Ibid., particularly Arts. 16 and 56.

EU Patent 105

and remedies.247 Furthermore, EU competences are affected regarding mediation,248 budgetary arrangements,249 and in respect of allowing preliminary references to the ECJ,250 to name but a few. At the same time, however, the EU cannot proceed on the ECPC without its Member States either. Member States retain a range of regulatory compe-tences required for the ECPC system, such as, prominently, the competence of transferral of patent litigation to the ECPC251 as a matter of national court organization, or the competence to exclude the enforceability of national pat-ent rights under certain circumstances252 as well as competences for any other provisions253 related to the scope of national patent rights.

6.6.3. Safeguarding the autonomy of the EU legal orderThe compatibility of the draft agreement with the EU law has been submitted to the ECJ for an opinion under the procedure of Article 300(6) EC (now Art. 218(11) TFEU).254 In principle, EU law permits courts or tribunals based on international law for the litigation of legal issues arising from an international agreement,255 subject to a number of conditions. The central condition in this regard is in fact the non-interference of the agreement and litigation system with the principles and the functioning of the EU legal order.256 The EU legal order must, in other words, be shielded against impact from the acts of non-EU organs. This shielding exercise essentially extends to two areas: first, “the essential character of the powers of the Community and its institutions as conceived in the Treaty [must] remain unaltered[.] Second, … the procedures for ensuring uniform interpretation of the rules of the [respective international agreement] and for resolving disputes [must] not have the effect of binding the Community

247. Ibid., particularly Arts. 34a to 44a. 248. Ibid., Art. 17, cited supra note 25; Directive 2008/52/EC of the European Parliament

and of the Council of 21 May 2008 on certain aspects of mediation in civil and commercial mat-ters, O.J. 2008, L 136/3.

249. Cf. particularly Art. 18 Agreement 2009, cited supra note 25.250. Ibid., Art. 48. 251. Ibid., Arts. 1 and 15. 252. Ibid., particularly Art. 14e(a) and (b). 253. Ibid., Art. 14f. 254. Press Releases of 24 March 2009, IP/09/460 and 4 Dec. 2009, IP/09/1880.255. Cf. Opinion 1/76, Draft Agreement establishing a European laying-up fund for inland

waterway vessels, [1977] ECR 741, paras. 17 et seq.; Opinion 1/91, cited supra note 239, paras. 40 and 70; Opinion 1/92, EEA-Agreement II, [1992] ECR I-2821, paras. 13 et seq.; Opinion 1/00, cited supra note 239, paras. 6 et seq.

256. Cf., for many, Opinion 2/94, Accession to the ECHR, [1996] ECR I-1759, para 20; Opinion 1/00, cited supra note 239, para 11.


and its institutions, in the exercise of their internal powers, to a particular inter-pretation of the rules of Community law referred to in that agreement.”257 Under Article 19 TEU, the exclusive responsibility for ensuring “that in the interpretation and application of the treaties the law is observed” is assigned to the Court of Justice (ECJ and General Court (the former CFI) and special-ized courts). No organ other than the Court of Justice may adjudicate on the validity or interpretation of EU law in a definitive, i.e. finally binding, man-ner.258 The powers thus assigned to the Court under the TEU (and before that the EC Treaty) may only be altered by a revision of the Treaty, but not by way of a specific international agreement.259 Insofar as EU law is applied by non-EU courts, the ECJ must therefore have the last word on any question of valid-ity and interpretation. The preliminary rulings procedure under Article 267 TFEU (ex 234 EC) is in fact an implementation of these principles in the rela-tionship between the ECJ and the Member States courts. The relationship of the ECJ and tribunals established under international law is governed by the same principles. The ECPC would not be the first litigation system set up for the execution of an international law instrument which provides for directly applicable deci-sions of an international body. Such litigation systems are, for example,260 con-tained in well-known instruments such as the European Economic Area Agreement261 and the European Convention on Human Rights,262 but also in less prominent documents like the Draft Agreement for a European laying-up fund for inland waterway vessels.263 The jurisdiction of the respective litiga-tion bodies does not just extend to the interpretation of the respective agree-ment and of the acts of organs set up under that agreement, but also to preliminary references or similar tools of interaction with national courts. Fur-thermore, most of these agreements provide for preliminary references to the ECJ from either the international court or directly from the national courts of the Member States of the agreement. Some international agreements, like, e.g., the Agreement on a European Common Aviation Area,264 also refrain from

257. Opinion 1/00, cited supra note 239, paras. 12 and 13; similarly Opinion 1/91, cited supra note 239, paras. 61 et seq.; Opinion 1/92, cited supra note 254, paras. 32 and 41.

258. Cf., e.g., Opinion 2/94, cited supra note 255, para 20; Opinion 1/00, cited supra note 239, para 11.

259. Cf. Opinion 1/92, cited supra note 254, para 32. 260. For another example and ensuing EU law concerns cf. Eilmansberger, “Bilateral invest-

ment treaties and EU law”, 46 CML Rev. (2009), 383, at 402 et seq.261. Cf. Opinion 1/91, cited supra note 239, para 32 (initial draft); Opinion 1/92, cited supra

note 254, para 4 (revised version). 262. Cf. Opinion 2/94, cited supra note 255, paras. 20 et seq. (not implemented in EU law).263. Cf. Opinion 1/76, cited supra note 254, para 17. 264. Opinion 1/00, cited supra note 239, paras. 29 et seq.

EU Patent 107

setting up a separate court and rely on the ECJ for the adjudication of disputes and validity questions and preliminary references. In short therefore, several precedents exist in EU law for systems for the lit-igation of individual rights derived from an international instrument before an international law-based tribunal or before national courts and with a proce-dural involvement of the ECJ. The tools to bring about that procedural link-age include preliminary references to the ECJ from international tribunals265 as well as from non-EU Member States courts266 and direct ECJ jurisdiction on the validity of acts of non-EU bodies established under a specific agree-ment.267 Against this diverse background, the ECPC and its particular design and its modus of interaction with EU law are far from being new creatures which EU law has not already seen. The principles established for these prec-edents provide sufficient guidance for assessing the ECPC system. As regards the compatibility of the ECPC with the EU legal order, first of all and importantly, the agreed 2009 Proposal268 establishes a relationship of subordination of the ECPC vis-à-vis the ECJ for all matters relating to EU law.269 The ECPC is compelled to refer EU law questions to the ECJ, the answers handed back to the ECPC are binding.270 This procedure appears suf-ficient to provide for the observance of EU law principles and the safeguard-ing of EU law-based rights for individuals by the ECPC and thus, in sum, for the non-interference and concurrence of ECPC jurisprudence with the EU legal order. Although the text of the 2009 Proposal does not explicitly state so, it is to be assumed that the wording “question of interpretation of Community law” is (or rather: must be) meant to also include questions of validity of EU acts. In this regard therefore, the Proposal’s wording perhaps needs some clarifica-tion still. As was pointed out above, this preliminary references obligation will extend to any instance of application and interpretation of a future EU patent regulation. It was, however, likewise pointed out that the case law limiting the obligation to request preliminary references under the similar procedure of Article 267 TFEU (ex 234 EC) will likely apply to the ECPC also, thus

265. Cf. Opinion 1/76, cited supra note 254, para 17. 266. Cf. Opinion 1/00, cited supra note 239, paras. 30 and 33; Opinion 1/91, cited supra note

239, para 56 (initial draft); Opinion 1/92, cited supra note 254, para 11 (revised version). 267. Cf. Opinion 1/00, cited supra note 239, para 44; Opinion 1/92, cited supra note 254,

para 9 (revised version).268. Cf. Art. 48 Agreement 2009, cited supra note 25..269. Cf. on this point Opinion 1/92, cited supra note 254, paras. 29 and 35 (revised version),

missing (thus incompatible) in the earlier draft, cf. Opinion 1/91, cited supra note 239, para 35; similarly Opinion 2/94, cited supra note 255, para 34.

270. Cf. on this point Opinion 1/00, cited supra note 239, paras. 34 et seq.; Opinion 1/91, cited supra note 239, paras. 39 and 61; Opinion 1/92, cited supra note 254, para 15 (revised ver-sion).


requiring ECPC references only in relation to new or non-obvious questions of EU patent law.271 In assessing the need for preliminary references, the ECPC will have to screen and assess the existing case law of the ECJ for general prin-ciples or analogies relevant to the patent cases before it. The 2009 Proposal does not yet include an explicit statement regarding relevance of ECJ case law for ECPC decision-making. For the sake of clarity and to reinforce the com-patibility of the 2009 Proposal with EU law,272 such wording should be included. A related question concerns the problem of de facto divergence of patent jurisprudence for EU patents and European patents. In principle, the ECPC is not obliged (nor competent) to refer questions relating to the application of the EPC to the ECJ for preliminary rulings. If, however, the ECPC developed a strand of jurisprudence for a specific question of patent law under the EPC, the ECJ will, if subsequently seized with that question under the EU patent regulation, be faced with the dilemma of either having to take over that juris-prudence or to accept a rift in the interpretation and application of patent laws in Europe. The same problem occurs for the ECPC, where it does not wish to take over jurisprudence developed by the ECJ for the EU patent and is not obliged to apply it in relation to European patents. If the ECJ or the ECPC choose to diverge from each other’s case law under the present system, the rift is permanent insofar as there is no reconciliation mechanism in place. As the ECJ has previously pointed out, however, in relation to the EEA Agreement, there is an interest in a harmonious application of bodies of law with parallel or similar norms (there: the EEA Agreement for EFTA States vis-à-vis the EEA Agreement and EC Treaty for EU States) even where those bodies of law are formally independent of one another (there: the EC Treaty and the EEA Agreement).273 The ECJ held that a first draft of that Agreement that did not provide for a reconciliation mechanism for diverging interpretations of those parallel bodies of law was not compatible with EC law.274 In particular, the ECJ found that in the absence of such a mechanism, a mere obligation for the courts involved (in the first draft: the EEA court, which was subsequently not established) to take account of each other’s case law was insufficient in that regard.275 By contrast, the revised draft for the EEA Agreement provided for a mixed committee charged with monitoring the bodies of case law of the two courts. If a divergence occurs, the mixed committee attempts reconciliation and takes a final decision on the interpretation of the norm at issue. Importantly,

271. Cf. the references cited supra in notes 191 to 193. 272. Cf., for many, Opinion 1/00, cited supra note 239, paras. 37 and 39; Opinion 1/92, cited

supra note 254, paras. 21 et seq. (revised version).273. Cf. Opinion 1/91, cited supra note 239, paras. 22 et seq.; Opinion 1/92, cited supra

note 254, paras. 17 et seq. (revised version). 274. Cf. Opinion 1/91, cited supra note 239, paras. 28 et seq.275. Cf. Ibid., paras. 24 et seq.

EU Patent 109

however, the interpretative freedom of the mixed committee is limited insofar as decisions must not run counter to existing case law of the ECJ. The ECJ held that this mechanism, provided that the condition of compatibility with pre-existing case law is binding, is compatible with EU law.276 For the ECPC, the 2009 Proposal neither stipulates a requirement of mutual respect for the bodies of case law already developed, nor does it provide for a reconciliation mechanism. This constitutes an incompatibility with EU law, which may, however, be easily remedied: the advisory committee277 foreseen under the 2009 Proposal to partake in the procedure for appointment of judges could also take on the task of reconciliation of case law similarly to what is provided for in the EEA Agreement. The advantage of assigning that task to the independent advisory committee instead of the political mixed commit-tee278 simply lies in the expertise of that body. In view of the preliminary reference obligation for the appeals court, the fact that the divisions of first instance are neither obliged nor competent to refer EU law questions to the ECJ is generally harmless. This assessment of compatibility includes the competence of first instance divisions to revoke EU patents granted by EPO. Under the Foto-Frost doctrine, national courts of any instance must generally refer questions of validity to the ECJ and may not invalidate EU law acts independently.279 As will be shown immediately below, the fact that the validity of the patent will be assessed after the conclusion of the EPA procedure, i.e. once the patent has been granted, means that decisions to grant EU patents are principally to be regarded as acts of EU law in spite of the fact that EPO is not an EU institution. However, the explicit authoriza-tion for first instance divisions to invalidate such acts likely frees them from observance of the Foto-Frost obligation. By the same token, national courts dealing with questions of invalidity of EU trademarks and designs are not bound by the Foto-Frost doctrine and need not request a preliminary ruling before declaring invalidity, apparently because of the respective authorizations in the legal bases governing their competences.280

Finally, jurisdiction of the ECJ is excluded in relation to acts of the EPO, since those acts cannot be reviewed by the ECPC and there is no other mech-anism in the 2009 Proposal allowing for ECJ review. In principle, however, all acts of EU institutions must be subject to review by an EU court by virtue of Article 19 TEU (previously, Art. 220 EC) and the general principle of rule of law. The ECJ has long emphasized that the EU is “a community based on

276. Cf. Opinion 1/92, cited supra note 254, paras. 20 et seq.277. Cf. Art. 57b Agreement 2009, cited supra note 25.278. Ibid., Art. 57, cited supra note 25. 279. Cf. Case 314/85, Foto-Frost, [1987] ECR 1499, paras. 17 et seq.280. Cf. Arts. 96 and 100 CTMR; Arts. 81 and 84 CDR; both cited supra note 29.


the rule of law, inasmuch as neither [the] Member States nor [the] institutions can avoid a review of the question whether the measures adopted by them are in conformity with the basic constitutional charter, the Treaty. … Natural and legal persons are thus protected against the application to them of general mea-sures which they cannot contest[.] Where the Community institutions are responsible for the administrative implementation of such measures, natural or legal persons may bring a direct action before the Court[.] Where imple-mentation is a matter for the national authorities, such persons may plead the invalidity … before the national courts and cause the latter to request the Court of Justice for a preliminary ruling.”281 The Treaties therefore establish a com-plete system of remedies, which requires any act of law adopted within that system to be reviewed by a court under the final supervision of the ECJ.282 The principles of rule of law and judicial review for administrative mea-sures do not only apply to the EU institutions in the proper sense, but must also be deemed to set the limits of action for a transfer of administrative power in the interpretation and application of an EU law act by a non-EU body – in our case therefore the interpretation and application of an EU patent Regula-tion by EPO. As the ECJ has repeatedly emphasized, the fundamental role of safeguarding the rule of law within the EU, which includes judicial control over administrative acts applying EU law and having an effect within the EU territory, cannot be encroached upon in an international agreement concluded under Articles 300 or 310 EC (now replaced by Arts. 218 and 217 TFEU).283 Those provisions do “not provide any basis for setting up under an interna-tional agreement a system … which conflicts with … the very foundations of the Community.” An agreement disregarding the principles of rule of law and of exclusive judicial review of questions of interpretation of EU law by the ECJ would thus fall foul of the very foundations of the EU Treaty. It may even be said that the Council and Commission would act ultra vires by concluding such an agreement insofar as they would try to endow an international with body rights or powers (i.e. the power to apply EU law without the control of an EU court) that they themselves do not possess. Consequently, the lack of review of EPO decisions (i.e. the decisions of EPO’s boards of appeal) on EU patents is incompatible with EU law insofar as those decisions belong to the sphere of EU law. In this regard, it could be argued that EPO is not an EU institution and that its decisions therefore stand

281. Case 294/83, Les Verts, [1986] ECR 1339, para 23; consistent jurisprudence.282. Cf., e.g., Case C-131/03 P, Reynolds, [2006] ECR I-7795, para 80; Case C-461/03,

Gaston Schul, [2005] ECR I-10513, para 22; Case C-50/00 P, Unión de Pequeños Agricultores, [2002] ECR I-6677, para 40; Opinion 1/91, cited supra note 239 para 21. Also Lenaerts and Arts, op. cit. supra note 193, para 4-002.

283. Cf. in this sense Opinion 1/91, cited supra note 239, para 45.

EU Patent 111

outside the framework of EU law and consequently do not require or permit ECJ scrutiny. This argument is, however, flawed both as regards decisions taken throughout the examination procedure up to its conclusion and EPO deci-sions taken after the patent was granted, i.e. in opposition proceedings. First of all, even if the EPO is not an EU institution, the fact that it applies and inter-prets EU law (the future patent regulation) when deciding applications must be regarded sufficient to require an ECJ review of those acts of application and interpretation. If this was not the case, the application and interpretation practice of EPO (and its boards of appeal) was likely to conflict with or com-promise any principles established by the case law of the ECJ, relevant for the application of the patent regulation (e.g. under the ECPC’s preliminary rulings procedure), thereby compromising the homogeneity of the application of a future EU patent regulation in the EU territory.284 Apart from this threat to the uniformity of application of substantive EU patent law, the question of legal-ity of the general measure on which EPO decisions will be based, i.e. of the EU patent regulation, would likewise not be subject to review. The ECJ has, however, made it clear that individuals must always be in a position to chal-lenge the legality of a general EU law measure underlying specific applica-tion decisions, irrespective of the procedure where this general measure is applied.285 In the proceedings before EPO, however, the question of legality of the EU patent regulation could not be raised because EPO (or its boards of appeal) are not competent to review that question while, at the same time, no alternative remedy to the ECJ is available. In sum therefore, a substantive approach is called for regarding EPO decisions situated within or outside the EU legal order, rather than a formal one. This substantive approach is also taken vis-à-vis national courts, where the application and interpretation of EU law is subject to review by the ECJ, although they are not formally EU organs. Where jurisdiction over all provisional, final and post-grant acts of EPO in relation to EU patents is transferred to the ECPC, EPO’s boards of appeal may remain competent for Munich patents or, with a view to allowing for as uni-form a review practice as possible, should be abolished and their tasks trans-ferred to the ECPC.

6.6.4. Flanking adaptations of EU lawFinally, the question arises whether the setting up of the ECPC system requires any flanking adaptations of EU law. Technically, the ECPC system presupposes the prior or parallel adoption of the EU patent regulation, including the still missing separate act on translation requirements, although it is possible and in fact advisable to contemplate a reverse chronology also – as will be done below.

284. Cf. Opinion 1/91, cited supra note 239, paras. 39 et seq. and 71. 285. Cf. Les Verts, cited supra note 280, para 23.


The EU patent regulation would in the past have had to be based on Article 308 EC, the legal basis chosen for other EU IP instruments, like the CTMR and the CDR.286 With entry into force of the Treaty of Lisbon, an explicit legal basis is contained in the new Article 118 TFEU.287 That Article lowers the threshold for adoption of the substantive part of the EU patent regulation to qualified majority (while Art. 308 EC required unanimity), but the unanimity requirement persists regarding language requirements for the EU patent. This may be the reason why the Council’s 2009 political agreement on the patent package does not extend to language requirements. As regards the setting up of the ECPC system as such, no further adapta-tions seem necessary. In particular, the system established under the 2009 Pro-posal does not presuppose the adoption of an act by the Council for conferral of jurisdiction for patent disputes on the ECJ, as is envisaged by Article 262 TFEU (ex 229a EC). After all, the ECJ would decide on preliminary references relating to patent disputes, but not rule in the disputes as such. The current sys-tem of legal remedies under the EC Treaty is sufficient to deal with this task. Likewise, since the ECPC would not be a panel of the CFI in the meaning of Article 257 TFEU (ex 225a EC), no decision pursuant to that Article is required either. In sum therefore, other than the conclusion of the agreement establish-ing the ECPC (and, as will be discussed below, an accession of the EU to the EPC), the setting-up of the ECPC system requires no prior adaptations in EU law.

6.6.5. A flexible roadmap toward the EU patent regulationThe next steps on the road to the implementation of the litigation system are the successful completion of the legislative procedure involving the Parliament under Article 294 TFEU (ex 251 EC) and the handing down of the already requested opinion by the ECJ on the compatibility and possible remaining incompatibilities of the ECPC structure with EU law.288 Only when these hur-dles are taken and any required amendments are likewise agreed and included, is the aforementioned mixed agreement between the EU and the participating EU and EPO Member states to be drawn up, and will subsequently need to undergo ratification. In parallel, negotiations for an accession of the EU to the EPC could be pre-pared and an accession agreement signed once the ECJ has given a positive opinion on the ECPC structure.289 Without accession, the EPO can, evidently,

286. Both cited supra in note 29.287. See supra note 223. 288. Cf. Press Releases of 24 March 2009, IP/09/460, and of 4 Dec. 2009, IP/09/1880;

EuZW editorial news, “Schaffung eines einheitlichen Patentgerichtssystems”, (2009) EuZW, 555, at 555.

289. Cf. also Press Release of 24 March 2009, IP/09/460.

EU Patent 113

not be instigated for the granting of EU patents as is envisaged. In that case, a separate office for the granting of EU patents would have to be set up, similar to or under the roof of the OHIM for trademarks and designs. Since a harmo-nious development of patent practice in Europe is desirable, such an option should at best form the backup plan if accession does not succeed. After all, an important obstacle to accession is the fact that the EPC still was not amended to allow international organizations such as the EU to join.290

Most importantly, however, the entry into force of an EU patent regulation is still impeded291 by the persisting lack of agreement in the Council over trans-lation requirements as well as some other formal details.292 As concerns lan-guage requirements, Article 118 TFEU as the relevant legal base for that regulation has not lowered the voting thresholds as compared to the formerly applicable Article 308 EC and still requires unanimity in the Council in this specific respect. In theory, the EU patent could be realized by way of enhanced cooperation if some Member States wish not to participate.293 Article 20(2) TEU (replacing Art. 11 EC) stipulates that a qualified majority and nine coop-erating Member States would be enough in this regard. However, from its his-toric background enhanced cooperation was clearly intended to be a means of last resort only.294 Furthermore and importantly, the main advantages of the EU patent lie in the creation of a uniform right covering the entire internal market and in cost-benefits compared to European patents which are regis-tered for a minimum number of States, starting at roughly thirteen.295 Enhanced cooperation therefore does not necessarily offer the key benefits required to render the EU patent attractive and competitive. Since the coming into existence of an EU patent may thus still take some time given this procedural and factual background, it should be ensured that unified litigation for European patents is not held up by that process. What is needed therefore is a flexibility clause in the ECPC agreement, allowing the ECPC to become operative in relation to European patents immediately and to include EU patents whenever the EU patent regulation is passed. EPO States which are not also Members of the EU could hardly be required to join the ECPC system and would thus probably retain the option to stick with their cur-rent litigation arrangements. Where some States choose to remain outside the

290. Cf. Art. 166 EPC; also Ullrich, op. cit. supra note 10, 460.291. Cf. Press Release of 4 Dec. 2009, IP/09/1880.292. Cf. also Press Releases of 24 March 2009, IP/09/460, and of 4 Dec. 2009, IP/09/1880.293. Cf. in this regard Ullrich, op. cit. supra note 10, 489 et seq.; Tilmann (2006), op. cit.

supra note 12, 825; Krieger (1987), op. cit. supra note 40, 732 et seq.; Willems, “The EPLA – Trojan Horse or Gift of The Gods?” in Kur, Luginbühl and Waage, op. cit. supra note 72, p. 325, at pp. 336 et seq.

294. Cf. Art. 43a TEU. The phrase has disappeared in the TFEU.295. Cf. Communication 2007, cited supra note 6, 3 and 22 et seq.


system, the existing structure for internal appeals within EPO would have to be retained in respect of such States for as long as necessary. Regarding EU Member States, by contrast, no such option exists: Articles 20(1) TEU and 329(1) TFEU (and before them Art. 43(1)(d) TEU) excludes enhanced coop-eration over matters resorting to an exclusive competence of the EU. As was shown earlier above, several aspects of the litigation system have, however, become exclusive competences of the EU because of legislation enacted in the fields of jurisdiction and court remedies. Enhanced cooperation is therefore, unlike for the substantive provisions on the EU patent, not an option with regard to the ECPC. The agreed 2009 Proposal, by contrast, is still vague on the issue of how the ECPC system should enter into operation.296 Since, how-ever, the risk of a permanent rift in patent laws and patent litigation in Europe is removed as soon as an agreement on the setting up of a unified ECPC is concluded and ratified, there is no justification for further delay with regard to litigation for European patents. This solution would also have the advantage of the ECPC being able to establish a set of patent case law already before the ECJ starts to deal with preliminary rulings on patent cases, so that the ECJ would be able to draw upon some of the principles thus elaborated where necessary.

7. Conclusions: Agree on the framework, leave the details for later

While the integration of national markets has come a long way since the EU’s early days and the internal market has been completed in many areas,297 European patent law is still in its infancy: it still relies on EPO-type intergov-ernmentalism298 and on a separation of national legal systems under presiding logic of the lex loci protectionis, thereby fragmenting and hampering the eco-nomic exploitation of patent rights by rightholders as well as access to patents for third parties. The intergovernmental method and the nationally fragmented approach to protection are unfit to deliver appropriate results in a European market that is dynamic, constantly evolving and adapting to new realities. The field of patents has thus so far missed out on the EU’s last fifty years of evolu-tion and achievement and is stuck in something like a pre-Van Gend en Loos and Costa299 loop.

296. Cf. Art. 60 Agreement 2009, cited supra note 25.297. Cf., e.g., Communication A single market for citizens, COM(2007) 60 final, pt. 1.298. Even if the EPO system may show some quasi-supranational streaks (cf. Ullrich, op. cit.

supra note 10, 437 and the references cited there), that system can in no way be likened to the sophisticated supranational regime of the so-called Community method.

299. The judgments in Case 26/62, Van Gend & Loos, [1963] ECR 3, 24 et seq., and Case 6/64, Costa / ENEL, [1964] ECR 1253, 1269, lie at the foundation of the EU’s suprana-tional legal order.

EU Patent 115

It is therefore time to break the spell of the roaring sixties on patents and innovation in Europe. The odds for success are also better than ever: the Coun-cil has reached an initial political agreement on the setting up of a patent system, and there is no fundamental opposition among the majority of other stakeholders against the plan to create a sui generis patent right and flanking litigation system for the internal market. Looming protectionist sentiments against the patent as such have long disappeared and the remaining points of disagreement are essentially all over details of the formal framework rather than over substance.300 Indeed, economic analysis today proves significant eco-nomic benefits from a single patent and litigation system for rightholders and users alike. Finally, in response to the urgent need for better litigation for EPO patents which cannot go ahead without EU consent, the Commission and Coun-cil have placed a non-paternalistic and well-calibrated proposal for an inte-grated EU/EPO patent litigation framework on the negotiation table that is clearly the most innovative and flexible model discussed since the 1950s and 1960s.301

The deal for the EU patent is now a package proposal, combining the rules on protection, on litigation and on the integration of systems in an “all or noth-ing” offer. But the comprehensive character of the combined EU patent Reg-ulation/integrated patent litigation-package naturally raises the threshold for reaching agreement. It may, in other words, be doomed because the size of the deal is too big to pass. Failure of to pass would administer a significant blow to the EU patent plans and probably take the drive out of the debate altogether. The solution here might be to opt for a flexible strategy that requires only agreement on the framework of the package and leaves agreement over detail for later. This means that we should strive to ensure the operability of the inte-grated patent judiciary as quickly as possible, without waiting for agreement on any missing rules for the EU patent, such as languages. In this way, the framework for an integration of the EU and Munich systems as well as for the litigation of patents would be agreed, while allowing the EU and Member States more time to negotiate the details of patent protection and court proce-dures within that system. The integrated judiciary would adjudicate Munich patents alone in the beginning, but subsequently incorporate EU patent litiga-tion once the regulation enters into force. The pressing issue of ineffectiveness of patent litigation in Europe would thereby be relieved immediately, yet the EU’s own patent plans not (unlike under EPLA) threatened by this increase in effectiveness.

300. Equally Schneider, op. cit. supra note 10, p. 646. 301. Appreciative therefore Jaeger, Hilty, Drexl and Ullrich, op. cit. supra note 26, paras. 30

et seq.

common market law review - wydzia‚ prawa i administracji uw

Documents