Reset and Go: The Unitary Patent System Post-Brexit

Abstract

Patent law in Europe is characterized by a historic rivalry between EU and non-EU patent systems. The EU for decades could not establish a working, attractive and balanced system of its own. After the failure of its well-tailored 2009 model, the Commission was determined to push ahead with the patent plans even at the cost of compromise that severely damaged the functionality of the patent system. The result was the 2012 Unitary Patent Package, which has since been cleared twice by the CJEU in spite of severe doubts concerning EU law compatibility. Just as the race seemed to near finish line, the June 2016 Brexit referendum put a spoke in the EU’s wheel. Against the backdrop of a brief review of the systemic rivalry, this paper recounts and assesses the CJEU’s recent case law on the legality of the UP Package, the implications of the Brexit vote and the prospects, if any, for the unitary patent post-Brexit.

1 Introduction

The history of patent cooperation in the EU is one of decade-long failures.¹ While there has always been agreement that the patent would be desirable in principle, the details of what it should look like and how cross-border patent litigation should be organized deeply divided EU (or at the time EEC) States.

1.1 Systemic Competition

Within the EU framework, early plans for a patent for the internal market stem from the 1950s.² While the EU had for decades been unsuccessful in passing legislation on a patent right and its cross-border litigation, the non-EU system of the European Patent Convention (EPC) filled in for that failure from 1973 onward (operational as of 1977). The EU and EPC legal systems were over the years competing as rivals: The more successful the EPC became over the years, the more it posed a threat to the realization of patent legislation within the EU framework. In 2003 that threat peaked with plans to complement the EPC framework by establishing a supranational court for cross-border patent litigation under the so-called European Patent Litigation Agreement (EPLA).³

1.2 Models 1975 to EPLA

The main point of disagreement hindering the establishment of an EU-own patent system used to be the shape of the litigation model. By contrast, the substantive patent right plans were hardly disputed over time⁴ and consistently resembled the sui generis type of right that was designed for, among others, EU trade marks and design rights.⁵ All of the historic models relied on the EPO for the grant and administration of the patent, with most models leaving the internal administrative appeals to the EPO (with the exception of the First and Second CPC, CJEU and CoPAC respectively).

Each time, court models were developed alongside the substantive patent proposal to ensure effective cross-border litigation. Other than the substantive and administrative proposals, the court proposals were very diverse and covered virtually any imaginable model.

The First Community Patent Convention (CPC)⁶ from 1975 proposed a combination of national courts under the preliminary reference control of the CJEU. This would, accordingly, have been a decentralized system similar to the one established for EU trade mark and design rights. Administrative appeals would have gone directly to the CJEU.

The 1989 Second CPC,⁷ by contrast, proposed a court of appeals based on international law only, the so-called Community Patent Appeal Court (COPAC). It foresaw centralized administrative appeals and a competence to take preliminary rulings from national courts.

The Patent Regulation proposal from 2000⁸ again devised a court based on international law, the Community IP Court (CIPC), however for direct litigation in first and last instance. It would have dealt with infringement and invalidity issues, had full cognition and was fully centralized. The only competence it did not hold was for administrative appeals, which remained within the EPO.

Next, the 2004 Patent Regulation proposal⁹ sought to put the newly (by the Nice Treaty) created legal basis for specialized EU courts below the level of the General Court (GC) into practice. It accordingly proposed to create an EU law-, Art. 257 TFEU-based specialized court, the Community Patent Court (CPC). It would have dealt with infringement and invalidity litigation and would have had full cognition in that regard. As an EU court, it would have been fully centralized. Administrative appeals, again, were kept within the EPO.

Between 2002 and 2004, the European Parliament designed an EP alternative proposal in the form of an EU law-based court of (direct) appeals ex national courts.¹⁰ That court, however, never came into being, mainly due to Member States’ qualms over the compatibility of the direct appeals mechanism with several constitutional systems (e.g. the French constitution).

In 2003 and (revised) 2005, the aforementioned EPLA was drawn up under the auspices of the EPO, i.e. as the first non-EU litigation model. It proposed creation of an international law-based, two-instance court, the so-called European Patent Judiciary (EPJ). The first instance of that court would have been decentralized. It would have been competent for infringement and invalidity proceedings and would have had full cognition. Administrative appeals, as with the EU models, would have remained within the EPO.

1.3 EEUPC Model and CJEU Opinion 1/09

EPLA sounded alarm bells with the Commission: An effective cross-border litigation branch outside the EU framework would have removed pressure from the need for EU harmonization. The Commission accordingly acted against EPLA, alleging an infringement of external competences in fields occupied by internal legislation.¹¹ In turn, it stepped up work on a compromise model.

In 2009, the Commission had fostered compromise among those Member States in favor of deepening non-EU integration under the EPC and those Member States that wished to bring patent integration under the mantle of the EU. It proposed to partially integrate the EPC system into the EU. However, one important point of disagreement had been deferred for later decision, namely the patent’s language regime: Keeping all languages would mean a very costly system, but which languages were to be dropped?

The 2009 proposal devised the so-called European and EU Patent Court (EEUPC).¹² It was an international law-based two-instance court, decentralized in the first instance. The EEUPC’s revolutionary feature was that unlike any model before, it would have extended the litigation system to both EU and EPC patents. The EEUPC would have dealt with both, thus responding to the EPC’s wish to make effective the litigation leg that had been the object of EPLA. The basis for this combined court would have been a mixed agreement between the EU, the EPO States and EU Member States. The EEUPC’s combined jurisdiction for EU and EPO patents would have replaced the respective jurisdiction of the national courts for infringement and invalidity proceedings. For EU patents in addition, there was a procedure for obtaining compulsory licenses. Integration of the EEUPC into the EU legal order would have been limited to a link to the CJEU via its preliminary rulings competence for EU patents only. Similarly, institutional links with the EPO were limited insofar as the EEUPC had no jurisdiction to review EPO decisions. The underlying substantive patent regime for the EEUPC model was, as with all the models before, a complete EU sui generis type of right.

The EEUPC was, however, declared incompatible with EU law in 2011 on grounds that it infringed central principles of the EU legal order.¹³ In particular, the CJEU held that the direct application of EU secondary law by the EEUPC jeopardized the autonomy of EU law and the completeness of the system of remedies and interfered with the cooperation model established by Art 267 TFEU. Indeed, whereas the EEUPC would have applied the EU Patent Regulation directly, other international courts typically apply only their own agreement, i.e. international law without (additional) EU law. Under a strict reading of Art. 19 TEU and Art. 267 TFEU, only EU courts and national courts may apply EU law directly. There is no third way. Some minor points additionally tainted the EEUPC model, such as the lack of Francovich-type liability and a potentially discriminatory language regime.

2 CJEU Authorization of Enhanced Cooperation and Its Wider Implications

Opinion 1/09 was the first major blow to the EU’s modern patent plans, but it wasn’t the last: A series of subsequent obstacles followed, prominently the Spanish language dispute and recently the Brexit vote. Each of these obstacles that added new compromise and additional dysfunctionalities to the original plans of the Commission.

2.1 Timeline 2009–2012

The 2009 EEUPC proposal had been celebrated as a significant breakthrough by the European Commission.¹⁴ As mentioned, it did not, however, cover the issue of languages (procedures and translations post-grant). The language regime is subject to unanimity under Art. 118 TFEU, whereas the rest of the patent rules require qualified majority voting only.

Following Opinion 1/09, the Commission was under pressure to come up with a quick follow-up to the (incompatible) EEUPC model in order not to lose the momentum gained.¹⁵ But the decision on a follow-up model was held up by the unresolved language dispute. Spain and Italy insisted on the five-language language regime (EN, FR, GER, IT, ESP) of the EU trade mark office (EUIPO), whereas the remaining States favored a reduced three-language regime (EN, FR, GER) for cost and simplicity reasons. Since that dispute could not be quickly overcome, the Commission and Council decided to circumvent the blockage by the objecting States Spain and Italy: In 2011, on a proposal from the Commission and after consent of the European Parliament, the Council authorized enhanced patent cooperation over the patent.¹⁶

In 2012, the implementing acts for enhanced patent cooperation followed. They consisted of the Unitary Patent (UP) Regulation 1257/2012¹⁷ laying down the substantive patent right, Language Regulation 1260/2012¹⁸ laying down the disputed language regime for translations and, finally, an international agreement on the creation of a so-called Unified Patent Court (UPC).¹⁹

2.2 Decision 2011/167/EU

The 2011 Decision authorizing enhanced cooperation is brief and basically consists of two main parts: One part explains that insurmountable differences existed in the Council in reaching agreement on the language issues. The second major part explains why the conditions for enhanced cooperation are fulfilled: “On 30 June 2010, the Commission adopted a proposal for a Council Regulation on the translation arrangements … At the Council meeting on 10 November 2010, it was recorded that there was no unanimity to go ahead … It was confirmed on 10 December 2010 that insurmountable difficulties existed, making unanimity impossible at the time and in the foreseeable future. In these circumstances, 12 Member States … addressed requests to the Commission … that they wished to establish enhanced cooperation”.²⁰ “The conditions laid down in Art. 20 TEU and in Arts. 326 and 329 TFEU are fulfilled.”²¹ Those conditions and their fulfillment (according to the Decision) are that the cooperation falls within an area covered by the Treaties, but that no regular decision could be reached within a reasonable period. Furthermore, the cooperation advances the internal market by way of the creation of a patent, while at the same time not undermining that market or EU cohesion. Another condition deemed fulfilled is that the cooperation falls within the internal market rules, which is a shared competence. Finally, it was deemed to sufficiently respect the rights of non-participating States, because their integrational status quo and their relevant acquis remained untouched.

Striking is the short time delay between the first language proposal tabled by the Commission and the Decision to authorize enhanced cooperation: It was roughly eight months. This illustrates the pressure the Commission was under to deliver evidence of progress of the patent plans in spite of Opinion 1/09. Nobody was going to wait for the Spaniards to cave in or to offer a compromise proposal – in spite of the fact that the language issue is protected by a procedure of unanimous voting in the Treaty precisely because of its sensitive character.

2.3 CJEU Review of Decision 2011/167/EU

The actions for annulment brought by Spain and Italy against the Authorization Decision were dismissed by the CJEU in 2013.²² The Court dealt with a few crucial points there. First, the Court answered the question whether the cooperation concerned a shared competence, which is a precondition for enhanced cooperation. Article 2 TFEU defines shared competences as a situation where “the Union and the Member States may legislate” and where, accordingly, the “Member States shall exercise their competence to the extent that the Union has not exercised its competence.” It is hard to see how “measures … to provide uniform protection of intellectual property rights throughout the Union”, as the wording of Art. 118 TFEU goes, could logically be anything but an exclusive Union competence. However, the Court was unconcerned: It took a purely formalistic approach by looking only at the Chapter that Art. 118 TFEU is part of, the internal market, and that the internal market, according to Art. 4(2) TFEU, is a shared competence.²³ There is much to be said against this, from the merely declaratory nature of the competence catalogue²⁴ to the Court’s disregard for the telos and history of the norm, i.e. a truncation of interpretative methods. What it boils down to, however, is a marked friendliness on the part of the Court to let the cooperation go ahead.

This observation is also true for other aspects of that judgment – prominently the Court’s approach to the term of last resort in Art. 20 TEU. That norm allows enhanced cooperation only “as a last resort” to overcome permanent blockages of integration initiatives. As was shown, the Council had only deliberated the issue for about eight months, which indeed begs the question whether that is already a permanent blockage – or rather the normal course of negotiations observed for many pieces of legislation eventually passed.

That judgment accords the Council utmost discretion to decide on the notion of last resort. Only a reduced judicial review for misuse of powers is carried out: “The Council … is best placed to determine whether the Member States have demonstrated any willingness to compromise [, as long as] the Council has carefully and impartially examined … this point and … adequate reasons have been given”.²⁵ Under that standard, judicial review is essentially reduced to a mere requirement to state reasons. It opens the door wide for enhanced cooperation whenever there is no hope for compromise in the foreseeable future. At the same time, the judgment highlights the political nature of the term last resort. A majority of Member States may use the threat as bargaining momentum in Council negotiations.

The assessment of other effects of the cooperation was essentially deferred to the implementation phase. Examples are the openness to all Member States for later participation (that is, non-preclusion), the absence of negative effects on the functioning of the internal market (that is, fragmentation, discrimination) and the absence of negative effects on cohesion in the EU. Whether those effects surface indeed depends on how the cooperation is conducted, so deferral of those issues is generally sensible. However, some effects already visible from the reasons or aims of the cooperation may not be remediable. The language regime that will be used for the cooperation is an archetypical example of such structural preclusion, which disallows later participation on the same terms.

The Court‘s position ignores this. Missing is a de minimis test for prejudice of interests: Some initial prejudice to interests must be accepted by non-participants, but not where the prejudice threatens to outweigh the benefits of subsequent participation.

2.4 Interim Summary

The judgment on enhanced patent cooperation shows that differentiation has reached the core of the internal market. Patent protection directly relates to the free movement of goods²⁶ and to the cross-border business opportunities of undertakings. The CJEU made it clear in that respect that the Council enjoys very wide political discretion, regarding both the need for enhanced cooperation (i.e. the question of “last resort”) and the best ways to implement the desired cooperation. This significantly strengthens the bargaining asset character of enhanced cooperation beyond the sphere of patents alone.

At the same time, however, the CJEU sets insufficient safeguards against the potential (or in this case, in view of the circumvention of Art. 118 TFEU’s unanimity requirement arguably, actual) misuse of the instrument of enhanced cooperation. In this regard, the judgment fails to establish clear limits and criteria to protect minority Member States’ negotiating interests. In particular, no de minimis test for features de facto forestalling future participation on acceptable terms (e.g. languages) is established.

3 CJEU Assessment of the 2012 Package

Two years after the judgment on enhanced cooperation, the CJEU dealt with the implementation details in two more judgments.²⁷ They each deal with one of the substantive legs of the 2012 patent package (the substantive patent and language regime respectively). The third, formal leg of the 2012 package, the Agreement on the establishment of the UPC, was not to date subject to CJEU review.

3.1 Package Overview

The 2012 package is the follow-up to the marred 2009 proposal. It consists of three legally distinct legs: Two Regulations²⁸ lay down the substantive and translation aspects of the so-called UP, while a flanking international law-based agreement undertakes the establishment of a Unified Patent Court (UPC).²⁹

The UP Regulation 1257/2012 incorporates a unique regulatory method hitherto unprecedented in EU law and all preceding patent proposals, in particular the 2009 patent proposal. The UP is defined as a regular EPC patent, however, granted with the same set of claims for the States participating in the enhanced cooperation. In terms of substance, it is a “stub patent“: The Regulation stipulates next to no patent rules in EU law. Not even the minima for determination of the type of right are spelled out, e.g. the requirements of protection, definitions of the scope of protection or property rules. In fact, the patent created is an empty shell because the Regulation does not determine its shape at all. Instead, Regulation 2157/2012 undertakes an extensive reliance on third law. Elsewhere,³⁰ this has been referred to as a “Hieronymus Bosch-approach”, because of the assembly of the creature from the body parts of other legal regimes.

Notably, the procedure and conditions for grant of the patent follow (and are directly based on) the EPC, i.e. international law. By contrast, the determination of the scope of protection flows from the UPC Agreement (UPCA), i.e. an act of international law transposed into the national laws of the Member States. There, the scope is laid down in a negative manner in the form of rights and remedies of the patent holder before the UPC. Finally, any property aspects of the patent (e.g. transfer, mortgaging and use as security, etc.) follow the various Member States’ laws where the patent applicant was domiciled at the time of the application or (subsidiarily) German law.

Based on these features, the UP is characterized by a high degree of legal fragmentation, yet a factual sameness of the right in all participating States. In other words, it is likely that the novel and unprecedented exercise to devise the unitary character of an EU right via extensive references to third law will factually work out. However, what this method implies for the principle of primacy of EU law, once the CJEU is called upon to interpret the Regulation’s rules, remains yet to be seen. It is unlikely that national or international law may make binding stipulations for the interpretation of the Regulation by the Court. This would be an inversion of the primacy rule. However, it should be noted that unlike for the unitary character, the Regulation does not also expressly stipulate the autonomous (sui generis) character of the UP in terms of its independence from those third law rules. In addition, the institutional links between the EU and the EPO, which will take on the task of granting the patent, remain unclear: First, the EU is not an EPO member; and second, most decisions of the EPO, in particular those concerning patentability, are not subject to a remedy before the EU courts.

As regards the flanking litigation leg for the UP and (combined) EPC patents under the UPCA, participation was reserved to Member States only. The EU was originally not intended to be a party. The reason for this exclusion of the EU as a party was the Commission’s reading of Opinion 1/09. There, the CJEU mentioned the Benelux Court as an example of an international law-based IP court that is compatible with EU law because of its links with the national court systems of the Benelux states. The Commission understood that mention – arguably without necessity³¹ – as an absolute requirement to exclude all non-Member State parties from the agreement, i.e. the EU as well as any non-EU members of the EPC.

The UPC is essentially an overhaul of the EEUPC model in order to fit Opinion 1/09, plus some extra compromise introduced in the course of remodeling. Therefore, like the EEUPC, the UPC has jurisdiction over both EPC and EU patents. As regards the national law side, purely national patents remain intact and stay outside of the system. Like the EEUPC, the UPC is a two-instance, international law-based, decentralized court. And it includes the same possibility (optional for first-instance, decentralized chambers) of splitting infringement and invalidity procedures (so-called bifurcation compromise).

There are, however, also some main differences between the UPC and the EEUPC. One is the limitation of participation to EU Member States, excluding the EU as well as any third-party EPC States. Also, the UPC enjoys only limited jurisprudence over certain EPO acts relating to the UP. Excluded, importantly, are all acts in the course of grant of the patent. Those decisions can therefore neither be attacked before the UPC, nor does the CJEU hold a competence of last resort to second-check the interpretation of the patentability criteria applied to an application that eventually becomes a unitary right. Another difference consists in the links between the UPC system and EU law, which were re-worked as regards, for example, the scope of preliminary references or the UPC’s liability for breaches. Also, several new compromises were woven into the Agreement at the occasion of the re-opening the EEUPC model. Among them are, for example, the now predominantly national composition of the bench in large divisions or the tripartite partition of the central division (located in London, Paris and Munich). The UPC also includes procedural rules, which are a first comprehensive attempt at the cross-border consolidation of the various national traditions of civil procedure and are actually quite good in that respect.³²

In sum, the 2012 Package brings an integration of the former EU (now unitary) patent into the EPC system insofar as the EPO is responsible for its grant and administration. The UP is granted on the basis of both EU law Regulation 1257/2012) and the EPC (all patentability and procedural questions). Yet, there are no direct procedural links between the EPO and the CJEU, but only indirect links via a limited review of those acts by the UPC under CJEU (preliminary reference) control.

In terms of the legal assessment of the 2012 package, a vast number of concerns exist.³³ They can be grouped here under three headings, namely illegality and dysfunctionality of the UP, concerns over the legality and effectiveness of the UPC and the fragmentation of the internal market resulting from the enhanced cooperation model. As will be shown in the next section, the CJEU brushed off all of these concerns in its 2015 judgments, except for those relating to the UPC (which was not at issue in the proceedings). The UPCA’s legality in terms of a correct implementation of Opinion 1/09 is still uncertain. However, no new CJEU opinion will be sought – or in fact, in view of the lack of EU participation, can be sought.³⁴

3.2 CJEU Review of Regulation 1257/2012

Spain seized the CJEU with two actions for annulment, directed against each of the two Regulations.³⁵ The pleas in the (more relevant) case regarding the annulment of Regulation 1257/2012 consisted essentially of a lack of judicial review over EPO decisions (i.e. an infringement of the principle of rule of law), the failure of Regulation 1257/2012 to establish uniform protection (i.e. a breach of the requirements of Art. 118 TFEU), an arbitrary design of the patent (i.e. a misuse of powers), an alleged delegation of discretionary powers to the EPO (i.e. an infringement of Art. 291(2) TFEU and a misapplication of the Meroni case law),³⁶ and finally, the illegality of the link between the Regulation and the UPCA as regards the coupling of entry into force (i.e. a breach of the principles of autonomy and uniformity). In 2015, the Court turned both actions and all of the aforementioned pleas down.

Regarding the lack of regulatory substance and judicial review, the Court held that the implementation of the cooperation did not infringe the autonomy of EU law or Art. 118 TFEU respectively. The crucial assumption in this regard can be found right at the outset of the judgment:

[T]he contested Regulation is in no way intended to delimit … the conditions for granting European patents – which are exclusively governed by the EPC and not by EU law – and that it does not “incorporate” the procedure for granting European patents … into EU law. … it necessarily follows from the characterisation of the contested Regulation as “a special agreement within the meaning of Article 142 of the EPC” … that that Regulation merely (i) establishes the conditions under which a European patent previously granted by the EPO pursuant to the provisions of the EPC may, at the request of the patent proprietor, benefit from unitary effect and (ii) provides a definition of that unitary effect. [T]he … plea … to contest the legality … of the administrative procedure preceding the grant of a European patent … must therefore be rejected.³⁷

The Court thus views the Regulation as a form of cooperation provided for in the EPC, namely Art. 142 EPC. That Article allows clusters of EPC states to introduce single patent rights among themselves: In other words, the Court regards the Regulation as a mere implementing act to the EPC. Because it is an implementing act, the fact that certain key features are not contained in the Regulation – for example the conditions for patentability – is fine for the Court. In fact, the Court concludes that because of the mere implementing character, the whole administrative procedure before the EPO is legal in terms of EU law, thereby including the lack of judicial review over EPO decisions.

What the Court does here is to embrace the fictional model of transformation, which was developed by the Commission and Council already for the 2009 model as an alternative to a delegation of powers to the EPC – with a view to overcoming the EU law problems associated with such a delegation: Transformation means that EU law and EU law principles only kick in at the time of declaration of the unitary effect of a bundle of EPC patents. Anything before that is off the limits of EU law, meaning that acts leading up to the creation of the unitary right do not need to be subjected to CJEU review either. The Court’s ready embrace of the transformation fiction comes despite the fact that any exculpatory character of the transformation fiction was sharply – and rightly³⁸ – rejected by the Advocates General just a few years earlier in their Joint Position for Opinion 1/09.³⁹

The Court’s assumptions, however, are very hard to subscribe to for two reasons. The first reason is one of legal principle: How can the Regulation be an implementing act of the EPC when the EU is not party to the EPC? While Art. 142 EPC is addressed to EPC states and entitles them to act, the Regulation was passed by the Council, which is an EU institution. In particular, the Council is not the same as the “representatives of the Member States reunited in the Council” – which are no institution and cannot legislate for the EU. The second reason is that the “transformation fiction” creates a factual vacuum for legal protection against administrative EPO decisions: Transformation means that EU law only kicks in at the end of that procedure. The EPO de facto decides over an individual right conferred by EU law without being subject to Court control.

After the initial embrace of the transformation fiction by the Court, dismissal of the remaining challenges was a piece of cake. This is best visible as regards the Court’s approach to an infringement of the principles for delegation established in the Meroni jurisprudence. In a nutshell, Meroni states that the Commission may not delegate to agencies powers involving wide discretion. Since the Court had already embraced the transformation doctrine, Meroni could only have a limited significance for the case at hand in that it is only to be applied to acts after the conferral of uniform character – like the setting of the level of fees.⁴⁰ Fee-setting in particular does not, according to the Court, involve discretion, as it is “intrinsically linked to the implementation of the UP protection”.⁴¹ In fact, the Court says that it is probably not even delegated either: The Regulation entrusts the Member States with ensuring compliance of the EPO with the Regulation. So they delegated that task to the EPO, not the EU, and there is no direct delegation from the EU, to which Meroni would apply. Apart from the weaknesses of the transformation fiction in the first place, that approach may be convincing insofar as fee-setting is, of course, a normal task for the EPO under the cooperation. But that observation fails to explain the concerns over its discretion regarding the fee levels.

Likewise, regarding the issue of whether Art. 118 or Art. 114 forms the correct legal basis, the Court showed marked generosity.⁴² It held that Art. 118 allowed for harmonizing acts and not just for the establishment of sui generis regimes. If that is so, however, one may wonder what the remaining difference between Arts. 114 and 118 should consist of. The legislative history behind the norm was clearly to remove sui generis IP regimes from the formerly necessary⁴³ recourse to the supplementary legal basis of Art. 352 TFEU.

Finally, as concerns the alleged inversion of the principle of primacy and an infringement of the autonomy of EU law, the CJEU held that making the entry into force of the Regulation dependent upon the ratification of the Court Agreement does not raise concerns. Again, on a critical note, one may observe that the CJEU did in essence not deal with the inversion of primacy resulting from the Regulation’s reliance on national law for basic patent features. Whether such a court would be compatible with EU law, however, remains to be seen. The most important difference to the model under review in Opinion 1/09 was the kick-out of the EU and its restriction to EU Member States only. If that is no longer so, what is the remaining difference between the UPC and the previous, incompatible model? Thus, vested in the UPC model is uncertainty for years to come, because the issue will likely only reach the CJEU once the system is operational (e.g. via a reference from a national court).

3.3 Interim Summary

The CJEU’s assessment of the UP Regulations again shows a markedly friendly, albeit not always convincing, approach towards allowing the cooperation to go forward. Particularly unconvincing is the Court’s ready embrace of the transformation fiction, its classification of the Regulation as an implementing act to the EPC, and the Court’s awkward reasoning regarding the delegation of post-grant powers to the EPO. Similarly to the enhanced cooperation judgment, the Court again resorted to a purely formalistic approach that puts formal arguments over arguments directed against the actual effects of the measure.

The immediate result from the patent perspective is that the Patent Cooperation may proceed. Entry into force of the Regulations hinges upon the ratification of the UPCA. Currently, from among the 25 signatories, 11 + 1 (France) of required 10 + 3 (France, Germany, UK) ratifications have been deposited.

The States participating in the enhanced cooperation over the substantive unitary patent and the States participating in the UPC system are not congruent: Whereas Spain and Italy remain outside the enhanced cooperation, Italy has subscribed to the UPC system. In its stead, Poland, which forms part of the enhanced cooperation, abstained from the Agreement. This political complexity crates additional legal confusion and economic uncertainties within the system.

As regards the UPCA itself, there are severe concerns over its legality, in particular its compatibility with Opinion 1/09. The missing link with Member States’ courts, intrinsic differences to the Benelux Court, and a lack of review of (most) EPO decisions seem very problematic in that regard. Since, however, the EU is not party to that Agreement, it cannot be subjected to a preliminary legality check by the Court under Art. 218 TFEU. So we will only learn of its legality once the UPC becomes operational. Tools for legality review at that later stage are infringement proceedings against the participating Member States⁴⁴ or preliminary references brought to the CJEU from national courts⁴⁵ on the grounds of a lack of UPC competence because of the EU law incompatibility of the UPCA.

4 Is Brexit Breaking It?

Since the British vote to leave the EU in a referendum held in June 2016, commentators have been trying to assess the implications of a Brexit for the UP package.⁴⁶ The entry into force of the UPCA,⁴⁷ and thus of the Regulations coupled to it in terms of their own entry into force,⁴⁸ are clearly in serious jeopardy following the UK’s vote to exit the EU. In fact, the UK had delayed its ratification to await the outcome of the referendum. Recently, the UK declared that it would push ahead with ratification in spite of the Brexit vote.⁴⁹ However, the signals coming out of the UK following the Brexit vote have in part been contradictory, so that the credibility and feasibility of that government announcement is hard to assess. The Prime Minister recently announced that she aimed to leave the single market altogether,⁵⁰ thus logically also ending participation in the UP cooperation. At the same time, the UK Supreme Court had strengthened the role of the UK Parliament in the Brexit process.⁵¹ In spite of political announcements suggesting the opposite, it appears somewhat unlikely that ratification of the UPCA would occur separately from an overall deal between the UK government and Parliament on the conditions of Brexit. Some more delay can thus be expected.

Whether ratification is at all legally possible upon Brexit and whether it might save the Package from failing is not entirely clear. The different approaches and suggestions made in that regard will be discussed and assessed here in the following. Some⁵² commentators see the possibility to devise remedies to allow the UK to benefit from the UP and remain within the UPC system, others⁵³ are less optimistic in that regard. All commentators, however, agree that any form of UK post-Brexit participation in the UP Package will require both amendments to the UPCA and flanking international agreements concluded between the EU and the UK. The major divergences between commentators in that regard basically concern the degree of change or amendment required and the likelihood of success in terms of the EU legality of such constructs.

4.1 The Framework of Art. 50 TEU and General Post-Brexit Perspectives

The procedure to leave the EU is laid down in Art. 50 TEU in a quite simple and straightforward manner.⁵⁴ The intention to exit needs to be formally notified to the European Council, (which was done at the end of March 2017), which triggers a two-year (extendable) deadline for negotiations over the future relationship between the exiting State and the EU. Article 50 TEU suggests that the agreement may not entail any alteration to primary law. Any extras negotiated with the UK which would require amendments to the TEU, TFEU or Protocols, would go beyond the framework set by Art. 50 TEU and would thus be a Treaty revision⁵⁵ rather than an exit agreement. If no agreement is reached, the UK will theoretically be kicked out of the EU in the sense that EU law ceases to apply overnight.

Whereas initial UK government estimates from February and March 2016,⁵⁶ i.e. before the referendum, assumed that the withdrawal process would take around ten years at minimum, more recent government statements point to the intention to stay within the two-year period⁵⁷:

Theresa May announced that Britain will go for a hard Brexit in 2019, prioritizing immigration control over access to the single market. However, in order to avoid the absolute mess that a hard Brexit would cause, Mrs. May has called for a simple and straight-forward Great Repeal Act that will incorporate all EU Law into British Law, allowing the Government to adjust the provisions of this “repatriated EU Law” in the way it considers appropriate. Britain will be an independent nation again after digesting sixteen tons of acquis communautaire in one single bite.⁵⁸

The rationale behind embrace of a two-year deadline and a potential “hard Brexit” is clearly to cancel out the bargaining advantage for the EU vested in that deadline. Full withdrawal on the one hand and a maintaining of the pre-Brexit (EU) statutes on the other would relieve the UK from EU negotiating pressure. Instead, it seems that the UK intends to reset its integration status to zero, in order to subsequently resume an association process with the EU through (ideally) pressure-free negotiations. Whether that will work or, by contrast, the “hard Brexit” will simply replace the current legal pressures by then economic pressures to agree to trading terms with the EU remains to be seen. In addition, that “hard Brexit” model offers no prospects for trade in goods and services and continued access of UK companies to the EU market – something which from today’s point of view appears to be a rather costly and risky venture.

Apart from the “hard Brexit”, i.e. the sudden end of UK integration into EU law and the internal market and its essential reduction to a WTO trading status vis-à-vis Europe, the UK government before the referendum⁵⁹ had contemplated EEA membership (euphemistically termed the “Norway option”) or a tailor-made model like the Swiss or CETA-like examples. Without going into details here, suffice it to say that all of those models seek to secure a certain amount of internal market access for the UK. However, the greater and easier that access is, the greater the limitations to UK sovereignty.

4.2 “Hard Brexit” Patent Perspectives

If the “hard Brexit” option is chosen, the immediate effect for the UK patent system would be that nothing changes: The UK would remain a member of the EPO, which is not an EU institution but an independent international organization. Thus, patent applicants could continue to apply for European patents with the UK as one of the designated states of protection. Upon grant, those patents would, as now, become genuine UK patents for all of their post-grant aspects (infringement, invalidity, property aspects and any litigation relating to these, which would be before the national courts).

However, if we now look at this latter aspect of litigation, one immediate problem pops up: If the UK ends its EU membership, the Brussels I Regulation⁶⁰ on cross-border jurisdiction, recognition and enforcement of judgments in civil and commercial matters would no longer be applicable. Cross-border infringements of patents involving the UK could thus no longer be joined before one competent court (outside the UK or within the UK, including infringements in other states). Likewise, UK infringement or invalidity judgments would no longer be enforceable outside of the UK. In short therefore, the UK would, in order to maintain its current patent enforcement regime for cross-border infringements, be forced to draw up a follow-up regime to the Brussels I Regulation. An option here would be its accession to the Lugano Convention.⁶¹ However, the EU is a party to that Convention, so that some degree of negotiation with the EU could not be avoided, even in this “hard Brexit” scenario for patents.

4.3 “Negotiated Brexit” Patent Perspectives

If a “hard Brexit” is not an option (at least in the patent area), the UK will be forced to negotiate with the EU. This is essentially the conclusion reached by all commentators so far.⁶² Their opinions differ only, as was pointed out, in the assessment of the extent of necessary amendments. Also, it should be pointed out here that the vast majority of commentators harbor some doubts (again in differing degrees)⁶³ whether there is a way compatible with EU law to keep the UK within the 2012 Package at all.

The discussion over keeping the UK in the 2012 Package essentially has three prongs, which are interrelated. They are, respectively, the UK’s participation in the substantive patent right under the UP Regulation, its participation in the UPC litigation model, and its continued eligibility to host one of the UPC’s three central divisions of first instance. As will be shown below, the latter two aspects are interrelated and will be dealt with together here.

4.3.1 Staying Onboard the UP Regulation

As regards the first aspect, the continued participation of the UK in the substantive UP, the initial assumption would be that the UK is out post-Brexit. This also seems to be what the Prime Minister suggests when she underlined recently that the UK was precisely not seeking “anything that leaves us half-in, half-out [nor] to hold on to bits of membership as we leave.”⁶⁴ If that is really so, the 2012 Package’s prospects are dim, both legally (UK ratification requirement) as well as factually (third largest patent jurisdiction out).

What is more, the UP is designed to be a uniform, single right throughout the participating EU Member States only. In turn, Brexit is defined as the end of an applicability of EU law to the UK.⁶⁵ Accordingly, the UK could no longer enjoy the benefits of the UP Regulation directly, since that Regulation is an instrument of EU law. EU law only applies to EU Member States.⁶⁶ Likewise, the tool of enhanced cooperation under Art. 20 TEU, which forms the legal framework for the two Regulations, is only open to EU Member States.

In short, a UK that was no longer a member of the EU would essentially be sized-down to the status of any other third non-EU state. In the patent context, this is the plain legal status of an EPC member outside the UP Package. Just as other third-party EPC members are not and were never intended to be able to participate in the setup of an EU sui generis patent for the internal market, the UK would not be able to participate either.

Another question is whether the effects of the UP Regulation could be extended via an international agreement to include the UK. There is indeed an example in EU law for such an extension of EU secondary law to third states, namely the Schengen border management regime. It is laid down as a policy in primary law⁶⁷ and implemented through secondary law measures, yet includes third states as participants. For some of them (Iceland and Norway) that association status is grounded on an explicit primary law basis.⁶⁸ For others, however, namely Liechtenstein and Switzerland, there is no such primary law basis. Yet they participate in both the application and development of Schengen rules on equal terms.

Because of the latter aspect, the Schengen constellation is not to be confused with those numerous, regular situations where an international agreement establishes parallel bodies of law in the EU and the respective other parties to that agreement. There, in spite of the common legal obligation flowing from the agreement, the laws created within each of the jurisdictions to implement the agreement are legally distinct and (at least formally) independent from one another. The most prominent example from the area of internal market law for this regular parallelism of laws is the EEA Agreement.

In the Schengen constellation, by contrast, the same body of secondary law applies immediately to the EU and the other parties. This is essentially what the proponents of a post-Brexit UK participation in the UP are looking for: In order to yield the benefits associated with the UP, namely the unitary character as one right in all of its post-grant aspects throughout all participating states.

From the Schengen example, it thus looks as if EU law did not categorically preclude the possibility of extending its direct applicability to third states. However, in view of the explicit primary law basis for the association of at least some third states, it is unclear to what extent that example might have a wider scope of application to other areas of EU law as well – like the UP. In addition, as the CJEU has pointed out on numerous occasions (including the assessment of the 2009 patent package),⁶⁹ any application of EU law must, irrespective of the context or body, always go hand in hand with sufficient CJEU control of last resort (i.e. at least through preliminary references).⁷⁰

In terms of the regulatory technique for an extension of the UP Regulation’s applicability to the UK, some commentators⁷¹ suggest that this could be done in the form of a simple international agreement between the EU and the UK. More specifically, they see such an agreement as a form of regional patent cooperation envisaged and authorized by Art. 142 EPC. The EPC foresees that any of its parties may enter into closer patent cooperation, and the UP is just that. Factually, that logic may be correct. Legally, however, as will be shown below, is problematic.

The reference to Art. 142 EPC might be intended to overcome the key problem that there is no primary law authorization to extend the patent acquis to third states like the UK. At first sight, reference to the EPC for authority to include third states might seem absurd: The EPC is, after all, no EU law act. How could it then authorize special arrangements relating to the application of EU law?

However, the Commission has used that fiction before and, more importantly, it was also endorsed by the CJEU in its jurisprudence⁷²: According to that fiction, the UP Regulation itself is a regional agreement in the sense of Art. 142 EPC. For the Regulation, that argument was important to explain why the Regulation did not need to include any rules on patentability and why, and that it could refer to the EPC instead. For the CJEU, acceptance of the Art. 142 fiction was a key element for its reasoning of EU law compatibility of the UP Regulation.⁷³

As was mentioned before, in spite of the CJEU endorsement, the Art. 142 fiction is legally all but convincing: It confuses the EU legislator, the Council, with its Member States’ representatives. Since the EU is not a party to the EPC, the Council cannot conclude a regional agreement within the sense of Art. 142 EPC. At the same time, the Council is a body and institution distinct from the representatives of the Member States reunited in the Council. Arguably, they could conclude an Art. 142 EPC agreement, but such an agreement could never take the form of an EU legislative act – such as the UP Regulation. That the CJEU nonetheless endorsed the Art. 142 fiction is more a sign of its acknowledgement of the strong political will to implement an EU patent solution after decades of failure rather than proof of its legally solid build.

If the UP Regulation was really an Art. 142 agreement and therefore could derive its legitimacy (and substantive determination) from the EPC, the suggestion that an agreement between the EU and the UK to keep the UK onboard the UP is stringent. That proposal might then indeed be a workable way out of the Brexit dilemma. If the EU, which is not an EPC party, may pass legislation to implement the EPC, it might also choose to amend that legislation to bring other EPC non-EU parties onboard as well: An agreement between the EU and the UK regarding the extension of the UP Regulation to the UK would just be another Art. 142 agreement next to the existing one (i.e. the UP Regulation).

The CJEU’s endorsement of the Art. 142 fiction is indeed a strong argument that an agreement between the EU and the UK regarding the extension of the UP Regulation to the UK might also be legal. Reservations against the legal construction of that fiction are de facto no longer relevant after the CJEU’s endorsement. The EPC as a legal basis might then fulfill a similar legitimizing function for secondary legislation as is in place for the extension of the Schengen acquis to third states.

In sum therefore, the commentators suggesting a duplication of the Art. 142 solution to overcome the Brexit dilemma might be correct. There is a fair chance, well beyond mere speculation, that the CJEU would also wave that second agreement through. All it would take for this would be a cursory reference to the preexisting case law that ascertains (albeit does not explain) the UP Regulation’s Art. 142 agreement character.

If that Art. 142 fiction-based construction was duplicated successfully for the UK case, this would open the way to include more EPC states in the UP cooperation – well beyond the UK. From the perspective of the coherence of patent law in Europe that might, in fact, be a positive prospect. It would, of course, come at the price of those third states’ subjection to the jurisdiction of the CJEU regarding the interpretation and application of the scope of the UP and the rights associated with it.

4.3.2 Staying Onboard the UPC

As regards the UPCA and litigation system, the situation is likewise complicated. Again, the initial impulse would be that the UK has to leave: Although the UPCA is formally an instrument of public international law, that Agreement is open to EU Member States only.⁷⁴

This is so at least in the Commission’s understanding of CJEU Opinion 1/09: Under a misguided⁷⁵ reading of the Court’s Opinion, the Commission understood Opinion 1/09 as meaning that the CJEU requires an international law-based patent court to remain restricted to EU members only.

The Commission’s narrow reading of Opinion 1/09 was the main reason why the well-designed 2009 EEUPC model had been downsized both territorially and substantively for the UPC proposal, which now affords comparatively less substantive patent law coherence and is intrinsically defunct.⁷⁶ In sum, the UPC was essentially intended to be a copy of the BENELUX Court, which the CJEU in the Opinion had mentioned as an example for an EU-law compatible international law-based court, i.e. “a court common to [EU] Member States and thus subject to the same obligations under Union law as any national court”.⁷⁷

What that approach overlooks, of course, is that the CJEU’s approach to the EU system of remedies under Art. 19 TEU and Art. 267 TFEU and to safeguarding the autonomy of EU law is an effects-based one, not a formal one. In other words, a merely formal designation of a court as being a Member States’ court which does not also change that court’s factual setup and structure will therefore not tackle concerns of a potential incompatibility with EU law. Call it BENELUX and it will be BENELUX doesn’t work.

Instead, there would have to be a real link between national courts and the respective international court, which neither the incompatible 2009 model nor the UPC show. As the CJEU put it in the Miles judgment passed shortly after Opinion 1/09:

[I]t is true that the Court of Justice has held, in [the Dior judgment⁷⁸], that there is no good reason why a court common to a number of Member States, such as the Benelux Court of Justice, should not be able to submit questions to the Court of Justice, in the same way as courts or tribunals of any of those Member States. … However … the Benelux Court … procedure … is a step in the proceedings before the national courts leading to definitive interpretations of common Benelux legal rules …, [whereas the international law-based institution at hand] does not have any such links with the judicial systems of the Member States. … Moreover, [the institution at hand] is a body of an international organisation which, despite the functional links which it has with the Union, remains formally distinct from it and from those Member States. … In those circumstances, the mere fact that [such an institution] is required to apply the general principles of EU law when it has a dispute before it is not sufficient to make [it] fall within the definition of “court or tribunal of a Member State”.⁷⁹

In fact, some of the proponents of a post-Brexit participation of the UK in the UPC system recognize that the UPC is in no way similar to the BENELUX Court – thereby acknowledging the potential EU law incompatibility of the UPC, but at the same time opening the door for the UK to remain in:

The UPC, as the product of an international agreement, is an international tribunal. … If the UPC were truly part of the Union legal order, it would already be subject to [obligations like the respect for EU law and the duty to ask the CJEU for preliminary references] without them needing to be spelled out in the Agreement. Whilst Art. 1 of the UPCA and Art. 71a of the Brussels Regulation designate the UPC as a “court common to a number of Member States”, we do not consider that such secondary legislation is capable of converting the UPC’s fundamental status as an international court into that of a court which is part of the national legal order.⁸⁰

The logic of that argument is apparently that if the UPC is currently deemed in line with Opinion 1/09, although it is in no way a court of the Member States, this apparently means that Opinion 1/09 did not mean to impose that restriction in the first place. Whether that argument flies remains to be seen.

Be that as it may, if the Commission’s narrow reading of Opinion 1/09 is upheld and no alternative visions for a court model are developed, there is no way that the UK could remain in that litigation system. In turn, if another reading for Opinion 1/09 were developed, that might not just open the way for keeping the UK inside of the litigation branch, but for bringing in all of the remaining non-EU EPC states that were removed from the litigation leg following its overhaul after Opinion 1/09.

Accordingly, some commentators⁸¹ suggest an alternative reading of Opinion 1/09 to allow for the participation of non-EU states. The more adventurous among them⁸² suggest really just to change the reading for Opinion 1/09 without altering anything in the UPCA: The UK is urged to ratify the UPCA quickly,⁸³ while it is still an EU Member. A subsequent Brexit would, according to that argument, not affect its position as a UPC participant:

However, if the UK leaves it is nowhere written that they have to leave the UPC. There is simply no provision in the UPC which states that they have to leave the UPC. It is true that the present UPCA does not allow the participation of non-EU Member States but there is no provision for the situation in which a Member State having ratified the UPCA becomes a non-Member State.⁸⁴

Apparently, as was mentioned, that call has now been heeded: The UK government declared that it would carry the ratification forward in spite of the Brexit vote.⁸⁵

That view seems problematic for a number of reasons. It is, again, based on formality (date of accession) and not effects (status of a state as inside or outside the EU legal order and the scope of Art. 267 TFEU). In essence, it expresses in itself a misconception of the concerns in Opinion 1/09. While that view is correct in pointing out that [i]t is a widely held misconception … that the European Court of Justice has ruled that only EU member states can be a member of the UPC. The European Court of Justice has only held that the Supremacy of EU law has to be recognized, it overlooks another important reason – next to primacy and autonomy – behind the CJEU’s concerns, namely legality.

Primary law establishes a complete system of remedies for the interpretation and application of EU law. As Art. 19 TEU and Art. 267 TFEU show, national courts are part of that system. As long as the UPC is a Member States’ court (which in essence it is not, but that’s on a different note),⁸⁶ it is situated within that primary law basis. If, however, the UPC includes non-EU states, that inclusion has no legal basis. Such a legal basis could, however, arguably be created, where the limits posed by EU law are taken into sufficient account,⁸⁷ by an EU accession to the UPCA. Without the EU being a party to the UPCA, however, there is no legal basis for an inclusion for its courts – directly or indirectly via the UPC detour – in the EU system of remedies.

Another commentator⁸⁸ sees that problem and thus suggests bringing the EU in not via an amendment of the UPCA but through the Brexit agreement pursuant to Art. 50 TEU:

The CJEU, in its Opinion 1/09, has decided, that the UPCA may be concluded only if the referral procedure under Art 267 TFEU is not jeopardized. Pursuant to Art 21 UPCA … the requirements of Art 267 TFEU are met to a full extent. It is true that under Art 267 TFEU only the courts of EU-MS are permitted to refer questions …, but the UPC is a common court of EU-MS … and would not lose that character, if a non-EU-MS (UK) who has ratified the UPCA being an EU-MS would continue to participate in the UPCA after leaving the Union, because that state, in ratifying, had fully accepted Art 21 UPCA and Art 267 TFEU and is bound to accept the Union law as defined by the CJEU. A statement to that effect could be included in the agreement based on Art 142 UPCA and also be endorsed in the exit-agreement EU-UK. … Therefore, Union law would allow a change of the UPCA permitting a [contracting state to the UPC], who had been an EU-MS at the time of ratification, to remain as a member state of the UPCA …. This change could be made by the Administrative Council of the UPC …, if the exit-agreement (having the legal status of Union law) would contain a parallel text. In that case, no ratification by the [UK] would be necessary.⁸⁹

This approach is admittedly more EU law sensitive, but it is still doubtful in terms of its EU law compatibility. After all, the CJEU has made it unequivocally clear in Opinion 1/09 and the jurisprudence preceding it, that a number of fundamental principles are to be respected in the setup of a court system. Those are, in particular,

• 1. EU law primacy over the laws of the Member States and its direct effect⁹⁰;

• 2. the autonomy of EU law and the exclusive role of the CJEU in safeguarding autonomy⁹¹;

• 3. the principle of sincere cooperation in the application and implementation of EU law⁹²;

• 4. safeguarding the judicial protection of individual rights based on EU law⁹³;

• 5. access of those individuals to a court where EU law-based rights may be adjudicated under the final control of the CJEU⁹⁴;

• 6. the principle of completeness of the EU law system of remedies and procedures, so that any EU law-based claim can be brought either before a national or an EU court⁹⁵;

• 7. the principle that a Member State is obliged to make good damage caused to individuals as a result of breaches of European Union law⁹⁶;

• 8. the principle that EU law infringements may be the subject of infringement proceedings before the CJEU.⁹⁷

These principles are quite demanding vis-à-vis the comprehensive character and density of an arrangement altering the court competences for the litigation of EU law-based rights. In fact, even the UPCA in its present form is very likely not in line with all of those principles.⁹⁸ Since, however, the EU is not a party to the Agreement, its legality cannot be checked beforehand under the procedure provided for in Art. 218(11) TFEU.

It does not seem entirely unrealistic that the required link between the UK adhering to the UPCA and EU law could be established not only by an amendment or alteration of that Agreement, but, as suggested,⁹⁹ via a reference to the UPCA in another international agreement concluded between the UK and the EU, namely the Brexit agreement.

That second agreement would thus need to have the effect of incorporating the UPCA into EU law by way of reference to it: The aforementioned principles can only be safeguarded where the UPCA is inside the EU legal order, and thus inside of its system of remedies and not outside of it.

Given that the UP Regulation incorporates a similar technique of extensive referencing to third law (EPC and national law, the latter partially harmonized by the UPCA and partially not harmonized) and given that it was waived through in that aspect by the CJEU, it may seem justified to extend such a referencing technique to the Brexit case. However, a closer look reveals the limits of such a referencing posed by the principles of legality and legal certainty: The CJEU needed to embrace the Art. 142 fiction and the so-called transformation fiction, used in Art. 3 of the UP Regulation, which are both unprecedented in EU law and seemed highly controversial, to overcome the problem of a legal basis for the UP Regulation. Looking at it as an “implementing [act for] the provisions”¹⁰⁰ of the EPC provided the legal basis needed.

The transformation fiction goes as follows¹⁰¹: UPs will be granted by the EPO according to the applicable rules under the EPC for the granting of European patents. From there on, there are two ways to classify this technique. Under first classification, the theory of delegation, administrative powers (to examine applications etc. and grant the patent) would be delegated to the EPO by the EU, which would grant UPs instead of and in the place of an EU Union agency. That theory was unattractive for the Commission because of concerns over the compatibility of such a delegation with the Meroni principles.¹⁰² Under the second classification, the theory of transformation, there is no delegation. The EPO would not act on behalf of the EU and only grants a European patent pursuant to the EPC. That European patent, however, is subsequently transformed in the Union’s legal system, to become a UP, automatically and solely through the effect of the UP Regulation. The hope was that this would avoid the Meroni dilemma. With reference to Art. 142 EPC, the CJEU eventually acknowledged the transformation fiction as feasible and apt to overcome the Meroni concerns,¹⁰³ which thus only kick in regarding the post-grant aspects of the management of UPs by the EPO.¹⁰⁴

Similarly, the UP Regulation does not incorporate the UPCA directly in the area of rights and remedies, but only via a reference to national law.¹⁰⁵ Of course, the UPCA stands behind those national laws and harmonizes them. Yet technically, the UP Regulation references to national law, not international law. National law’s function to fill in the blanks left by EU harmonizing legislation is, however, a regular feature of the EU legal order and in fact an expression of the principles of subsidiarity, conferral of powers and procedural autonomy. None of this is true for references to international agreements.

In short, the solution proposed by some commentators to allow for a continued UK participation in the UPC system post-Brexit by a mere reference in the Brexit agreement concluded pursuant to Art. 50 TEU (or any other Brexit agreement, for that matter), does not seem viable. Such a reference does not seem to respect the principle of legality and moreover does not appear to incorporate the UPCA into the EU legal order in a manner sufficient to safeguard the key principles of EU law which might be affected by a change to the system of litigation of EU-based rights.

Although politically not desired by the proponents of a quick implementation of the 2012 Package, the legally most reliable way to allow for a continued participation of the UK and other third states in the UPC system is to reopen the Agreement and to change it to reverse the misguided Commission reading that only EU Member States may participate in such court system.¹⁰⁶ Actually, given that the UPC is structurally really not a court of the Member States like the BENELUX Court anyway and thus in essence is still incompatible with EU law under the strict reading, reopening would be a sensible option from that point of view also. Simply put: The Brexit opportunity should be seized to remedy the flaws of the all-too-quick modifications to the 2009 model after Opinion 1/09 and to bring non-EU EPC states back on board.

How could this be done? The Commission’s reading of Opinion 1/09 seems misguided in two major respects. Firstly, the CJEU’s mention of the BENELUX Court is only one example of possible litigation models, not the only possible model. The Court has made this quite clear already in the Opinion.¹⁰⁷ Secondly, the crucial point regarding the legality or illegality of the court model is not whether it is called BENELUX or whether it is concluded between EU Member States only, but rather, whether that court has the power to apply EU law directly and if so, under what conditions. Those conditions were listed just above, in particular such a system’s unconditional respect for the CJEU’s exclusive role under Art. 19 TEU and, consequently, the autonomy of the EU legal order.¹⁰⁸

The Court has made it clear that where those conditions are met,

an international agreement providing for the creation of a court responsible for the interpretation of its provisions … is not, in principle, incompatible with European Union law. The competence of the European Union in the field of international relations and its capacity to conclude international agreements necessarily entail the power to submit itself to the decisions of a court which is created or designated by such agreements as regards the interpretation and application of their provisions[.] Moreover, … an international agreement concluded with third countries may confer new judicial powers on the Court provided that in so doing it does not change the essential character of the function of the Court as conceived in the EU and FEU Treaties … The Court has also declared that an international agreement may affect its own powers provided that the indispensable conditions for safeguarding the essential character of those powers are satisfied and, consequently, there is no adverse effect on the autonomy of the European Union legal order.¹⁰⁹

Cornerstones of alternatives to the current UPC setup have been elaborated elsewhere¹¹⁰ and will not be duplicated here. Possible ways forward consist in (alternatively)

• 1. an abandonment of the Commission’s narrow reading of Opinion 1/09;

• 2. the creation of an explicit primary law basis for an international court for cross-border patent litigation;

• 3. the creation of a genuine Benelux-type court (i.e., unlike now, procedurally linked to the national courts)¹¹¹;

• 4. revisiting Art. 257 TFEU-based models like the one proposed 2004¹¹²;

• 5. handing the plans to create a single patent right and court back to EPC law (i.e. an Art. 142 EPC-type and EPLA-type solution).

Solution numbers two to five would require an entirely new court model and agreement. Solution number one also likely requires some amendments to the UPCA, but, if successful, would allow keeping the UPCA in principle. Rethinking the UPC model in such a way involves modifying some, albeit crucial, details safeguarding the autonomy of EU law. The eventual model might actually look quite similar to what was proposed in 2009, just reinforced in terms of its respect for the aforementioned EU law principles. Such reinforcement might (in addition to responding to the immediate points of criticism under Opinion 1/09, like responsibility for damages and infringements) particularly take the form of a reinforcement of the respect for the autonomy of EU law and a watertight revision of the preliminary reference procedure.¹¹³ Any overhaul would have to ensure EU law primacy, autonomy and an exclusive role for the CJEU in all aspects of the interpretation and application of the UP throughout all of its lifecycle (from application to expiry and from validity over contracts and licensing issues, including compulsory licenses), thereby safeguard full judicial protection of individual rights based on EU law in respect of every legal and commercial aspect of the UP. If, by contrast, solution number one is not viable because the narrow reading of Opinion 1/09 is really the only correct reading, a court based on international law will never be in the position to litigate UP issues. As a consequence, one of the solutions numbers two to five would have to be chosen.

A legal basis for solution number three, which might even go beyond a mere preliminary references jurisdiction and involve the establishment of a direct appeals system ex the national courts, might be found in Art. 262 TFEU, which allows to entrust the EU courts with patent infringement proceedings. Arguably, that norm has specialized EU patent courts in the sense of Art. 257 TFEU and direct appeals to the GC as its primary focus, as devised by the (unsuccessful) 2004 proposal.¹¹⁴ However, Art. 262 TFEU does not explicitly limit its application to such a model. Consequently, a generous reading of that norm might include an international-law based patent court with subsequent direct appeals to the EU courts. Such a generous reading might be applied in particular with a view to overcoming the CJEU’s concerns in Opinion 1/09 that there was no primary law basis for an international law-based court unless it could be brought within the scope of Art. 267 TFEU. If Art. 19 TEU and Art. 262 TFEU are read in conjunction and in a generous manner, that primary law basis might well be there.

It should also be added on a final note that the safest way out of the dilemma of a missing primary law basis for an EEUPC-/UPC-type of patent court would be an amendment of primary law. Potential candidates in that regard are Art. 257 TFEU (to include international law-based courts there or in a new Art. 257a TFEU) and Art. 262 TFEU (in terms of a clarification along the lines just suggested). Of course, since such a revision of primary law could not be brought under the simplified procedure,¹¹⁵ such a process might be lengthy. However, should other primary law changes be needed alongside in the Brexit procedure, these changes could also be undertaken, thereby providing legal certainty for the continued participation of the UK in that system.

If, by contrast, the UPC model would have to be abandoned for whatever (legal¹¹⁶ or political) reason, that would not necessarily be the end of the story either: Alternative models, e.g. a mere joint court of appeals ex the national courts,¹¹⁷ seem feasible – at least from an EU law point of view (national constitutional laws might however have to be screened for such a model).¹¹⁸ That latter model of a mere appeals court in particular would, in addition and unlike the current UPC, boast sufficient links with the national judiciary to render it a Member State court similar to the BENELUX.

Where the Commission is keen to save the 2012 Package from another – and this time likely terminal – failure, it should start rethinking its litigation model. This can be done on the go through another revision of the agreement. Admittedly, this will slow down the ratification process once more. But that might still look like the better option compared to the looming alternative of an inevitable collapse of the system should the UK not be kept in post-Brexit for legal or political reasons.

To be clear: There will de facto be no UP if one of Europe’s three largest patent jurisdictions remains out. So any pains to keep the UK in should be taken. However, any solutions must be legally solid and cannot be built upon uncertainties regarding their EU law compatibility. It is well known that the economy does not appreciate uncertainty. The same is true for patent holders as the would-be users of the new system.

4.3.3 The London UPC Division

The third initial question, that of the fate of the institutional setup of the UPC, which foresees a life sciences and chemistry section of the UPC’s central division to be located in London, is intrinsically linked to the UK’s continued participation in the UPCA: If that participation can be secured, any of the required or agreed UPC’s (regional, local and central) divisions may be kept within the UK.¹¹⁹ Likewise, UK citizens could continue to act as judges there and on the benches of other divisions throughout the participating Member States.¹²⁰

However, it should also be noted that the partition of the UPC’s central division in three is the result of a political compromise aiming at satisfying the power and economic interests of the three biggest patent jurisdictions in Europe (Germany, UK and France). It is not functionally warranted and probably even counter-productive in terms of the efficiency and quality of jurisprudence. The reopening of the UPCA to bring it in line with Opinion 1/09 and to accommodate for post-Brexit UK membership, as suggested just before, would provide a welcomed opportunity to cut back this and other dysfunctional compromises (e.g. also regarding bifurcation or the predominantly national composition of larger division benches).

4.3.4 Link Between UK Participation in the Regulation and the UPC

It was highlighted before that the UP Package has no political future if the UK cannot be kept on board. This is, on a final note, even more true legally: As was mentioned, the two UP Regulations and the UPCA are coupled in terms of their coordinated entry into force.¹²¹ Only once the UPCA was ratified by Germany, the UK and France plus ten other contracting states will that Agreement, and with it the two Regulations, enter into force.¹²²

The UK has (just like Germany) not yet ratified the UPCA. Initially, the intention was to await the outcome of the Brexit referendum. Since Brexit has become a fact, it is doubtful whether ratification is still legally possible and sensible. Initially, those doubts delayed ratification in the UK. Recently, the UK declared that it would finally go ahead.¹²³

We may still expect that domestic discussion to go on for a while for reasons well beyond patent law: The UK Supreme Court strengthened the rights of Parliament in the Brexit process.¹²⁴ In addition, any post-Brexit legislation, starting already with the repeal of the European Communities Act of 1972, depends on parliamentary cooperation and consent. The Prime Minister’s former strategy of presenting Parliament with a fait accompli in order to force it into post-Brexit cooperation did not work out. Now that parliamentary involvement has been confirmed, ratification of the UPCA might become part of a broader solution on the terms of, and strategies for, Brexit.

If the UK doesn’t ratify the UPCA or if it exits from the EU before such ratification, the UPCA (and with it the entire Package) is permanently barred from entering into force.¹²⁵ The only option then would be a change to the UPCA’s clause on entry into force to eliminate its (indirect) reference to the UK – which would, of course, require UK consent. In short, if the UK chooses not to ratify the UPCA in view of the current Brexit turmoil and within two years from the time of its exit notification, the UPCA will have to be reopened anyway to adapt it to the new circumstances. If the UK does not lend its consent to such a reopening, a new Agreement would have to be drawn up altogether, marring the UPCA, the 2012 Patent Package and very likely the EU’s sui generis patent ambitions altogether.

The call for a quick ratification by the UK therefore at first sight seems to be in the interest of the remaining contracting UPC states, since the UPCA can only then enter into force. In turn, once ratified, a subsequent Brexit would not legally affect the functioning of that system. Upon closer look however, it might seem wiser not to, like so many times before in the EU patent saga,¹²⁶ hurry to reassemble the leftovers of a crumbling system patchwork-style. In view of the 2012 Package’s extensive flaws in terms of both its substantive law and court system,¹²⁷ one step back should be taken to reconsider the way forward and assess alternatives. The Commission’s attempt to devise a quick and questionable solution by compromising on all levels of the functionality of the patent system has failed. The Brexit provides an opportunity to turn the UP into a system that will deliver actual, attractive and balanced results for its users and society instead of its mere usability in political rhetoric.

5 Conclusions

After decades of historic rivalry between the EU and the non-EU path of patent law integration in Europe and so many setbacks for the EU plans, implementation of the 2012 UP Package recently seemed so close. The EU legislator had gone to unprecedented lengths and accepted compromise to an extent never seen before in internal market legislation to make the EU patent finally reality. That outstanding determination was subsequently honored by the CJEU, where both the Decision to implement the Package via enhanced cooperation and the implementing acts in the form of two Regulations were cleared in spite of severe prior reservations regarding their EU law compatibility on several levels and the overall functionality of the system thus created. Then came the Brexit vote and the huge institutional and political effort is poised to fall apart once more – this time likely for good.

It is not surprising that commentators friendly to the political compromise incorporated by the UP Package, despite its extensive legal and functional flaws, quickly jumped to its rescue. The proposals developed vary in terms of sophistication and self-assessed likelihood of success. What they have in common is, in particular, a new reading of Opinion 1/09 and an abandonment of the former (Commission-initiated) fiction that access to the UPC had to be restricted to EU Member States and that the UPC was indeed a court of the Member States in the sense of Art. 267 TFEU, similar to the BENELUX Court. The UPC, of course, never was anything like that in terms of its tasks and structural setup.¹²⁸ The Brexit vote forced proponents of the system to acknowledge this, change their perspectives and abandon the legally convenient fiction: If the UK is to stay onboard the UPC and UP at all, it would inevitably do so as a non-EU state. If, therefore, the UK does ratify and subsequently Brexits, a non-EU MS will enter that system, so the UPC will be a different type of court from what the Commission propagated since Opinion 1/09: The claim that the UPC is compatible with EU law, because it is really not at all an international court, but rather part of the national court systems, could no longer be made. This would open the way for other non-EU EPC states back into the system. However, the legality of the UPC model still involves much uncertainty for years to come. This is already true without Brexit, and even more so if the UPC system includes a post-Brexit UK.

The change of perspective on the UPC that now seems to be spreading among commentators, however, precisely offers a dire opportunity to reopen the 2012 compromise and remedy its flaws, both in substantive terms (i.e. for the UP Regulation, which lacks all determination of the shape of the right and neglects third-party interests) and in terms of the litigation system (which is ridden by compromise and thus dysfunctional). One of the major flaws in that regard was the original sin committed after Opinion 1/09 to compromise the coherence of patent law in the EEA market by kicking certain EU (enhanced cooperation) and non-EU (UPC) states out in a misguided attempt to push the patent project over the finish line more easily. However, as the US experience with the older patent case law of the Court of Appeals for the Federal Circuit (CAFC) has taught us,¹²⁹ a bad patent system is worse having than no patent system. Brexit provides the opportunity to take a step back, reset the table and start afresh.

Such a fresh start does not have to begin at zero. It is true that the race is back on for the development of a pan-European patent court system that allows third states to participate, thereby remedying the key deficit of the existing (non-EU) EPC path of patent integration and providing coherence of the body of patent law (which the 2012 model does not). Nonetheless, alternatives have long been proposed in literature.¹³⁰

Those existing proposals should be revisited, double-checked for their EU law compatibility, functionality and political acceptance and eventually be reassembled for a new patent package. A look at these models suggests that the UPC would not have to be entirely overthrown to accommodate the changes necessary to open it for the participation of third states, like the UK and any other non-EU member of the EPC.

It does not seem unrealistic, in particular, to contemplate a model similar to the 2009 proposal,¹³¹ but reinforced for its EU law compatibility. Such a model would, in particular need to include a watertight revision of the preliminary reference procedure or, alternatively, its abandonment and replacement by direct appeals to the GC or the CJEU ex the patent court for EU patents only. Their EU law compatibility could be ensured by a clarification of the primary law basis of such a court, e.g. in or after Art. 257 TFEU or in Art. 262 TFEU.

Should a reinforcement of the 2009 model not prove workable, the next best option seems a common court of appeals ex the national patent courts, as was proposed in 2002.¹³² That latter option seems well workable from an EU law perspective and would certainly be in line with Opinion 1/09, as it is in fact (and unlike the UPC) very similar to the BENELUX model. The main difference between that appeals court and the BENELUX court is that the latter takes preliminary references from the national courts, whereas the former would deal with direct appeals.

It should be noted that the initial proposal for the mentioned appeals court failed in particular because of national constitutional reservations (particularly under the French constitution) against such a system. Those concerns may be valid. However, the follow-up we are now faced with is a court model that may be compatible with national constitutional laws (particularly the French constitution, under which the UPCA was already ratified), but seems incompatible with EU primary law. For national politicians, that may be the more easily acceptable option, but from the EU perspective it is clearly not.

The work regarding implementation of the UPC is already far advanced. The Court’s preliminary set of provisions for the Rules of Procedure¹³³ have been elaborated and seem quite well-drafted. The process for recruitment of legally and technically qualified judges is underway irrespective of the Brexit vote, with the finalization of the appointment procedure, however, dependent on the entry into force of the UPCA.

Therefore, all signs for the UPC are on “go”. Except for the most important ones, namely its intrinsic functionality, legality and implementation prospects after the Brexit vote. The Commission’s decade-long determination to come up with an EU patent in spite of stiff political opposition has evidently not done the project any good: There is so much compromise in the system, that it is now really just an emergency patchwork. The most visible example for this is the fact that both the substantive right (UP Regulation) and the litigation system (UPC) are limited to certain states for participation. This is so far removed from what the Commission initially had in mind and what boosting business in Europe would actually need. Every new obstacle that pops up adds new compromise and new dysfunctionalities – CJEU Opinion 1/09, the Spanish language dispute and now Brexit. Let’s thus take the time for another loop to reset, rethink and readjust the process to make necessary adjustments. After that, go ahead with full force and the EU will come up as a winner in its systemic rivalry with the EPC – but this time as a winner in terms of quality, not just speed.