UNITED STATES DISTRICT COURT
DISTRICT OF COLUMBIA
|
CEF Energia, B.V. Petitioner,
The Italian Republic Respondent. Greentech Energy Systems A/S (now known as Petitioners,
The Italian Republic Respondent. |
Case No. 1:19-cv-03443 |
1. I, Piet Eeckhout, make this declaration based upon my personal knowledge, except as to those statements where I have expressly stated that my opinion is based upon information and belief, and I believe all such statements, and the information upon which they are based, to be true.
2. I am a Belgian citizen, born on 30 January 1962.
3. I am currently Professor of European Union (“EU”) Law, and Executive Dean of the Faculty of Laws at University College London (“UCL”). I am also Academic Director of the UCL European Institute. Before joining UCL (in 2012), I taught at the Universities of Ghent and
[Page 2]
of Brussels (Belgium), and was Professor of European Law at King's College London, where I directed the Centre of European Law (between 1998 and 2012). I also worked as a référendaire (law clerk) in the Chambers of Advocate General Francis G. Jacobs, at the European Court of Justice (the “ECJ”), between 1994 and 1998. My academic interests revolve around EU law generally, and EU external relations law (including the interaction between EU law and public international law) more particularly. I have published extensively in this field. I am the sole author of a leading monograph, widely used in the field: EU External Relations Law (2nd ed, Oxford EU Law Library, 2011). I am further co-editor of the Oxford EU Law Library, a series of monographs published by Oxford University Press and aimed at both academics and practitioners. I am co-founder and co-editor of an open-access journal, Europe and the World – A Law Review (UCL Press). I have also taught international trade law and have written on questions of public international law, particularly those connected to the relationship between EU law and public international law. Next to my academic work, I regularly advise barristers and law firms in the context of litigation involving questions of EU law. Mostly that litigation is before the ECJ. The most noteworthy cases in which I have been involved are Kadi, on the relationship between United Nations (“UN”) law and EU law in the context of counterterrorism, and Wightman and Others, on the question of whether the United Kingdom could unilaterally revoke its notification to withdraw from the European Union.1,2 In both cases the ECJ fully endorsed the arguments I contributed to, and rejected the positions of the EU Council of Ministers and of the European Commission.
1 While Professor Bungenberg submitted two expert reports (CEF Dkt. 26-20 and Greentech Dkt. 35-2), I will refer to these two reports collectively as the “Expert Report of Marc Bungenberg”. Attachments to Professor Bungenberg's declaration are hereinafter referred to as “Bungenberg Exhibits”. ↩
2 Joined Cases C-402/05 P Kadi and Al Barakaat v Council and Commission EU:C:2008:461 (“Bungenberg Exhibit 52”); Case C-621/18 Andy Wightman and Others v Secretary of State for Exiting the European Union EU:C:2018:999 (“Exhibit 2”). ↩
[Page 3]
4. My full CV, together with a list of my publications, is attached hereto as Exhibit 1.
5. I am being compensated at a rate of GBP 500 per hour to prepare this expert declaration and, if required, to testify in this case.
6. I have no familial or business relationship or affiliation with any of the parties in the above-captioned matter. I have never provided legal advice to them or represented them in any capacity.
7. I have been asked by counsel for the Petitioners in the above-consolidated action, CEF Energia, B.V. v. The Italian Republic and Greentech Energy Systems A/S (now known as Athena Investments A/S), Novenergia General Partner S.A. (acting as liquidator of Novenergia II Energy & Environment (SCA) SICAR) Novenergia II Italian Portfolio v. The Italian Republic, to provide an expert declaration on certain legal issues in this action by CEF Energia, B.V. (“CEF”) and Greentech Energy Systems A/S (now known as Athena Investments A/S), Novenergia General Partner S.A. (acting as liquidator of Novenergia II Energy & Environment (SCA) SICAR) Novenergia II Italian Portfolio (“Greentech”) to enforce the arbitral awards entered in their favour against the Italian Republic (“Italy”) in CEF v. Italy and Greentech v. Italy (the “Awards”).3 In particular, I have been asked to give my expert opinion on the following issues:
3 S.D.N.Y. Case No. 1:19-cv-09153, Dkt. 1-1; S.D.N.Y. Case No. 1:19-cv-04398, Dkt. 1-1. ↩
[Page 4]
8. In preparing my expert declaration counsel for CEF and Greentech provided me with the parties' submissions, the expert declarations of Professor Bungenberg and the EC's amicus brief submitted in this matter. I reviewed these documents when preparing this declaration.
9. The EC claims that the ECT was never intended to apply in an intra-EU context, and that it was concluded by the EU as a single block. However, neither the terms of the ECT (interpreted in their context and in light of the ECT's objective and purpose) nor the EU's own practice in the field of so-called mixed agreements confirms this proposition. Nor have any international tribunals, duly set up pursuant to the ECT's provisions, accepted this proposition; quite the contrary, each relevant tribunal has rejected challenges to its jurisdiction to hear investment arbitration cases in an intra-EU context.
10. Professor Bungenberg and the EC make the extraordinary and wholly unwarranted claim that so-called intra-EU arbitration under the Energy Charter Treaty (hereafter “ECT”) is contrary to EU law, because the judgment of the EU Court of Justice (hereafter “ECJ”) in Achmea governs such arbitration.4 In Achmea the ECJ ruled that EU law precludes intra-EU investment arbitration of the kind included in the Dutch-Slovak bilateral investment treaty (“BIT”). I disagree fundamentally with that claim for the following reasons.
11. First, the ECT is an agreement concluded by the EU itself, together with its Member States and a number of so-called third countries (the EU law term for non-member States). It is
4 Expert Report of Marc Bungenberg, ¶ 59; EC Brief, Section C and passim. ↩
[Page 5]
fully in force, and an integral part of EU law. The EU's membership of the ECT is a major and crucial difference with an intra-EU BIT purely between two EU Member States. Such a BIT sits outside the EU legal order. Moreover, in Achmea the ECJ expressly distinguished intra-EU BITs between Member States from agreements concluded by the EU itself.
12. Second, no EU court has, to the best of my knowledge, ever established any incompatibility between the ECT and EU law. In fact, as the ECT is an agreement concluded by the EU, only the ECJ has jurisdiction to establish such an incompatibility, and to find that the act through which the EU concluded the ECT is invalid under EU law. The ECJ has never done so. This means that the ECT is not only fully in force on the international plane, but that it is also an integral and valid part of EU law as it presently stands.
13. Third, Professor Bungenberg's and the EC's reliance on Achmea is entirely analogical, and upon closer scrutiny the analogy completely breaks down, for the following reasons:
[Page 6]
14. For those reasons it is my firm opinion that the ECT investment arbitration system is compatible with EU law, including in those cases where an investor from an EU Member State has recourse to that system against another Member State. However, even if there were any such incompatibility, it could not affect the validity of any actual ECT arbitration under public international law; and international law is the law governing ECT arbitrations, not EU law.
15. First, notwithstanding the fact that EU law is founded in international treaties, EU law and public international law are distinct in the sense that rules and principles of EU law are merely conventional and are not part of the general body of public international law, which includes customary international law and general principles. EU law is therefore distinct from intra-EU BITs between Member States.
16. Second, the ECJ's ruling in Achmea on intra-EU investment arbitration is confined to EU law, and does not purport to govern the conflict between EU law and public international law, of which the relevant BITs form part. Nor could it be otherwise, as the ECJ's jurisdiction
5 Opinion 1/17 re CETA EU:C:2019:341 (“Bungenberg Exhibit 60”). ↩
[Page 7]
does not go beyond EU law. The ECJ has no jurisdiction to rule on the validity of those BITs on the international plane and under international law.
17. Third, contrary to what Professor Bungenberg and the EC suggest, the ECJ has never stated that EU law has primacy over public international law; indeed it has confirmed that there is no such primacy. The principle of the primacy of EU law governs exclusively the relationship between EU law and the domestic (or municipal) laws of the EU Member States.
18. Fourth, the judgment in Achmea does no more than establish an incompatibility between EU law and the system of investment arbitration in BITs of the type in issue in that case. The judgment means that the EU Member States are under an EU law obligation to remove that incompatibility, by amending or terminating those BITs. As long as this has not happened, the conflict may persist, but it cannot, on the international plane, affect the validity of those BITs and of any arbitral awards which apply them.
19. The claims in Professor Bungenberg's Expert Declaration and in the EC Brief are also extraordinary because the awards against Italy whose enforcement is sought do not include any substantive interpretation or application of EU law. They do not therefore affect the mutual trust between EU Member States, as a principle of EU law. They also do not interfere with the preliminary rulings system as the essential instrument for ensuring that EU law is respected and correctly applied in the relations between EU Member States. And they do not interfere with the ECJ's supreme authority to interpret EU law.
20. It should further be pointed out that that the EC's position in these proceedings is not supported by all EU Member State Governments, as Exhibits E, G and F in the EC Brief show. Five Member State Governments are of the view that it is at this stage “inappropriate, in the absence of a specific judgment on this matter, to express views as regards the compatibility with
[Page 8]
Union law of the intra-EU application of the Energy Charter Treaty” (Exhibit F). One Member State Government considers that such intra-EU application is compatible with EU law (Exhibit G). These are declarations made by the Representatives of the Governments of the Member States, not by the Council of Ministers of the European Union. I have seen no evidence that the Council itself has adopted any position in the matter.
21. Nor does the EC's Notice of 19 July 2018 serve as binding authority concerning the scope and applicability of Achmea or any other CJEU judgment. While the EC claims to understand the “correct” interpretation of the Achmea judgment, its opinion is binding neither on the EU Member States, the CJEU, or this Court. The EC's Notice is simply a policy declaration urging EU Member States to “formally terminate their intra-EU BITs,”6 but does not amount to a legislative directive. Moreover, the fact that such “urging” is necessary underscores that such agreements entered into by Member States are not automatically void as a result of the Achmea decision.
22. In fact, under principles of international law, neither of these statements (the January 2019 statements by certain EU Member States and the EC's July 2018 Notice) did (or could) modify Italy's obligations under the ECT. As the International Law Commission has explained, where a State makes a statement which “purports to exclude or modify the legal effect”
6 Communication from the Commission to the European Parliament and the Council: Protection of intra-EU investment (2018), https://ec.europa.eu/transparency/regdoc/rep/1/2018/EN/COM-2018-547-F1-EN-MAIN-PART-1.PDF (“Exhibit 3”). ↩
[Page 9]
of provisions of a treaty after the State has ratified that treaty, that statement can be treated as nothing more than an “interpretative declaration”7 that “does not modify treaty obligations.”8
23. Thus, while the EC is fully in its rights when it claims to be representing the EU in these proceedings, because it is generally the EC which represents the EU in any legal proceedings. However, that does not mean that its understanding of the law is of necessity shared by the other EU institutions, including the ECJ. The EC claims special expertise,9 but that does not mean that its arguments should be subjected to any lesser scrutiny. As I pointed out above, I have been involved in some high-profile ECJ litigation on matters concerning the relationship between EU law and public international law,10 and in both cases the arguments of the parties I counseled prevailed over those submitted by the EC.
24. It is lastly extraordinary, and frankly wholly unpersuasive, to ask this US Court to refuse the enforcement of an ECT award, valid under that Treaty and under international law, on the basis of a wholly presumptive and frankly implausible violation of EU law – a violation which no court in the EU has ever established. The ultimate paradox of this request is that this Court is asked to do exactly what the principle of the autonomy of EU law does not permit: To interpret and apply EU law, outside the EU judicial system.
25. As explained above, I consider it may assist the Court in following my analysis to provide some background information on: (i) the relationship between EU law and international
7 International Law Commission, Guide to Practice on Reservations to Treaties (2011), ¶ 1.1 (defining “interpretive declaration” as “a unilateral statement, however phrased or named, made by a State or an international organization, whereby that State or that organization purports to specify or clarify the meaning or scope of a treaty or of certain of its provisions.”) (“Exhibit 4”). ↩
8 Id. at ¶ 1.2. ↩
9 EC Brief, p. 8. ↩
10 See ¶ 3 above. ↩
[Page 10]
law; and (ii) the operation of agreements to which the EU and its Member States are all parties (so called mixed agreements), such as the ECT. Those topics are addressed in turn in this section.
26. The European Union is an organisation created by the conclusion of a series of international treaties – chiefly the Treaty on European Union (“TEU”) and the Treaty on the Functioning of the European Union (“TFEU”). In that sense, the law of the European Union (“EU law”) has its foundations in public international law, particularly treaty law. The EU respects international law. Article 3(5) TEU provides that the EU shall, in its relations with the wider world, contribute to “the strict observance and the development of international law”.11 The EU has legal personality,12 which encompasses international legal personality. It therefore has the capacity to act under international law, particularly by means of the negotiation and conclusion of international agreements.13 Article 216(2) TFEU provides that “[a]greements concluded by the Union are binding upon the institutions of the Union and on its Member States”. This provision ensures that the EU complies with its international commitments, and respects international law. The EU institutions are bound by any agreements the EU has concluded, including in their legislative function. The ECJ has confirmed, on numerous occasions, that acts of the institutions, including legislative acts, need to comply with the EU's international obligations.14 The Member States are equally bound, simply because agreements concluded by the EU are an integral part of
11 See also Art. 21(2) TEU (“Exhibit 5”). ↩
12 Art. 47 TEU (“Exhibit 5”). ↩
13 See Art. 216(1) TFEU (“Exhibit 6”). EU law uses the term “agreements” in a generic sense, as encompassing any conventional instruments such as treaties, conventions, agreements, etc. ↩
14 See e.g. Case C-344/04 The Queen ex parte IATA v Department for Transport EU:C:2006:10 ¶ 35 (“Exhibit 7”); Case C-308/06 Intertanko EU:C:2008:312 ¶ 42 (“Exhibit 8”). ↩
[Page 11]
EU law.15 These points are significant for the present case because the ECT is an agreement concluded by the EU (together with its Member States),16 in contrast with the BIT which was in issue in the Achmea.
27. The fact that international agreements concluded by the EU prevail over acts of the EU institutions, including legislative acts, does not mean that such agreements take priority over the EU's founding treaties (TEU and TFEU) – at least not as a matter of EU law. In this respect the TEU and the TFEU are functionally equivalent to most States' constitutions, which do not allow the State's institutions to conclude international agreements which derogate from, or conflict with constitutional norms. The ECJ has established, in a very limited number of cases, that the provisions of a particular international agreement violate EU constitutional norms.17 In such cases it will annul the EU's internal act by which the agreement was concluded, or declare that act invalid.18 It cannot, however, annul the agreement itself, quite simply because that is beyond its jurisdiction.19 The ECJ's jurisdiction is confined to EU law. International agreements concluded by the EU, or non-conventional international norms such as general principles and customary international law which are relevant to the EU's international activities, can be interpreted and relied upon by the ECJ. However, the ECJ cannot, as the court of one of the parties to an international agreement, declare that agreement invalid as a matter of public international law. This is in conformity with Article 46 of the Vienna Convention on the Law of Treaties (hereafter
15 Case 181/73 Haegeman EU:C:1974:41 ¶ 5 (“Bungenberg Exhibit 7”). ↩
16 See infra § III(B). ↩
17 See e.g. Case C-122/95 Germany v Council EU:C:1998:94 (“Bungenberg Exhibit 16”). ↩
18 Those two remedies – annulment and a declaration of invalidity – are equivalent. The difference lies in the judicial procedure by which the agreement is challenged before the ECJ: either in a (direct) action for annulment (Art. 263 TFEU); or on a reference from a Member State court asking whether the act concluding an agreement is invalid (indirect action – Art. 267 TFEU). ↩
19 Case C-327/91 France v Commission EU:C:1994:305 ¶¶ 13-17 (“Exhibit 9”). ↩
[Page 12]
“VCLT”).20 Where a breach of the EU treaties is found, the EU institutions need to either renegotiate the EU's commitments, or terminate them, so as to remove the breach.
28. In order to avoid such difficulties, the drafters of the EU Treaties have, with great foresight, created a special procedure in Article 218(11) TFEU. This provision allows any Member State, as well as the European Parliament, the Council or the Commission, to ask the ECJ whether an agreement which the EU envisages to conclude, but has not yet concluded, “is compatible with the Treaties”.21 As the ECJ has stated, this procedure aims “to forestall complications which would result from legal disputes concerning the compatibility with the [EU Treaties] of international agreements binding upon the Community [now the EU]”.22 It has been used for CETA, resulting in Opinion 1/17, an ECJ ruling which is highly relevant for the purpose of answering some of the objections to the enforcement of the Awards in the present case.23 It is also worth noting that Article 218(11) TFEU confirms that the EU Treaties cannot, on the international plane, override any treaties or agreements the EU concludes. It does so by stating that “[w]here the opinion of the Court is adverse, the agreement envisaged may not enter into force unless it is amended or the Treaties are revised”.24 In other words, any incompatibilities need to be removed before the agreement can be concluded, and this may include a revision of the EU Treaties. This is necessary to avoid a conflict between the EU's international commitments and the EU Treaties. Such a
20 Art 46 provides that “1. A State may not invoke the fact that its consent to be bound by a treaty has been expressed in violation of a provision of its internal law regarding competence to conclude treaties as invalidating its consent unless that violation was manifest and concerned a rule of its internal law of fundamental importance.” VCLT (“Exhibit 10”). ↩
21 Art. 218(11) TFEU (“Exhibit 6”). ↩
22 Opinion 1/75 re Understanding on a Local Cost Standard EU:C:1975:145, at 1360 (“Exhibit 11”). ↩
23 Opinion 1/17 re CETA EU:C:2019:341 (“Bungenberg Exhibit 60”). ↩
24 Art. 218(11) TFEU (“Exhibit 6”). ↩
[Page 13]
conflict could be difficult to resolve if the third countries concerned were not amenable to renegotiating those commitments.
29. The EU Member States continue to be international actors, capable of concluding international agreements with other States or with international organizations, independently from the EU. This includes agreements with other Member States. A good example is double-taxation treaties.25 It is only in matters covered by so-called exclusive EU treaty-making competences that the Member States have lost this capacity.26 Trade policy (the “common commercial policy”) is the most significant area of exclusive EU competence.27
30. The EU Member States must comply with their EU law obligations when concluding international agreements, acting on their own. There is in this sense no difference between what the Member States do within the four corners of their domestic laws, and what they commit to internationally. The ECJ may therefore establish that certain provisions in treaties or agreements concluded by one or more Member States violate EU law. The Achmea Judgment is one such instance.28 Again, however, the ECJ is incapable of annulling or invalidating such a treaty or agreement on the international plane. Where a breach is established, the relevant Member State is simply under an EU law obligation to remove the breach, by either renegotiating the relevant terms, or by denouncing or terminating the agreement in issue.
25 For the ECJ's recognition of such treaties, see e.g. Case C-648/15 Austria v Germany EU:C:2017:664 (“Exhibit 12”). ↩
26 See Art. 3 TFEU (“Exhibit 6”). ↩
27 See Arts. 3(1)(a), (e) TFEU (“Exhibit 6”). ↩
28 See also (as regards BITs) Case C-205/06 Commission v Austria EU:C:2009:118 (“Bungenberg Exhibit 23”); Case C-249/06 Commission v Sweden EU:C:2009:119 (“Bungenberg Exhibit 31”); and Case C-118/07 Commission v Finland EU:C:2009:715 (“Bungenberg Exhibit 15”). ↩
[Page 14]
31. A closer reading of the Achmea judgment confirms this: The ECJ only established that the dispute settlement system in the Netherlands-Slovakia BIT was not in conformity with EU law.29 It is therefore completely orthodox that the European Commission has advised the Member States to terminate all intra-EU bilateral BITs: The Achmea judgment, in and of itself, cannot have any effect on the existence and validity of such agreements under international law. For the EU-law incompatibility to be removed, the Member States need to act under international law, by either amending or terminating these BITs. Moreover, the Achmea judgment, delivered pursuant to a preliminary reference from the German Bundesgerichtshof, could not, on its own, invalidate even that particular BIT. All the ECJ did was to interpret EU law, in such a way as to assist the referring court when deciding whether to give effect to the set-aside application before it. The ECJ ruled that the arbitration provisions of an agreement “such as” the Netherlands-Slovakia BIT are in breach of EU law.30 The Bundesgerichtshof then applied that EU law finding to the actual BIT, and more specifically to the application for the set aside of an award made under that BIT.31 It ensured that EU law was complied with, and that the award was not enforced, by ruling that there was no valid agreement to arbitrate.32 That was the German court's ruling, in that particular case, which involved the BIT that had been specifically addressed by the ECJ. It was not the ECJ's ruling, nor could it have been, as the ECJ's jurisdiction did not extend to the BIT as such.
29 See Achmea Judgment (“Bungenberg Exhibit 31”) ¶ 7(i), in operative part. ↩
30 Id. ↩
31 The Bundesgerichtshof was empowered to consider the application for set aside because the arbitration had been conducted under the UNCITRAL Rules and was seated in Frankfurt, Germany. See also Achmea Judgment (“Bungenberg Exhibit 31”). ↩
32 Bundesgerichtshof [BGH] [Federal Court of Justice] Oct. 31, 2018, I ZB 2/15, Slovak Republic v. Achmea B.V. (Ger.) (“Bungenberg Exhibit 67”). ↩
[Page 15]
32. The fact that the EU's founding Treaties, or EU law more generally, cannot invalidate treaties or agreements which the Member States have concluded, on their own, is confirmed by Article 351 TFEU. That provision states, in relevant part:
The rights and obligations arising from agreements concluded before 1 January 1958 [the date of entry into force of the original EEC Treaty] or, for acceding States, before the date of their accession, between one or more Member States on the one hand, and one or more third countries on the other, shall not be affected by the provisions of the Treaties.
To the extent that such agreements are not compatible with the Treaties, the Member State or States concerned shall take all appropriate steps to eliminate the incompatibilities established.
As can be seen, this is not a provision which establishes a primacy rule in favor of EU law, contrary to what Professor Bungenberg suggests.33 As Advocate General Mischo established in Commission v Portugal,34 the first paragraph of this provision is merely declaratory, in the sense that it confirms the international law principle of pacta sunt servanda. It is obvious that the EU Treaties cannot modify the obligations which Member States have entered into towards third countries.
33. The EC claims that “the Court of Justice has held in a number of judgments (...) that such treaties do not apply within the EU if they are contrary to any rule of EU law (...).”35 This misrepresents the ECJ's findings in these cases. First, those findings are confined to EU law, and do not extend to the validity of such agreements on the international plane. Indeed, the second
33 See Expert Report of Marc Bungenberg, ¶ 36. ↩
34 Opinion of Mischo AG in Case C-62/98 Commission v Portugal EU:C:1999:509 ¶ 56 (“Exhibit 13”). This statement has strong judicial authority, in light of the role and function of Advocates General at the ECJ: they are full members of the Court, and their role is to deliver reasoned opinions in cases pending before the Court. They are completely independent, and the Opinions of Advocates General are followed by the Court in a significant majority of cases. ↩
35 See EC Brief, p. 21, with case law cited. ↩
[Page 16]
subparagraph of Article 351 TFEU confirms that, in case of any incompatibilities, specific action by the states involved is needed to cure them. Second, all of the relevant cases concerned international law rights which a Member State claimed to have, pursuant to an agreement with another Member State, and which were in conflict with an EU law obligation imposed on the rights-claiming Member State. The ECJ decided that, in such cases, that Member State would be required to refrain from exercising those rights, and thereby comply with its EU law obligation without this interfering with the rights of any third countries. The ECT, however, confers rights on private investors as against States. Those private investors are third parties, who benefit from a multilateral treaty which has the EU, its Member States, and a series of third countries as Contracting Parties. This circumstance materially distinguishes the present scenario from the cases discussed by the EC in its amicus brief.
34. It may also assist this Court to say a few words about the reasons for so-called mixed agreements (such as the ECT), and about some of the characteristics and effects of such agreements.
35. Mixed agreements, i.e. agreements which have the EU, its Member States, and third countries as contracting parties, are a prevalent phenomenon in the EU's foreign affairs. The need for such agreements arises because the EU is not a sovereign State with full and complete treaty-making powers. The EU's competences are limited by the principle of conferral: “the Union shall act only within the limits of the competences conferred upon it by the Member States in the Treaties to attain the objectives set out therein. Competences not conferred upon the Union in the Treaties remain with the Member States”.36 This principle extends to the EU's external action (as
36 Art. 5(2) TEU (“Exhibit 5”). ↩
[Page 17]
it is called), including its treaty-making activities. A mixed agreement is needed where the matters covered by the agreement do not all come within the EU's competence.
36. The ECT is such a mixed agreement. At the time of its conclusion, it was considered that not all of its provisions came within the EU's exclusive external competences. That continues to be the case, notwithstanding the further development of the EU's competences in the field of energy. It is indeed a fact that, since the conclusion of the ECT, the EU has adopted extensive legislation in energy matters, regulating its internal energy market. It has also been conferred specific legislative competences.37 However, those competences are not exclusive of Member States' competences, but shared with them (Arts. 4(2)(a), (i) TFEU). What that means is that the EU is able to claim an exclusive treaty-making competence in the field of energy only where an international agreement affects EU legislation in such a way as to trigger Article 3(2) TFEU.38 It is clear that a range of aspects of international energy policy do not come within the EU's exclusive treaty-making competence, and that the ECT, if it was concluded today, would again have to be concluded as a mixed agreement. Moreover, the ECJ recently established that provisions in an international agreement on investor-State dispute settlement which have the effect of excluding the jurisdiction of the ordinary courts of the Member States, in favour of international arbitration, are not within EU competence.39 It made that finding in its Opinion on the EU-Singapore Free Trade Agreement. The dispute settlement provisions of that agreement in the sphere of investment
37 See Art. 194 TFEU (“Exhibit 6”). ↩
38 Art. 3(2) TFEU provides that “[t]he Union shall also have exclusive competence for the conclusion of an international agreement when its conclusion is provided for in a legislative act of the Union or is necessary to enable the Union to exercise its internal competence, or in so far as its conclusion may affect common rules or alter their scope” (“Exhibit 6”). In Case C-66/13 Green Network SpA EU:C:2014:2399 (“Exhibit 14”), the ECJ established exclusive EU competence on the basis of EU Directive 2001/77, as regards the purchase of green electricity certificates. ↩
39 Opinion 2/15 re EU-Singapore FTA EU:C:2017:376 ¶ 292 (“Exhibit 15”). ↩
[Page 18]
protection are, in this respect, equivalent to those of the ECT. This means that the dispute settlement provisions of the ECT continue to be a matter of mixed competence.
37. When the EU and its Member States conclude a mixed agreement, there are, in theory, a number of options open to them. They could identify, in the agreement itself, the parts or provisions which are concluded by, respectively, the EU, and by the Member States. That option is rarely used, if at all. They could also indicate the scope of their respective competences in some other way, for example by means of a “declaration of competences”. That is an option which the EU and the Member States regularly employ.40 They have done so, for example, when concluding the UN Convention on the Law of the Sea (UNCLOS). It is also open to the EU and the Member States to negotiate a mixed agreement which operates in an essentially bilateral way. This is the case for most of the free-trade agreements which the EU concludes. Such agreements usually provide that they apply, on the one hand, in the territory of the third country (say Canada), and on the other, in the territory of the EU and its Member States.41 The EU could also negotiate a “disconnection clause,” which would assert the non-applicability of certain provisions to treaty disputes exclusively between EU Member States. Such a disconnection clause may be used where the EU considers there to be potentially overlapping obligations between an EU Member States’ treaty obligations and its EU obligations, and therefore a “disconnection clause” is needed to express (and ensure) the primacy of EU law.
40 See M Cremona, “Disconnection Clauses in EU Law and Practice”, in C Hillion and P Koutrakos (eds), Mixed Agreements Revisited (Oxford and Portland, Oregon: Hart Publishing, 2010) at 160–186 (“Exhibit 16”). ↩
41 See e.g., Art. 1.3 of CETA (“Exhibit 17”); see Council Decision (EU) 2017/38 of 28 October 2016 on the signing on behalf of the European Union of the Comprehensive Economic and Trade Agreement (CETA) between Canada, of the one part, and the European Union and its Member States, of the other part (EU OJ 2017 L 11 at 1, italics added) (“Exhibit 18”). ↩
[Page 19]
38. None of those options were used in the case of the ECT. Giving meaning to the ordinary meaning of the terms used, in their context and in the light of a treaty's object and purpose, is the first customary international law rule of treaty interpretation.42 A simple reading of the terms of the ECT, in their ordinary meaning, confirms that all of the Contracting Parties are bound by all of the ECT's provisions, and have entered into obligations towards all other Contracting Parties. In other words, nothing suggests that the ECT does not apply in an intra-EU context.
39. Contrary to what Professor Bungenberg appears to claim, there is no inherent conflict between the application of EU law, particularly internal market law, in the relations between EU Member States, and the application of the ECT. The objectives of the ECT are broadly aligned with the basic policy principles that guide EU economic regulation.43 The ECT provisions on investment protection are complementary to EU internal market law, in the sense that they provide an investor with additional protection. Between the Member States, the right of establishment, the freedom to provide services and the free movement of capital offer significant guarantees against discriminatory treatment and unwarranted regulatory restrictions. Those basic freedoms also underpin sector-specific legislation. EU competition and state aid policy ensure that distortions of competition caused by anti-competitive behaviour or government subsidies are combated. But none of this means that there is no room left for the ECT provisions, particularly those on investor protection. The additional protection of those provisions is complemented by strong remedies, particularly as regards financial compensation. Those remedies are arguably stronger than those for which EU law provides: It is only where a Member State has committed a
42 See Art. 31(1) of the Vienna Convention on the Law of Treaties, which is regarded as codifying the relevant customary international law rules. (“Exhibit 10”). ↩
43 See e.g. Electrabel S.A. v. Republic of Hungary, ICSID Case No. ARB/07/19, Award, 25 November 2015 (“Bungenberg Exhibit 65”). ↩
[Page 20]
“sufficiently serious” breach of EU law that an investor from another Member State is entitled to compensation (“Member State liability”).44 It is not at all clear on what basis the Claimants in this case would have been able to obtain compensation in the Italian courts pursuant to the EU law principles governing Member State liability.
40. As a matter of policy, it is of course open to the EU to aim at the exclusive application of EU law in an intra-EU investment context. The EU can ask the Member States not to conclude inter se agreements, and to terminate the existing BITs; and it could ensure that multilateral agreements on investment protection do not apply within the European Union, even when they are mixed. A simple disconnection clause – i.e. a clause confirming that the agreement does not create rights and obligations in the relations between the Member States – would achieve that objective. But the ECT does not contain such a clause. The glaring absence of a disconnection clause is striking and undermines the claim that the ECT cannot apply in an intra-EU context. The EC's argument that disconnection clauses have no bearing on intra-EU relations is wide of the mark. Where such a clause is inserted, it clarifies to all parties, including the EU and its Member States, that the agreement does not create inter se rights and obligations.
41. The EC advances a whole series of reasons for why the ECT should be interpreted as not applying in an intra-EU context and as not creating international law obligations between EU Member States, despite the fact that they are full Contracting Parties. None of those reasons withstand closer scrutiny, and none have been accepted by any arbitral tribunal.
44 See e.g., Joined Case C-6/90 and C-9/90 Francovich and Bonifaci EU:C:1991:428 (“Bungenberg Exhibit 55”). ↩
[Page 21]
42. The EC argues that the EU and its Member States acted as a single block in the ECT negotiations.45 That may be true. It is in fact an EU obligation for the EU and its Member States to aim to be unified in the positions they take up in a multilateral negotiation.46 But that does not mean that the resulting agreement cannot create inter se obligations. There are plenty of examples of agreements which create such obligations. Perhaps the most prominent one is the UN Convention on the Law of the Sea (UNCLOS). No one has ever suggested that the provisions of that Convention, e.g. those defining territorial waters and exclusive economic zones, are inapplicable in relations between two Member States. In fact, Ireland was found to have violated EU law by not bringing a case, claiming breach of UNCLOS, before the ECJ, rather than before an UNCLOS tribunal.47 The point I make here is not about the ECJ's jurisdiction in such cases (which results from a combined reading of certain UNCLOS provisions and Article 344 TFEU); but rather about the fact that the provisions of UNCLOS apply in an intra-EU context.
43. None of the authorities to which the EC refers in its brief says otherwise. The literature listed in footnote 9 of the EC Brief is not concerned with the ECT; the monograph by Bleckmann in fact predates the ECT's creation. The references do not concern the question whether an agreement governs intra-EU relations. The EC also appears to suggest that in Ruling 1/78 the ECJ found that mixed agreements do not create inter se obligations, but nothing in that Ruling confirms this in any shape or form.48
45 See EC Brief, p. 6-7. ↩
46 See Ruling 1/78 [1978] ECR 2151 ¶ 36 (“Exhibit 19”). ↩
47 Case C-459/03 Commission v Ireland EU:C:2006:345 (“Bungenberg Exhibit 41”). ↩
48 See EC Brief, pp. 7-8, with reference to Ruling 1/78 [1978] ECR 2151 ¶ 36 (“Exhibit 19”). In fact, the ECJ's statement in that paragraph comes close to recognizing that the convention in issue in those proceedings also applied in an intra-EU context. ↩
[Page 22]
44. Contrary to what the Commission claims, no arguments can be derived from the fact that, in the ECT, the EU is characterized as a “Regional Economic Integration Organization”.49 That is simply a traditional characterization of international organizations like the EU. It does not mean that, where the EU and its Member States conclude an agreement which contains this denomination, the agreement cannot create obligations between EU Member States.
45. The EC's argument that the EU Treaties “do not permit EU Member States (or, indeed, the EU itself) to modify or replace EU law by an international treaty such as the ECT”50 misses the point. The Awards in issue here did not interpret the ECT as modifying or replacing EU law. The provisions of the ECT and EU internal market law are perfectly capable of co-existing in intra-EU relations.
46. The EC also mischaracterizes the ECJ judgment in Green Network.51 That judgment did not establish EU exclusive competence “for renewable energy”. The judgment is confined to questions of certification of green electricity, and does not established a broader exclusive EU competence.
47. To conclude this section, there is nothing in the terms of the ECT which suggests that the EU Member States did not enter into inter se obligations. Those States are full Contracting Parties, and the provisions of the ECT are aligned with and complement EU internal market law.
48. In this Section, I address whether the Achmea Judgment precludes intra-EU arbitration under the ECT. In my opinion, it does not. First, the Achmea Judgment does not address
49 See EC Brief, p. 8. ↩
50 See id., p. 9. ↩
51 Case C-66/13 Green Network SpA v Autorità per l'energie elettrica e il gas EU:C:2014:2399 (“Exhibit 14”); see EC Brief, p. 9. ↩
[Page 23]
the ECT. Second, even if the Achmea Judgment were to prohibit intra-EU arbitration under the ECT, that would not affect the consent to arbitration provided by EU Member States in Article 26 of the ECT under international law. It is international law which governs the jurisdiction of a tribunal established under the ECT. Therefore, any arbitration agreement formed between an EU Member State and an EU investor under the ECT would still be valid.
49. My analysis is divided as follows: First, I address the scope of the Achmea Judgment's findings; second, I discuss Opinion 1/17 on CETA, which has clarified the position of investor-State arbitration in EU law; third, in light of the findings of both that ruling and the Achmea Judgment, I explain my view that intra-EU arbitration under Article 26 of the ECT remains possible under EU law; and fourth, I explain why, even if EU law were to preclude intra-EU arbitration, the consent to arbitration provided by EU Member States in Article 26 would remain valid (which is because international law – not EU law – is the applicable law).
50. The Achmea Judgment is one of the latest ECJ rulings in a series in which the ECJ has imposed significant restrictions on the extent to which (a) the Member States can participate in, and (b) the EU can conclude, an agreement setting up systems of international dispute settlement which concern the interpretation or application of EU law.52 Those restrictions are aimed at safeguarding what the Court describes as the “autonomy of EU law”: An autonomy both from the laws of the Member States and from international law. This autonomy includes the protection of the essential characteristics of EU law and of the essential powers of the EU
52 See in particular Opinion 1/91 re the EEA Agreement EU:C:1991:490 (“Bungenberg Exhibit 61”); Opinion 1/00 re the ECAA Agreement EU:C:2002:231 (“Bungenberg Exhibit 57”); Opinion 1/09 re Unified Patent Convention EU:C:2011:123 (“Bungenberg Exhibit 58”); and Opinion 2/13 re Accession to the ECHR EU:C:2014:2454 (“Bungenberg Exhibit 63”). ↩
[Page 24]
institutions. As far as the ECJ itself is concerned, it has highlighted its own role of ensuring the observance of EU law.53 The preliminary rulings system is aimed at guaranteeing such observance, particularly in the laws of the Member States. Pursuant to Article 267 TFEU any court or tribunal of a Member State may refer questions of interpretation or validity of EU law to the ECJ, and courts or tribunals against whose decisions there is no judicial remedy under national law (i.e. supreme and highest courts) are obliged to refer questions of EU law. It is further worth noting that the principles of the direct effect and primacy of EU law oblige national courts and tribunals to give domestic effect to EU law. Next to the preliminary-rulings procedure, Article 344 TFEU instructs the Member States “not to submit a dispute concerning the interpretation or application of the Treaties to any method of settlement other than those provided for therein”.
51. In Achmea the ECJ further expanded its case law on the autonomy of EU law. It is important to highlight that the case concerned a BIT between the Netherlands and Slovakia, concluded before Slovakia's accession to the EU; and that Article 8(6) of this BIT provides that the arbitral tribunal “shall decide on the basis of the law”, taking into account inter alia “the law in force of the Contracting Party concerned” and “any other relevant agreements between the Contracting Parties”.
52. In the Achmea Judgment, the ECJ first reiterated a number of general principles and past judicial statements about the autonomy of EU law and the principle of mutual trust. It then applied those principles to the BIT in issue, and divided its analysis in three parts.
53. First, and crucially, the ECJ examined whether an arbitral tribunal set up pursuant to Article 8 of the BIT was liable to rule on the interpretation or application of EU law.54 It found
53 See Art. 19(1) TEU (“Exhibit 5”). ↩
54 Achmea Judgment ¶¶ 39–42 (“Bungenberg Exhibit 31”). ↩
[Page 25]
that EU law is both “the law in force of the Contracting Party” (as EU law is an integral part of the domestic laws of the Member States) and constitutes an agreement between the Parties. It followed that “on that twofold basis the arbitral tribunal ... may be called on to interpret or indeed to apply EU law, particularly the provisions concerning the fundamental freedoms, including freedom of establishment and free movement of capital”.55 This sentence is particularly significant, in light of the differences between the definition of the applicable law in the BIT at issue and that definition in the ECT. I discuss that difference further below. Moreover, the insertion of the adverb “indeed” is noteworthy, as it shows that the ECJ was particularly concerned about a non-EU tribunal applying EU law to the relations between two Member States, and outside the EU judicial system. The ECT of course does not call for the application of EU internal market law; nor do the Awards in issue in this case address EU internal market law in any way.
54. In the second part of its reasoning, the ECJ established that the BIT arbitral tribunal is not situated within the EU judicial system: It is not a court or tribunal of the Member States, capable of making a reference to the ECJ pursuant to Article 267 TFEU.56
55. In the third part the ECJ examined to what extent an arbitral award giving effect to the BIT was “subject to review by a court of a Member State, ensuring that the questions of EU law which the tribunal may have to address can be submitted to the Court by means of a reference for a preliminary ruling”.57 The ECJ established that this was not the case, in light of the limits which German law imposed on such review in the case at hand. The ECJ concluded by finding that the dispute settlement system of this BIT could not guarantee the full effectiveness of EU law.
55 Id. ¶ 42. ↩
56 Id. ¶¶ 43-49. ↩
57 Id. ¶ 50. ↩
[Page 26]
But the ECJ did not stop there. It added two further paragraphs, which are crucial for the present case.
56. First, the ECJ recalled the principle that “an international agreement providing for the establishment of a court responsible for the interpretation of its provisions and whose decisions are binding on the institutions, including the [ECJ], is not in principle incompatible with EU law”.58 The EU's external competence “necessarily entail[s] the power to submit to the decisions of a court which is created or designated by such agreements as regards the interpretation and application of their provisions, provided that the autonomy of the EU and its legal order is respected”.59
57. Second, and this finding is worth quoting in full:60
“In the present case, however, apart from the fact that the disputes falling within the jurisdiction of the arbitral tribunal referred to in Article 8 of the BIT may relate to the interpretation both of that agreement and of EU law, the possibility of submitting those disputes to a body which is not part of the judicial system of the EU is provided for by an agreement which was concluded not by the EU but by Member States. Article 8 of the BIT is such as to call into question not only the principle of mutual trust between the Member States but also the preservation of the particular nature of the law established by the Treaties, ensured by the preliminary ruling procedure provided for in Article 267 TFEU, and is not therefore compatible with the principle of sincere cooperation ... .”
58. In other words, the ECJ made a clear distinction between intra-EU BITs such as the one in issue in Achmea, and a system of dispute settlement set up by an agreement which the EU itself has concluded. That distinction means that the Court sought to confine its ruling to intra-EU BITs between Member States, and did not rule on an agreement like the ECT. It cannot sensibly be argued that, in making that distinction, the ECJ was only considering investment protection
58 Id. ¶ 57. ↩
59 Id. ↩
60 Id. ¶ 58 (emphasis added). ↩
[Page 27]
agreements between the EU and third countries, and not agreements with intra-EU application: The ECJ's own Advocate General had pointed out in his Opinion in Achmea that the ECT applies in an intra-EU context.61 The Court was therefore clearly aware of the existence and potential intra-EU application of the ECT and had the opportunity to address it. It nevertheless distinguished Member State BITs from agreements concluded by the EU.
59. It is remarkable that Professor Bungenberg, as well as the EC, hardly mention this distinction.
60. The distinction is in my view explained, not just as an exercise in judicial economy or a desire to confine the ruling to the facts of Achmea, but by the ECJ's emphasis on mutual trust. Where the Member States set up a system of dispute settlement which sits outside the EU Treaties and is nevertheless capable of interpreting and indeed applying EU law, they undermine the principle of mutual trust between them. They undermine the principle that they must trust their respective domestic courts and tribunals to guarantee the full effectiveness of EU law. By contrast, where the EU itself is a contracting party to an agreement, it takes a conscious decision to participate in, and to be bound by a new dispute settlement system. That is not to say that this system cannot be contrary to the autonomy of EU law. However, it does mean that the ruling in Achmea cannot be extended to the ECT, in the absence of a further ECJ ruling which is focused on the particular characteristics of that agreement. There are moreover other, highly relevant differences between the Achmea-type BIT and the ECT, including, for example, the law applicable to disputes under those treaties. I address those differences below.
61 Opinion Advocate General Wathelet in Case C-284/16 Slowakische Republik v Achmea EU:C:2017:699 ¶ 43 (“Exhibit 20”). ↩
[Page 28]
61. It is also important to note that in Achmea the ECJ did not declare the Netherlands-Slovakia BIT invalid under international law. What it established was that “Articles 267 and 344 must be interpreted as precluding a provision in an international agreement concluded between Member States, such as Article 8 of the Agreement (...)”.62 As analysed above, the ECJ's jurisdiction is limited to EU law, and the ECJ fully recognises this in the formulation of its final ruling. The ECJ is incapable of making findings under international law as to the validity of an international agreement between two Member States.
62. The above analysis of the distinction in Achmea between a BIT between Member States to which the EU is not a party, and an investment agreement concluded by the EU, is wholly confirmed and strengthened by the recent ECJ ruling in Opinion 1/17, on the Comprehensive Economic and Trade Agreement between Canada and the EU (CETA).63 That Opinion (pursuant to Article 218(11) TFEU) was requested by Belgium, which raised a number of concerns about the compatibility of CETA's investment protection provisions, and the dispute settlement system which CETA sets up, with EU law. One of those concerns related to the Achmea issue: The extent to which a CETA Tribunal could be called upon to interpret or apply EU law.
63. Article 8.31 CETA defines the applicable law. Article 8.31.1 provides that CETA must be applied, “as interpreted in accordance with the Vienna Convention on the Law of Treaties, and other rules and principles of international law applicable between the Parties”. Article 8.31.2 clarifies:
“The Tribunal shall not have jurisdiction to determine the legality of a measure, alleged to constitute a breach of this Agreement, under the domestic law of a
62 Achmea Judgment ¶ 62 (“Bungenberg Exhibit 31”). ↩
63 Supra note 33. ↩
[Page 29]
Party. For greater certainty, in determining the consistency of a measure with this Agreement, the Tribunal may consider, as appropriate, the domestic law of a Party as a matter of fact. In doing so, the Tribunal shall follow the prevailing interpretation given to the domestic law by the courts or authorities of that Party and any meaning given to domestic law by the Tribunal shall not be binding upon the courts or authorities of that Party.”
64. Belgium argued that these provisions were inadequate for safeguarding the autonomy of EU law.64 It submitted that a CETA Tribunal would still, when ruling on whether an EU measure is compatible with CETA, “be compelled to interpret the effect of that measure”, and would not necessarily be able to rely on a previous ECJ interpretation. Further, a CETA Tribunal would be empowered “to engage in the assessment of issues of substantive law that involve (...) EU primary law”. Belgium further pointed out that a CETA Tribunal could not seek a preliminary ruling from the ECJ.
65. The ECJ did not accept Belgium's objections. Its reasoning shows the limits of its ruling in Achmea, in the sense that the ruling does not extend to agreements which the EU itself concludes, with third countries, provided that the tribunals set up by those agreements do not have the power to interpret and apply EU law. As further analysed below, the ECT is in this respect similar to CETA.
66. In Opinion 1/17 the ECJ first reiterates the relevant principles, revolving around the concept of the autonomy of EU law. It then accepts that the CETA dispute settlement mechanism “stands outside the EU judicial system”.65 However, that “does not mean, in itself, that that mechanism adversely affects the autonomy of the EU legal order”.66
64 Opinion 1/17 re CETA EU:C:2019:341 ¶ 48 (“Bungenberg Exhibit 60”). ↩
65 Id. ¶ 113. ↩
66 Id. ¶ 115. ↩
[Page 30]
67. The ECJ starts its analysis of why that is so by pointing out that the jurisdiction of the EU courts and tribunals (the ECJ itself and Member State courts, as referred to in Article 19 TEU) to interpret and apply an agreement concluded by the EU “does not take precedence over either the jurisdiction of the non-Member States with which those agreements were concluded or that of the international courts and tribunals that are established by such agreements”.67 Even if such agreements may be the subject of preliminary references, “they concern no less those non-Member States and may therefore also be interpreted by the courts and tribunals of those States”.68 The ECJ then adds that it is “precisely because of the reciprocal nature of international agreements and the need to maintain the powers of the Union in international relations that it is open to the Union (...) to enter into an agreement that confers on an international court or tribunal the jurisdiction to interpret that agreement without that court or tribunal being subject to the interpretations of that agreement given by the courts or tribunal of the Parties”.69 EU law therefore does not preclude the setting up of CETA Tribunals (and indeed the CETA Appellate Tribunal). But on the other hand, “since those Tribunals stand outside the EU judicial system, they cannot have the power to interpret or apply provisions of EU law other than those of CETA or to make awards that might have the effect of preventing the EU institutions from operating in accordance with the EU constitutional framework”.70 These findings further articulate the ECJ's fundamental acceptance of EU participation in international agreements providing for binding dispute settlement.
67 Id. ¶ 116. ↩
68 Id. ¶ 117. ↩
69 Id. ↩
70 Id. ¶ 118. ↩
[Page 31]
68. In a subsequent section of the Opinion the ECJ establishes that the CETA Tribunals do not have jurisdiction to interpret and apply rules of EU law other than the provisions of CETA itself.71
69. It first points out that a CETA Tribunal does not have jurisdiction to determine the legality of a measure under the domestic law of a Party.72 In that respect CETA must be distinguished from the draft agreement which was in issue in Opinion 1/09, which included directly applicable Community law (now EU law) in the applicable law. It must also be distinguished from the investment agreement at issue in Achmea: “that agreement established a tribunal that would be called upon to give rulings on disputes that might concern the interpretation or application of EU law”.73 And the ECJ also recalls that Achmea concerned an agreement between Member States, and this must be distinguished from agreements concluded “between the Union and a non-Member State”.74 The principle of mutual trust applies between the Member States, and it implies that Member States accept that all the other Member States comply with EU law, including the right to an effective remedy before an independent tribunal. That principle of mutual trust is not, however, “applicable in relations between the Union and a non-Member State”.75
70. The ECJ then accepts the limitations which CETA places on the extent to which a CETA Tribunal may take the domestic law of the Parties (and therefore EU law) into account.76 The relevant section is worth quoting in full:77
71 Id. ¶¶ 120-136. ↩
72 See Art. 8.31.2 CETA (“Exhibit 17”). ↩
73 Opinion 1/17 re CETA EU:C:2019:341 ¶ 126 (“Bungenberg Exhibit 60”). ↩
74 Id. ¶ 127. ↩
75 Id. ¶ 129. ↩
76 See Art. 8.31.2 CETA (“Exhibit 17”), supra ¶ 23. ↩
77 Opinion 1/17 re CETA EU:C:2019:341 ¶ 131 (“Bungenberg Exhibit 60”). ↩
[Page 32]
“Those provisions serve no other purpose than to reflect the fact that the CETA Tribunal, when it is called upon to examine the compliance with the CETA of the measure that is challenged by an investor and that has been adopted by the investment host State or by the Union, will inevitably have to undertake, on the basis of the information and arguments presented to it by that investor and by that State or by the Union, an examination of the effect of that measure.
That examination may, on occasion, require that the domestic law of the respondent Party be taken into account. However, as is stated unequivocally in Article 8.31.2 of the CETA, that examination cannot be classified as equivalent to an interpretation, by the CETA Tribunal, of that domestic law, but consists, on the contrary, of that domestic law being taken into account as a matter of fact, while that Tribunal is, in that regard, obliged to follow the prevailing interpretation given to that domestic law by the courts or authorities of that Party, and those courts and those authorities are not, it may be added, bound by the meaning given to their domestic law by that Tribunal.”
71. The ECJ further finds that, as the jurisdiction of the CETA Tribunals and Appellate Tribunal is limited to the interpretation of CETA, no provision needs to be made for preliminary references to the ECJ.78 It also finds that it is consistent with EU law for CETA not to allow for any re-examination of awards by domestic courts (including the ECJ), and not to allow the investor to bring parallel or subsequent proceedings before such courts.79
72. The ECJ subsequently analyses other objections against CETA, discarding all of them, but those parts of the Opinion are less relevant to the issues that arise in the present proceedings.
73. The ECT is analogous to CETA in the following respects.
74. First, the ECT is, like CETA, an agreement between the EU and one or more third countries. The ECT and CETA are both mixed agreements, with the EU and the Member States as contracting parties. There is a difference as well here, because CETA is an essentially bilateral
78 Id. ¶ 134. ↩
79 Id. ¶ 135. ↩
[Page 33]
agreement, “between Canada, of the one part, and the European Union and its Member States, of the other part”.80 I have been conscious of that in drawing conclusions from the ECJ's Opinion. I do not accept the EC's reading of Opinion 1/17 as being confined to an agreement such as the one with Canada, which does not have intra-EU application.81 What the ECJ did was to distinguish between Achmea, which “concerned (...) an agreement between Member States” and “an agreement between the EU and a non-Member State”.82 The ECT is in the latter category.
75. Second, the ECT defines the applicable law in an analogous way. Article 26(6) ECT makes no reference whatsoever to the domestic laws of the Parties, nor to other treaties or agreements applicable between the Parties. Its formulation is more succinct than the CETA provisions, but the underlying principle is clearly the same: The jurisdiction of an ECT tribunal is limited to interpreting and applying the ECT. That follows from the phrase: “shall decide the issues in dispute in accordance with this Treaty and applicable rules and principles of international law”. The “issues in dispute” are, according to Article 26(1), investment disputes “concerning an alleged breach of an obligation (...) under Part III” of the ECT. EU law says nothing on the interpretation of the obligations under Part III of the ECT. The only rules and principles of international law “applicable” to determining a dispute under the ECT are general principles and customary rules of international law, such as the customary international law rules on treaty interpretation. EU law also cannot have been intended to form part of the “applicable rules and principles of international law”, because it is not binding on the non-EU Contracting Parties to the ECT.
80 Council Decision (EU) 2017/37 of 28 October 2016 (“Exhibit 21”), supra note 33; see also supra ¶ 23. ↩
81 See EC Brief, p. 16. ↩
82 Opinion 1/17 re CETA EU:C:2019:341 ¶ 127 (“Bungenberg Exhibit 60”). ↩
[Page 34]
76. These analogies are crucial for the present proceedings. It is in my opinion most likely that, if the ECJ were ever asked whether the ECT provisions on investment arbitration are compatible with EU law, its answer would be affirmative because of those analogies. I explore this in the following subsection.
77. The ECJ judgment in Achmea establishes that arbitration provisions in BITs between Member States “such as” Article 8 of the Netherlands-Slovakia BIT are contrary to EU law. The judgment applies erga omnes, and courts and tribunals in the EU Member States need to give effect to this finding, in any cases involving such BITs. However, the ECJ clearly distinguished these kinds of BITs, which do not have the EU itself as a contracting party, from agreements which do. The clear and express distinction which the ECJ made in Achmea is confirmed by Opinion 1/17, in which it accepted the lawfulness of the CETA dispute settlement system. CETA is a mixed agreement, like the ECT.
78. To this date, the ECJ has not been asked to rule on the validity of the act by which the EU concluded the ECT. Acts of the EU institutions can be annulled or declared invalid only by the EU General Court and by the ECJ.83 Annulment actions need to be brought within a two-month period after the adoption of the pertinent act or law,84 and that period has obviously expired in the case of the Council Decision concluding the ECT.85 However, any court or tribunal of a
83 The General Court (the “GC”) has jurisdiction over certain types of direct actions for annulment; its judgments can be appealed to the ECJ. Together the GC and the ECJ form the Court of Justice of the European Union (“CJEU”) (although there is variable use of this terminology: the abbreviation CJEU is often used for the ECJ). ↩
84 See Art. 263 TFEU (“Exhibit 6”). ↩
85 See Council and Commission Decision on the conclusion, by the European Communities, of the Energy Charter Treaty and the Energy Charter Protocol on energy efficiency and related environmental aspects EU [1998] OJ L 69, at 1 (“Exhibit 22”). ↩
[Page 35]
Member State may at any time ask the ECJ whether a particular act is invalid, and the ECJ could therefore still be requested to rule on the validity of the ECT under the EU Treaties. But it is vital to add that in the Foto-Frost judgment the ECJ established that it has sole authority to declare an EU act invalid; Member State courts cannot make such a declaration.86 There is therefore a presumption of validity afforded to acts of the EU in any proceedings before courts in the EU Member States, and it is only once the ECJ has declared the relevant EU act invalid (in this case the act through which the EU Council concluded the ECT) that Member State courts may proceed on the basis that there is an incompatibility between the agreement in issue (here the ECT) and EU law. It is in my opinion most remarkable that, in the present proceedings, this Court in a non-EU Member State is effectively asked to do what a court in a Member State could never do, on its own: To rule on the invalidity of the ECT under EU law.
79. The above means that there is no judicial authority whatsoever, in EU law, in support of the proposition that the ECT investment arbitration provisions are invalid in so far as intra-EU investments are concerned, or that such provisions are precluded.
80. How the ECJ would rule on the validity of the ECT's provisions on investment arbitration, in so far as they apply between Member States, is a matter of conjecture. In my opinion, the Court would be most unlikely to issue a ruling of invalidity, for the following reasons.
81. First, the ECT defines the applicable law in essentially the same way as does CETA. ECT investment disputes do not extend to any violations of EU law, because the ECT does not include agreements between the parties, or the parties' domestic law, in the applicable law.87 An
86 Case 314/85 Foto-Frost EU:C:1987:452 (“Exhibit 23”). ↩
87 I leave to one side the question whether, even if this finding were wrong, any use of “agreements between the parties” would need to be construed as the parties to the ECT rather than the parties to the dispute. In the former case, ↩
[Page 36]
ECT tribunal must therefore treat EU law in the same way as a CETA Tribunal: As a factual matter. This is confirmed by the reference, in Article 26(6) ECT, to “the issues in dispute”, which constitutes a cross-reference to Article 26(1), which is limited to “breach of an obligation under Part III”, i.e. the ECT provisions on Investment Promotion and Protection.88 As explained above, EU law is not relevant to the determination of whether there has been a violation of Part III of the ECT.
82. Tribunals have indeed looked at EU law in cases where their very jurisdiction was challenged.89 It cannot be the case, however, that a challenge to a tribunal's jurisdiction as being contrary to EU law is sufficient to conclude that any decisions by such a tribunal violate EU law, because they interpret or apply it. That would mean that, in any investment cases involving the EU or any of its Member States, it would be sufficient for these parties to challenge the tribunal's jurisdiction as being contrary to EU law for such a violation becoming the necessary outcome. Such jurisdictional circularity or tautology can never be a good-faith interpretation of the relevant international law provisions.
83. Moreover, and in line with the analysis above, there is in my opinion no inherent or necessary incompatibility between the ECT provisions, on the one hand, and EU internal market and EU energy law, on the other. Both sets of provisions aim at free and open markets, and at undistorted competition. Professor Bungenberg is unable to point to any actual conflicts.
the EU Treaties would of course not constitute applicable law either, as they do not bind the ECT Contracting Parties that are not EU Member States.
88 See the reasoning of the tribunal in Vattenfall AB v. Federal Republic of Germany, ICSID Case No. ARB/12/12, Decision on the Achmea Issue (Aug. 31, 2018) (“Bungenberg Exhibit 66”) ¶ 116. ↩
89 See the Awards in issue in this case. ↩
[Page 37]
84. Nor is there an inherent or necessary incompatibility between the remedies available under the ECT and those which EU law offers. As mentioned above, EU investors may claim damages for a Member State's sufficiently serious breach of EU law, including the TFEU provisions on the right of establishment or the free movement of capital. An ECT Tribunal may of course also award compensation, as did the Tribunals in this case.
85. There is also no incompatibility between the ECT and EU State aid law, contrary to what the EC claims.90 The thesis that an international arbitral award that compensates an investor for not having received fair and equitable treatment may constitute a State aid has been firmly rejected by the EU General Court, in a judgment which articulates the requirements which need to be satisfied in order for an arbitral award to constitute State aid.91 Specifically, the General Court confirmed an older ECJ judgment, Asteris and Others. In that judgment the ECJ stated that “State aid, that is to say measures of the public authorities favouring certain undertakings or certain products, is fundamentally different in its legal nature from damages which the competent national authorities may be ordered to pay to individuals in compensation for the damage they have caused to those individuals”.92 The General Court built on this finding, by establishing that “compensation for damage suffered cannot be regarded as aid unless it has the effect of compensating for the withdrawal of unlawful or incompatible aid”. There is no suggestion in this case that Italy withdrew unlawful or incompatible aid.
86. Nor is it easy to see in what way the non-applicability of the ECT in an intra-EU context would make sense from the perspective of avoiding any conflicts between the ECT's
90 See EC Brief, p. 17. EU State aid law regulates subsidies by Member States which distort competition. ↩
91 Case T-624/15 European Food and Others v Commission EU:T:2019:423, ¶ 104-106 (“Exhibit 24”). ↩
92 Joined Cases 106 to 120/87 Asteris v Greece EU:C:1988:457 ¶ 23 (“Exhibit 25”). ↩
[Page 38]
investment protection provisions and EU law. The ECT would continue to bind the EU and its Member States in relation to claims by investors from non-EU Contracting Parties. Those claims could be identical to any intra-EU claims. The nationality of an investor plays no role here. Moreover, an investor from an EU Member State, such as a company which has invested in another Member State, may well be owned or controlled by a national from a non-EU ECT Contracting Party. To illustrate this further, if the Claimants in this case were owned or controlled by such a non-EU national from an ECT Contracting Party, any claims by the latter investor would presumably not come within the intra-EU non-application – even if, in substance, the claim would be identical to the one which led to the present Awards.
87. Even if there were a conflict between the EU Treaties and the ECT (which, as I have explained above, there is not), that would not affect the validity of an EU Member State's offer to arbitrate under Article 26(3) of the ECT. First, as I have explained, the Achmea ruling cannot change the fact that the ECT is still in force.
88. Second, EU law cannot affect the application of the offer to arbitrate extended by the Contracting Parties to the ECT in Article 26(3) of that treaty. EU law is simply not applicable to that offer. As explained above, the applicable law provision in the ECT (Article 26(6)) does not address EU law. Even if it did, however, it is relevant only to the determination of the “issues in dispute” which, as I have also noted, constitutes a cross-reference to Article 26(1), and, as such, is limited to “breach of an obligation under Part III”, i.e. the ECT provision on Investment Promotion
[Page 39]
and Protection. Article 26(3) is in Part V of the ECT, on the Settlement of Disputes between an Investors and a Contracting Party. Article 26(6) therefore has no relevance to it.93
89. Third, in the event of any conflict between the ECT and EU law (deriving from the EU Treaties), the ECT would prevail.
90. Customary international law rules govern the resolution of conflicts between treaties. These are enshrined in Article 30 of the VCLT. The general rule is: (i) that the provisions of the later treaty prevail (the lex posterior principle),94 unless (ii) the treaties themselves indicate which one should take precedence.95 In the case of the ECT and EU Treaties it is difficult to see how this rule could favour the EU Treaties. The ECT contains an express conflict rule in Article 16(2), which states that where two or more Contracting Parties have entered into another treaty, whose terms “concern the subject matter of Part III or V of this Treaty”, the terms of Part III and Part V prevail, provided they are “more favourable to the Investor or Investment.” As explained above, I disagree that the EU Treaties (or EU law) conflict with Part III or Part V of the ECT (i.e. that there is subject matter coincidence). If Professor Bungenberg were correct as to his interpretation of the Achmea judgment, and there was such a conflict, EU law is clearly not more favourable to the investor, as it removes the opportunity for intra-EU arbitration.
91. Professor Bungenberg and the EC claim that EU law has primacy over the provisions of the ECT, including Article 26. There is, however, no authority for that proposition.
93 In the absence of any relevant choice of law clause, any questions regarding the interpretation of Article 26 would be considered applying international law principles of treaty interpretation. ↩
94 VCLT, Arts. 30(3), 30(4)(a) (“Exhibit 10”). ↩
95 Id., Article 30(2). ↩
[Page 40]
92. First, the ECJ has never used the term “primacy” for any such conflicts. As described above, the ECJ's jurisdiction is limited to matters of EU law. All parties in these proceedings agree that the EU Treaties form part of international law, even if according to the ECJ they set up an autonomous legal order. The ECJ is the ultimate interpreter of those Treaties, and needs to ensure that they are properly enforced. But the ECT does not give the ECJ jurisdiction to rule on any conflicts between its provisions and the EU Treaties, under international law, as opposed to EU law. The ECJ is, for the purpose of interpreting and applying the ECT, simply not an international court. It has the function of a supreme domestic court, no more. It is, for purposes of the ECT, the highest court of one of the Contracting Parties, not the court tasked with enforcing the treaty between the Contracting Parties.
93. The ECJ has never, to the best of my knowledge, made any statements to the opposite effect – in relation to the ECT, or to any other agreement to which the EU is a party, or to any agreements between EU Member States. It has in fact confirmed the limits on its own jurisdiction, in the seminal Kadi case, on the relationship between the UN Charter and EU law. That case (which, as I mentioned above, I was involved in) concerned a UN Security Council Resolution, imposing sanctions on Mr Kadi post 9/11, for purposes of counterterrorism. The ECJ established that Mr Kadi's fundamental rights under EU law had been violated by the EU's attempt to implement the sanctions. Crucially, however, it stated that “any judgment given by the Community judicature deciding that a Community measure intended to give effect to such a [UN] resolution is contrary to a higher rule of law in the Community legal order would not entail any challenge to the primacy of that resolution in international law”.96
96 Kadi ¶ 288 (“Bungenberg Exhibit 52”). ↩
[Page 41]
94. Instead, the ECJ characterizes any conflicts between a Member State’s international commitments and EU law as a breach of EU law, thereby imposing on the relevant Member State an obligation to remove that breach,97 for example by renegotiating the agreement or denouncing it.
95. By contrast, the “principle of the primacy of EU law” is concerned with the relationship between EU law and the domestic laws of the Member States, as indeed Professor Bungenberg recognizes.98 But contrary to what he and the EC suggest, the primacy of EU law does not extend to any conflicts between the EU’s internal rules and international laws or obligations.99 Neither the Treaty articles to which they refer, nor the case law they cite confirm this proposition. Indeed, those authorities fully recognize that the EU is bound by international law. Article 216(2) TFEU provides that international agreements are binding on the EU institutions and on the Member States. Article 218(11) recognizes that, in case of a conflict between an envisaged international agreement and the EU Treaties, either the agreement must not be concluded, or the Treaties must be amended. That provision would be unnecessary if EU law had primacy over an incompatible agreement. Article 351 TFEU recognizes that Member State agreements predating their accession take priority over incompatible EU law.100 The Germany v Council judgment established an incompatibility between a WTO protocol on trade in bananas and EU law; but the protocol was renegotiated in recognition of the fact that the incompatibility had to
98 Expert Declaration of Marc Bungenberg ¶ 35. ↩
99 Id., ¶ 36; EC Brief, pp. 20-21. ↩
100 It may be noted that the EU Treaties do not contain any provisions regulating the relationship between Member State agreements postdating their accession. That however does not mean that the EU Treaties have primacy over such agreements. ↩
[Page 42]
be removed, and that EU law could not simply prevail on the international plane. And as just analysed, the Kadi judgment also contradicts Professor Bungenberg’s and the EC’s statement.101
96. The EC refers to one sentence in the International Law Commission’s report on Fragmentation, which states that: “[t]he EC Treaty takes absolute precedence over agreements that Member States have concluded between each other.”102 Read in context, however, this sentence addresses treaties between EU Member States only, and the effect of Article 351 TFEU.103 It does not refer to a general rule of primacy of the EU treaties over any other treaties to which EU Member States are parties. The ECT is, of course, an agreement concluded between EU Member States and other States, and the EU.
97. For the reasons explained above, I do not consider that the Achmea Judgment has any bearing on intra-EU arbitration under the ECT. As such, it has no effect on the validity of the arbitration agreement concluded between Masdar and Spain, and therefore on the enforcement of the Award which is the subject of these proceedings.
98. In light of this finding, I cannot see on what basis international comity would favor a decision by this Court not to enforce the Awards. The EC claims that “intra-EU arbitration is fundamentally incompatible with EU law – a conclusion that clearly follows from the Court of Justice’s decision in Achmea”. It adds that “[n]evertheless, controversy remains, within academic
102 See EC Brief, p. 21, referring to Report of the Study Group of the International Law Commission, Fragmentation of International Law: Difficulties Arising from the Diversification and Expansion of International Law, UN Doc. No. A/CN.4/L.682 (2006) ¶ 283 (“Exhibit 26”). ↩
[Page 43]
circles as well as in arbitral practice”.104 In my opinion the reverse is the case. International arbitral practice does not accept that the Achmea judgment affects the validity of intra-EU arbitration under the ECT. The various intra-EU awards constitute international judicial authority. If there is controversy, it is fueled by the EC, supported by a majority of Member States, in whose interest it is to limit the scope for ECT arbitration.
99. Professor Bungenberg’s claim that the enforcement of the Awards would offend against the protection of public policy in Article V(2)(b) of the New York Convention is puzzling, to say the least. Article V(2)(b) refers to the public policy of the country in which enforcement of the award is sought – in this case the United States. However, Professor Bungenberg’s analysis does not refer to US public policy; it focuses on the EU constitutional principles of autonomy and primacy.105
100. In so far as this claim is targeted at the incompatibility with EU law which the ECJ articulated in Achmea, it cannot succeed for the same reasons as identified and explained above. Achmea is not concerned with the ECT, which is an EU agreement which does not include EU law as the law to be applied by arbitration tribunals. No EU court has found any incompatibilities between intra-EU arbitration under the ECT and EU law. The ECT is a valid agreement which binds the EU and the EU Member States. And even if there were any incompatibility, EU law does not have primacy over the ECT. Nor could one see in what way EU law could affect the interpretation of the New York Convention, to which the EU is not even a party.
105 Expert Declaration of Marc Bungenberg, ¶ 74. ↩
[Page 44]
101. It is true that the ECJ has, as Professor Bungenberg points out, expressed general concerns in Opinion 1/17 on the extent to which international investment arbitration may interfere with the EU’s right to regulate, and to determine the level of protection of its public policies.106 It is also true that the EU Council has mandated the EU Commission to participate in negotiations on the ECT’s modernization, including with a view to clarifying in what way the ECT safeguards that right to regulate. However, Professor Bungenberg is unable to show in what way the current ECT, and any arbitration under this treaty, offend against the EU’s right to regulate. Nor does he show in what way the Awards in issue in this case offend against that right. Moreover, the Awards concern purely Italian acts, and it is not clear at all whether the ECJ’s statements in Opinion 1/17 extend to the EU Member States’ right to regulate.
102. I declare under penalty of perjury of the laws of the United States of America that the foregoing is true and correct.
Executed on February 5, 2020 in London, United Kingdom.
Signature
Piet Eeckhout
[Page 45]
| Exhibit | Description |
|---|---|
| 1 | CV of Professor Piet Eeckhout |
| 2 | Case C-621/18 Andy Wightman and Others v Secretary of State for Exiting the European Union EU:C:2018:999. |
| 3 | Communication from the Commission to the European Parliament and the Council: Protection of intra-EU investment (2018), https://ec.europa.eu/transparency/regdoc/rep/1/2018/EN/COM-2018-547-F1-EN-MAIN-PART-1.PDF |
| 4 | International Law Commission, Guide to Practice on Reservations to Treaties (2011). |
| 5 | Treaty on European Union (“TEU”). |
| 6 | Treaty on the Functioning of the European Union (“TFEU”). |
| 7 | Case C-344/04 The Queen ex parte IATA v Department for Transport EU:C:2006:10. |
| 8 | Case C-308/06 Intertanko EU:C:2008:312. |
| 9 | Case C-327/91 France v Commission EU:C:1994:305. |
| 10 | Vienna Convention on the Law of Treaties (“VCLT”). |
| 11 | Opinion 1/75 re Understanding on a Local Cost Standard EU:C:1975:145. |
| 12 | Case C-648/15 Austria v Germany EU:C:2017:664. |
| 13 | Opinion of Mischo AG in Case C-62/98 Commission v Portugal EU:C:1999:509. |
| 14 | Case C-66/13 Green Network SpA EU:C:2014:2399. |
| 15 | Opinion 2/15 re EU-Singapore FTA EU:C:2017:376. |
| 16 | M Cremona, “Disconnection Clauses in EU Law and Practice”, in C Hillion and P Koutrakos (eds), Mixed Agreements Revisited (Oxford and Portland, Oregon: Hart Publishing, 2010). |
| 17 | Comprehensive Economic and Trade Agreement (CETA). |
| 18 | Council Decision (EU) 2017/37 of 28 October 2016 on the signing on behalf of the European Union of the Comprehensive Economic and Trade Agreement (CETA) between Canada, of the one part, and the European Union and its Member States, of the other part (EU OJ 2017 L 11 at 1, italics added). |
| 19 | Ruling 1/78 [1978] ECR 2151. |
| 20 | Opinion Advocate General Wathelet in Case C-284/16 Slowakische Republik v Achmea EU:C:2017:699. |
[Page 46]
| Exhibit | Description |
|---|---|
| 21 | Council Decision (EU) 2017/37 of 28 October 2016. |
| 22 | Council and Commission Decision on the conclusion, by the European Communities, of the Energy Charter Treaty and the Energy Charter Protocol on energy efficiency and related environmental aspects EU [1998] OJ L 69. |
| 23 | Case 314/85 Foto-Frost EU:C:1987:452. |
| 24 | Case T-624/15 European Food and Others v Commission U:T:2019:423. |
| 25 | Joined Cases 106 to 120/87 Asteris v Greece EU:C:1988:457. |
| 26 | Report of the Study Group of the International Law Commission, Fragmentation of International Law: Difficulties Arising from the Diversification and Expansion of International Law, UN Doc. No. A/CN.4/L.682 (2006). |