• 1. Introduction

  With the negotiations between the EU and Canada on the conclusion of the CETA, a hitherto little-noticed legal area moved to the forefront of the general public’s attention, namely, the law of bilateral investment treaties, the so-called BITs. In its Chapter 8, the CETA contains extensive provisions for the protection of Canadian investors’ investments in the EU and vice versa. It provides for both substantive investment protection standards and an investor-state arbitration mechanism in the form of an investment tribunal. This investment tribunal differs significantly from the arbitration mechanism of an investor-state dispute settlement (ISDS) system that has been used in international investment protection law up until now. It has a more judicial structure and also provides for a second instance.¹x See Opinion of 30 April 2019, Opinion 1/17 pursuant to Article 218(11) TFEU, CETA, ECLI:EU:C:2019:341 para. 108; Christian Riffel, ‘The CETA Opinion of the European Court of Justice and Its Implications – Not That Selfish After All’, Journal of International Economic Law, Vol. 22, Issue 3, 2019, p. 504. Since this investment court model was developed by the EU in response to the legitimacy crisis of investment protection law²x Susan D. Frank, ‘The Legitimacy Crisis in Investment Treaty Arbitration: Privatizing Public International Law Through Inconsistent Decisions’, Fordham Law Review, Vol. 73, Issue 4, 2005, pp. 1521-1625; Charles Brower & Stephan Schill, ‘Is Arbitration a Threat or a Boon to the Legitimacy of International Investment Law?’, Chicago Journal of International Law, Vol. 9, 2009, pp. 471-498. and represents the gold standard for the EU, serving also as template for other bilateral investment protection agreements of the EU (e.g. with Singapore or Vietnam), the question of its compatibility with EU (constitutional) law is of utmost importance. This is because investment protection was already highly controversial during the negotiations on the CETA, particularly with regard to ISDS. Investment protection mechanisms were met with various constitutional objections.³x For this discussion in Germany see Claus-Dieter Classen, ‘Der EuGH und die Schiedsgerichtsbarkeit in Investitionsschutzabkommen’, Europarecht, Vol. 47, Issue 6, 2012, pp. 611-627; Claus-Dieter Classen, ‘Die Unterwerfung Demokratischer Hoheitsgewalt Unter eine Schiedsgerichtsbarkeit’, Europäische Zeitschrift für Wirtschaftsrecht, Vol. 25, Issue 16, 2014, pp. 611-616; Claus-Dieter Classen, ‘Die Unterwerfung unter Völkerrechtliche (Schieds-)Gerichte: Kein Verfassungsverstoss!’, Zeitschrift für Europarechtliche Studien, Vol. 19, Issue 4, 2016, pp. 449-458; Steffen Hindelang, ‘Repellent Forces: The CJEU and Investor-State Dispute Settlement’, Archiv des Völkerrechts, Vol. 53, Issue 1, 2015, pp. 68-89; Christoph Ohler, ‘Die Vereinbarkeit von Investor-Staat-Schiedsverfahren mit Deutschem und Europäischem Verfassungsrecht’, Juristenzeitung, Vol. 70, Issue 7, 2015, pp. 337-346; Stephan Schill, ‘Investor-Staat Schiedsverfahren Sind Verfassungskonform’, Recht und Politik, Vol. 51, Issue 1, 2015, p. 11; Stephan Schill, ‘Investitionsschutz in EU Freihandelsabkommen: Erosion Gesetzgeberischer Gestaltungsmacht?’, Heidelberg Journal of International Law, Vol. 78, Issue 1, 2018, pp. 33-92; Christian Tietje, ‘Investitionsschutzgerichtsbarkeit in CETA und anderen Freihandelsabkommen der EU: Völkerrecht als Verfassungsverstoß?’, Zeitschrift für Europarechtliche Studien, Vol. 19, Issue 4, 2016, pp. 421-432. There were fears of an impairment of the legislator’s room for maneuver. Allegedly, the mere possibility that an investor may regard a state measure to be an impairment of his investment (and thus as an indirect expropriation or as unfair treatment in the sense of an investment protection agreement) and therefore, relying on this, could claim damages before an arbitral tribunal, restricts the legislator’s decision-making freedom. Hence, a situation could arise in which the state authorities refrain from taking e.g. environmental, consumer or climate protection measures that could impede economic activity. This fear is further substantiated by the fact that the interpretation of investment protection standards by arbitration tribunals is often perceived as quite arbitrary and extremely business-friendly. In legal literature, the resulting reluctance of the legislator is referred to as the so-called chilling effect.⁴x Or ‘regulatory chill’, Peter-Tobias Stoll et al., Investitionsschutz und Verfassung, Mohr Siebeck, Tübingen, 2017, p. 115. Although these negative effects has not been the subject of extensive empirical research,⁵x For nuanced view see Kyla Tienhaara, ‘Regulatory Chill and the Threat of Arbitration: A View From Political Science’, in Chester Brown & Kate Miles (eds.), Evolution in Investment Treaty Law and Arbitration, Cambridge University Press, Cambridge, 2011, p. 606. there are known examples, such as the events surrounding the Moorburg coal-fired power plant in Hamburg,⁶x Markus Krajewski, ‘Umweltschutz und Internationales Investitionsschutzrecht am Beispiel der Vattenfall-Klagen und des Transatlantischen Handels- und Investitionsabkommens (TTIP)’, Zeitschrift für Umweltrecht, Vol. 25, Issue 7-8, 2014, pp. 396-403. where authorities failed to carry out an environmental impact assessment in conformity with EU law,⁷x Judgment of 26 April 2017, Case C-142/16, Commission v. Germany, ECLI:EU:C:2017:301. probably also for fear of the negative consequences. The criticism of the negative consequences of possible compensation claims for the openness of the democratic process is combined with very fundamental constitutional objections to the investor-state dispute settlement itself. ISDS is regarded as inadmissible special justice for the benefit of multinational companies, in violation of equality before the law. Furthermore, there are complaints about rule of law deficits in the judicial process, such as judicial independence, transparency of the proceedings, and insufficient legitimation of the arbitrators.⁸x For the latter see e.g. Katrin Groh, ‘Endstation Karlsruhe! – Schiedsgerichtsbarkeit in Freihandelsverträgen’, Zeitschrift für Europarechtliche Studien, Vol. 19, Issue 4, 2016, p. 442. In Germany, the criticism has led to a still pending constitutional complaint before the Federal Constitutional Court, in which, inter alia, the indeterminacy of investment protection standards, namely of indirect expropriation and fair and equitable treatment, are criticized.⁹x For the argumentation in the pending constitutional complaint see at https://www.mehr-demokratie.de/fileadmin/pdf/2016-08-30_CETA-Klage.pdf. The German Federal Constitutional Court (BVerfG) did not comment on this in its decisions on interim relief (BVerfG, Beschl. v. 13.10. 2016 – 2 be 3.16 – BVerfGE 143, 65-101; BVerfG, Beschl. v. 07.10.2016 – 2 BvR 1444.16 – BVerfGE 144, 1-17).
  The decisions of the EU Council on the signing and provisional application of the CETA in October 2016 were highly controversial in Belgium too. Belgium has therefore linked its approval inter alia with an application to the CJEU for an opinion under Article 218(11) TFEU. The concerns raised related to the compatibility of the provisions on the Investment Court System (ICS) in the CETA (Chapter 8, Section F, Article 8.18 et seq.) with EU primary law, including fundamental rights. Specifically, compatibility with the autonomy of EU law and the related competence of the CJEU to interpret EU law in an authoritative, binding manner was questioned. It further impugned its compatibility with the principle of equality, the right of access to an independent court and the effet utile of EU law.¹⁰x See at https://diplomatie.belgium.be/de/newsroom/minister_reynders_reicht_antrag_auf_gutachten_zum_ceta_abkommen_ein. On 30 April 2019, the CJEU, sitting as full court, accepted the compatibility of the ICS with primary EU law.¹¹x Opinion of 30 April 2019, Opinion 1/17 pursuant to Article 218(11) TFEU, ECLI:EU:C:2019:341 - CETA. Since the CJEU’s Achmea decision at the latest, which rejected the ISDS system in intra-EU BITs,¹²x Judgment of 6 March 2018, Case C-284/16, Achmea, ECLI:EU:C:2018:158. Advocate General Wathelet, however, had no objections to the compatibility of intra-EU arbitration mechanisms with EU law, see Opinion of Advocate General Wathelet delivered on 19 September 2017, Case C-284/16, Achmea, ECLI:EU:C:2017:699. He even regarded such arbitral tribunals as having the right to refer a case to the CJEU under Article 267 TFEU. this decision had been eagerly awaited.¹³x See Panos Koutrakos, ‘More on Autonomy – Opinion 1/17 (CETA)’, European Law Review, Issue 3, 2019, pp. 293-294. (Koutrakos 2019a).
  The present contribution analyses the CETA Opinion of the CJEU, which, in comparison with earlier decisions, in a surprisingly uncritical, and, in terms of argumentation and result, quite surprising¹⁴x See Marc Bungenberg & Catharine Titi, ‘CETA Opinion – Setting Conditions for the Future of ISDS’, EJILTalk!, 5 June 2019, at www.ejiltalk.org/ceta-opinion-setting-conditions-for-the-future-of-isds/. consideration of conceivable conflicts between the competences of the CETA Tribunal on the one hand and those of the CJEU on the other hand, did not identify any objections against CETA ICS. This was notwithstanding the fact that the CJEU had been quite hostile to external jurisdiction over the past years.¹⁵x This applies both to the EEA Court in its initial form, to the Unified Patent Court and the ECtHR, see Opinion of 14 December 1991, Opinion 1/91 pursuant to Article 228(1) EC, ECLI:EU:C:1991:490; Opinion of 8 March 2011, Opinion 1/09 pursuant to Article 300(6) EC, ECLI:EU:C:2011:123; Opinion of 18 December 2014, Opinion 2/13 pursuant to Article 218(11) TFEU, ECLI:EU:C:2014:2454. Initial reactions have welcomed the CETA Opinion as an extension of the EU’s room for maneuver in the area of investment protection. Consequently, the CETA ICS – and thus also plans for a comparatively structured Multilateral Investment Court – are compatible with EU law, albeit with certain conditions which are not made explicit in the operative part of Opinion 1/17 but in its text instead. These conditions are likely to restrict the scope for EU external contractual relations to a considerable extent. These restrictions on the autonomy of the Union’s legal order, as established by the CJEU, are the central focus of the present contribution.
  The CJEU examined the complaints against the ICS in three steps: (i) First, it addressed the question of compatibility with the autonomy of the EU legal order (paras. 106-161), (ii) then the alleged violation of the principle of equal treatment (paras. 162-188), and (iii) finally, the compatibility of the ICS with the right of access to an independent court (paras. 189-244). Apart from the explicit use of constitutional terminology in the area of trade law, there are two remarkable points in this opinion. (i) On the one hand, the CETA Opinion extends the concept of the autonomy of EU law to include the protection of certain democratic decision-making processes against interference by international jurisdiction. More concretely, the CJEU has protected not so much the processes themselves, but rather the political results of the process, as far as the determination of the level of protection of public interests is concerned. This raises the question what this restriction of the EU’s capacity to enter into international agreements and be bound be international law, denoted a fundamental constitutional statement by the President of the CJEU, Koen Lenaerts¹⁶x Koen Lenaerts, Modernising Trade Whilst Safeguarding the EU Constitutional Framework: An Insight Into the Balanced Approach of Gutachten 1/17, 2019, p. 16, at https://diplomatie.belgium.be/sites/default/files/downloads/presentation_lenaerts_Gutachten_1_17.pdf. may mean for other international treaties, in particular when it comes to EU obligations under WTO law.¹⁷x According to Riffel 2019, p. 519, as a consequence of the CETA Opinion, a necessity test in the narrower sense under WTO law (as required by Article XX GATT, Article XIV GATS) must now be regarded as in violation of EU law. Similarly, the criticism of Patricia Sarah Stöbener de Mora & Stephan Wernicke, ‘Riskante Vorgaben für Investitionsschutz und Freihandel – Das CETA-Gutachten des EuGH’, Europäische Zeitschrift für Wirtschaftsrecht, Vol. 30, 2019, p. 973, which criticize a contradiction between the leeway parliaments have in a democracy with regard to their legislation, on the one hand, and their capacity to enter into international legal obligations, on the other hand. Such statements neglect the differences between WTO dispute settlement and CETA ICS. More on this below. (ii) On the other hand, it follows from the CJEU’s assessment of fundamental rights that those included in the EU Charter on Fundamental Rights (FRC) also clearly extend to external action.¹⁸x On this externalization see Heiko Sauer, ‘Europarechtliche Schranken Internationaler Gerichte’, Juristenzeitung, Vol. 74, Issue 19, 2019, p. 928 et seq. The CJEU takes two concrete requirements from Article 47 FRC. On the one hand, a kind of legal aid for small and medium-sized enterprises is to be established until CETA enters into force, and on the other hand, no retroactive interpretative decisions are to be issued to safeguard the independence of the CETA Tribunal.¹⁹x Opinion 1/17, paras. 221 and 237.
  Owing to their fundamental importance for the international treaty-making capacity of the EU, this article focuses on the autonomy line of arguments of the CJEU in the CETA Opinion, while the fundamental rights issues just mentioned are reserved for a different analysis. In what follows, the new accentuation of the external autonomy of EU law in the CETA Opinion is analyzed (Section 2). Subsequently, the considerations of the CJEU on the delimitation of its competences from those of the CETA Tribunal are critically examined. The rather superficial analysis carried out by the CJEU observed in this context stands in stark contrast with the approach of the CJEU in earlier decisions. Indeed, it misjudges problems and therefore only seemingly gives a clear delimitation of competences by limiting the CETA Tribunal to the interpretation and application of the CETA only (Section 3). This section is followed by a consideration of the last part of the CJEU’s autonomy assessment, in which it turns to the criticism of regulatory chill (Section 4). Here, by referring to the unimpeded functioning of EU institutions in accordance with the EU constitutional framework, the CJEU identifies the new barrier for investment protection mechanisms just mentioned, taking back the earlier affirmation of the comprehensive competences of the CETA Tribunal for the CETA in one point. This raises many questions about the concrete significance, consequence, and scope of application of this barrier, which considerably limits the extension of the EU’s competence granted by the CJEU to act under international law.

• 2. The Autonomy of EU Law as a Yardstick: The CJEU on New Paths

  A key question for the compatibility of the ICS with EU primary law is the impact of the jurisdiction of an international court on the EU legal order. In this respect it is an issue going to the heart of the autonomy of EU law, i.e. the fundamental autonomy and independence of EU law in its development and unfolding of law. The CJEU has always claimed this autonomy both internally with respect to the national constitutional law of the Member States and externally with respect to international law.²⁰x See Judgment of 5 February 1963, Case C-26/62, Van Gend en Loos, ECLI:EU:C:1963:1, para. 12; Judgment of 15 July 1964, Case C-6/64, Costa v. E.N.E.L., ECLI:EU:C:1964:66. For more on the external autonomy see Loic Azoulai, ‘Structural Principles in EU Law: Internal and External’, in Marise Cremona (ed.), Structural Principles in EU External Relations Law, Hart Publishing, Oxford, 2018, p. 31. The CJEU has developed the autonomy of EU law in its decisions on international courts into a core yardstick for assessing their jurisdiction’s compatibility with the legal order of the EU, most clearly in Opinion 2/13 on the Draft Accession Agreement to the ECHR.²¹x Opinion 2/13, para. 158. The fact that autonomy was particularly important there and that the CJEU was expansive in its conceptualization had its primary legal basis in Protocol No. 8 on accession to the ECHR, but the decision went far beyond that. The autonomy of EU law had already been used previously as a yardstick for the assessment of the jurisdiction of international courts, starting with the EEA Court, although there the CJEU had essentially limited its deliberations on autonomy to its relevance for the constitutional order of the EC.²²x Opinion 1/91, para. 35. The CJEU was primarily concerned with preserving its own ultimate responsibility for the interpretation and application of EU law.²³x Id. para. 46. In the CETA Opinion, the autonomy of EU law was also put into context with international law. It refers to the “essential characteristics of the European Union and its law”, which the CJEU identifies as being an independent source of law (namely the Treaties), characterized

  “by its primacy over the laws of the Member States, and by direct effect of a whole series of provisions that are applicable to their nationals and to the Member States themselves. Those characteristics have given rise to a structured network of principles, rules and mutually interdependent legal relations binding the European Union and its Member States.”

  The CJEU refers to this as “a constitutional framework” in its own right encompassing

  “founding values set out in Article 2 […], the general principles of EU law, the provisions of the Charter, and the provisions of the EU and FEU Treaties, which include, inter alia, rules on the conferral and division of powers, rules governing how the EU institutions and its judicial system are to operate, and fundamental rules in specific areas, structured in such a way as to contribute to the implementation of the process of integration described in the second paragraph of Article 1 TEU.”²⁴x For the quotes see Opinion 1/17, paras. 109-110.

  Thus, autonomy of EU law now, going far beyond safeguarding the jurisdiction of the CJEU, includes the central elements of the legal effects of EU law in and between the Member States, which are in turn judged by the CJEU.²⁵x The CJEU emphasizes its particular importance for safeguarding these special characteristics, Opinion 1/17, para. 111. In addition, it also includes fundamental procedural and substantive rules of EU primary law of constitutional dignity, such as the protection of fundamental rights, the institutional order and the allocation of competences between the institutions.²⁶x Also the function of the national courts as Union courts is one of these essential elements of the constitutional order of the EU, see Opinion 1/09, paras. 68 et seq.; Opinion 2/13, para. 175. This is the corollary of the CJEU’s competence to secure its exclusive jurisdiction for preliminary rulings under Article 267 TFEU and thus again serves the CJEU. The role of the national courts was wrongly ignored in the CETA Opinion; see below. In the CETA Opinion this definition of the EU’s autonomy is expanded: the CJEU not only examines whether the jurisdiction of the CETA Tribunal interferes with the jurisdiction of the CJEU to decide on the validity and interpretation of EU law, but also discusses whether the jurisdiction of the CETA Tribunal affects the competence and functioning of the democratic legislator of the Union as governed by the EU constitutional framework (paras. 137-160). Thus, a violation of EU autonomy also occurs if the CETA Tribunal influences the EU legislator’s exercise of competence. The unimpaired functioning of other EU institutions thus becomes a further yardstick for reviewing the legality of international courts’ jurisdiction. In the CETA Opinion, the CJEU for the very first time declares the protection of EU political processes (although, as will be shown, limited to the definition of the level of protection) from the effects of international court rulings to be an element of the autonomy of EU law. Thus, the CJEU now goes beyond the self-referential standards of past decisions in determining what the autonomy of the EU legal order actually implies.²⁷x See Koutrakos 2019a. For criticism of the self-referential attitude of the CJEU in previous decisions on international courts, which in effect denies legal dialogue, see Bruno de Witte, ‘A Selfish Court? The Court of Justice and the Design of International Dispute Settlement Beyond the EU’, in Marise Cremona & Anne Thies (eds.), The European Court of Justice and External Relations Law, Hart Publishing, Oxford, 2014, p. 46; Bruno de Witte, ‘The Relative Autonomy of the European Union’s Fundamental Rights Regime’, Nordic Journal of International Law, Vol. 88, Issue 1, 2019, p. 71; Jed Odermatt, ‘The Principle of Autonomy: An Adolescent Disease of EU External Relations Law?’, in Marise Cremona (ed.), Structural Principles in EU External Relations Law, Hart Publishing, Oxford, 2018, p. 316; Steve Peers, ‘The EU’s Accession to the ECHR: The Dream Becomes a Nightmare’, German Law Journal, Vol. 16, Issue 1, 2015, pp. 213-222; Riffel 2019, p. 506; Eleanor Spaventa, ‘A Very Fearful Court?’, Maastricht Journal of European and Comparative Law, Vol. 22, Issue 1, 2015, p. 47. The CETA Opinion, however, establishes new barriers, within the concept of autonomy, to the participation of the EU in international law as far as treaties establishing international courts are concerned. What exactly the protection of democratic decision-making actually means, however, remains quite unclear, even on closer inspection of the CJEU’s opinion. Before going into this in more detail under Section 4, the first, more traditional part of the autonomy analysis related to the CJEU’s jurisdictional competences must be analyzed.

• 3. Preserving the CJEU’s Jurisdiction to Interpret EU Law: The Court’s Cursory Examination

  3.1. Starting Point: Interpreting CETA Is for the CETA Tribunal

  It is settled case-law that the EU’s participation in international treaties establishing their own international courts is as such in conformity with the EU Treaties.²⁸x Opinion 1/91, para. 40. The CJEU’s definitive jurisdiction over EU law is preserved if an international court has jurisdiction only to interpret the specific agreement, especially as this is done according to interpretive rules of international law, and not of EU law, and if this Court’s activities do not affect the interpretation of EU law.²⁹x Id. paras. 49 et seq.; Opinion 1/09, paras. 73 et seq. The CJEU emphasizes that the exclusive jurisdiction of the CJEU and of national courts to interpret EU law, as laid down in Article 19 TEU, does not conflict with this. Although the agreements concluded by the EU form part of EU law³⁰x See Judgment of 30 April 1974, Case C-181/73, Haegeman, ECLI:EU:C:1974:41, paras. 5 et seq. and are therefore also subject to Article 19 TEU, the CJEU considers that its jurisdiction insofar does not take precedence over the jurisdiction of the international courts established by international treaties.³¹x Opinion 1/17, para. 116. The CJEU does not give any reasons for this statement. This statement is probably due to the distinction between jurisdiction over EU, internal or international law: the CJEU cannot claim a monopoly on the interpretation of international agreements of the EU as such.³²x Riffel 2019, p. 513. It has the task of interpreting the EU’s international agreements internally, and only with internal effect, and of assessing their effects within the EU. This, however, leaves open the question as to what significance the interpretation of a treaty by an international court will have for the application of this treaty in the EU. The CJEU also stresses that the EU must be capable of concluding agreements which establish an autonomous court independent from national courts in order to maintain its competence to enter into international relations.³³x CJEU, Opinion 1/17, para. 117. The CJEU, however, for the first time restrains the EU’s capacity by requiring that these courts must not be empowered to issue judgments that prevent EU institutions from functioning in accordance with the constitutional framework of EU law.³⁴x Id. paras. 118 et seq.

  3.2. The CJEU’s Cursory Analysis of the CETA Provisions on Jurisdiction

  When examining whether the CETA provisions ensure that the CETA Tribunal only has jurisdiction to interpret the CETA, the CJEU takes a purely formal, textual, and rather uncritical approach. Indeed, the CJEU rightly emphasizes the provisions of Articles 8.18 and 8.31(1) and (2) CETA, which attribute the interpretation of the CETA to the CETA ICS and explicitly exclude the application of the national law of a contracting party from its jurisdiction. In this respect, the CJEU also rightly highlights the factual differences between assessing the jurisdiction under the CETA and the jurisdictions the CJEU had to assess in its decisions on the Unified Patent Court and on the Intra-EU BITs in Achmea.³⁵x Id. paras. 120-126. However, Article 8.31(2) CETA does allow an interpretation of national law by the CETA ICS when it states that it is not binding on national courts. The CJEU does not consider this unclear (‘fuzzy’)³⁶x Giulia Claudia Leonelli, ‘CETA and the External Autonomy of the EU Legal Order: Risk Regulation as a Test’, Legal Issues of Economic Integration, Vol. 47, Issue 1, 2020, p. 50. provision to be relevant because it allegedly only deals with the use of national law as fact.³⁷x Opinion 1/17, para. 131. This distinction between the application of national law as fact and its interpretation as law is not further discussed by the CJEU as if it was easy to apply and generally accepted, which it is not.³⁸x See Hindelang 2015, pp. 76 et seq.; Leonelli 2020, pp. 55 et seq. The assessment required by Article 8.31(2) CETA as to whether the CETA Tribunal follows the prevailing interpretation of national law, is less a question of fact than of interpretation. Of course, it is the task of an investment tribunal to examine national measures for compliance with the protection standards for foreign investments enshrined in the international treaty; whether these measures are lawful under national law is largely irrelevant in the assessment before the tribunal. However, the exact determination of the national measure and the assessment whether it is in conformity with the protection standards may require a more detailed examination of the substance and effect of national law and may thus lead to its interpretation by the tribunal. The Advocate General in his opinion on the CETA recognized this, but did not assess it as a problem because of the lack of binding force.³⁹x Opinion of Advocate General Bot delivered on 29 January 2019, Opinion 1/17, ECLI:EU:C:2019:72, paras. 136 and 142 (AG Bot’s Opinion). The problem that the CETA Tribunal then determines the substance of the EU act in a decisive way, since this determination is binding and unalterable in the investment protection trial, is also ignored by Advocate General Bot, see para. 143. Both the CJEU and the AG fail to see that the interpretation provided by the CETA Tribunal is the basis for its judgment on damages, which is absolutely and irrevocably binding on the national authorities and courts in the context of enforcement under Article 54 of the ICSID Convention.
  The CJEU also regards the provision in Article 8.21 CETA on the designation of the correct defendant as unproblematic. For the CJEU it merely states the competence of the EU to designate whether a claim should be brought against the EU or a Member State.⁴⁰x Opinion 1/17, para. 132. Indeed, the EU is given the opportunity under Article 8.21(3) CETA to determine the correct defendant, so that, were this possibility to be used, the CJEU’s jurisdiction over the allocation of competences between the EU and the Member States is preserved. However, Article 8.21(4) CETA determines what happens if the EU does not make such a designation within 50 days: in that case, the determination of the defendant will depend on whether the measure is imputed to a Member State or the EU, the determination of which is up to the plaintiff company. This determination cannot be questioned neither by the CETA ICS nor by the EU or the Member States.⁴¹x Cf. Article 8.21(5-7) CETA. Thus, the determination of the correct defendant does not fall within the jurisdiction of the CETA Tribunal, but in the absence of a designation by the EU, the choice of the plaintiff is binding and is then taken as a basis by the CETA Tribunal, without the CJEU being able to rectify it.

  3.3. Constellations Not Considered by the CJEU

  In the absence of pertinent complaints by Belgium, the CJEU did not address further problem constellations, although a comprehensive legal analysis is certainly the task of the court. These problem constellations include the following:
  (i) Interpretation of reservations. CETA contains definitions and reservations that refer to EU secondary legislation.⁴²x See e.g. Article 1 of the Protocol on Mutual Recognition of the Results of Conformity Assessment regarding the notion of in-house body, or Annex I on Reservations for Existing Measures in which the EU in several instances refers to specific EU legislation. The same applies to Annex II regarding reservations for future measures. Assessing the compatibility of EU or Member State measures falling within the scope of the reservations with CETA investment protection standards will require the CETA Tribunal to determine the scope of such EU and national measures. Insofar as the scope of the EU’s reservations in Annexes I and II to the CETA is determined by reference to existing EU secondary law, it is inevitable that, in order to determine the substance of a measure when reviewing it under CETA standards, the provisions and scope of the relevant secondary law will have to be considered, which may lead to its interpretation by the CETA Tribunal. This differs from the situation in which EU law, as a contested measure, is examined for its compatibility with the CETA investment protection standards. While in the latter case EU law is the object of the examination, in the former case EU law itself is the yardstick for the CETA Tribunals examination as the EU reservations in CETA determine the scope of the exception. The treatment of EU law by the CETA Tribunal as a mere fact hardly seems possible in the latter case. Since the CETA Tribunal also has no right to make a reference for a preliminary ruling to the CJEU,⁴³x So explicitly Opinion 1/17, para. 134. the question remains open as to how the exact scope of EU reservations can be determined by the CETA Tribunal, if it has no power of interpretation. To answer whether an investor was treated in a manner incompatible with CETA standards, it is important to clarify whether the EU is entitled to an exception in this respect by virtue of the reservation. For, if so, the obligations triggering liability according to Article 8.18 CETA in connection with Section C and D of Chapter 8 CETA are not applicable due to Article 8.15 CETA.
  (ii) Function of Article 267 TFEU and the role of national courts. As noted above, in earlier decisions, the safeguarding of the function of the preliminary ruling procedure under Article 267 TFEU and of the role of national courts has been an important focus of the examination of the jurisdiction of international courts in light of the autonomy of EU law. In the CETA Opinion, the CJEU does not address this issue, although there was reason to do so. Also with regard to the CETA Tribunal, the repercussions of the ICS on the ability of national courts to submit references for a preliminary ruling to the CJEU must be critically examined, as the CJEU had done with regard to the Unified Patent Court⁴⁴x Opinion 1/09, para. 80. Admittedly, the jurisdiction of the Patent Court to interpret relevant Union law was intended to be exclusive (see Riffel 2019, p. 512), which is not the case with the CETA Tribunal. or with regard to accession to the ECHR.⁴⁵x See Opinion 2/13, paras. 236 et seq. The CJEU raised concerns even though the Draft Accession Agreement to the ECHR provided for a preliminary referral mechanism to the CJEU, which is completely absent in CETA. For, the initiation of proceedings before the CETA Tribunal presupposes that the plaintiff terminates or waives national proceedings,⁴⁶x See Article 8.22(1)(f) and (g) CETA. either of which deprives national courts of the possibility to clarify questions of EU law by means of a referral; termination of national proceedings leads to a termination of an already pending referral before the CJEU.⁴⁷x See para. 24 of the CJEU’s Recommendations to National Courts and Tribunals in Relation to the Initiation of Preliminary Ruling Proceedings. By raising a claim before the CETA Tribunal, the plaintiff is thus in a position to prevent a decision by the CJEU or of a national court which could eliminate or confine the disputed measure, or which could define the facts to which the CETA Tribunal is bound (it must follow the interpretation of the Union act given to it by the EU courts).⁴⁸x See Article 8.31(2) CETA. Here, the CJEU would certainly have had reason to derive requirements from the restriction on jurisdiction enshrined in Article 8.31 CETA as to how the interaction with national courts is to be handled by the plaintiff. The CJEU could have shown limits to Article 8.22 CETA that gives the plaintiff considerable leeway with regard to the relationship with national courts. But the CJEU failed to do so in the CETA Opinion. Instead, the opinion may even aggravate the threat to the uniform interpretation and application of EU law. This is rooted in the fact that the CETA Tribunal itself, in the absence of an opinion of the CJEU, interprets the EU legal situation in a certain way and, on this basis, identifies unfair treatment or expropriation of a Canadian investor in the EU, and awards compensation, even though the CJEU may later be given the opportunity in different proceedings to adopt a different interpretation of Union law, which could remove the factual basis for the finding of unfair treatment or expropriation by the CETA Tribunal.⁴⁹x For a case scenario insofar see Leonelli 2020, pp. 59-61. In its opinion on the Patent Court, the CJEU had also foreseen that the jurisdiction of international courts must not

  “affect the powers of the courts and tribunals of Member States in relation to the interpretation and application of European Union law, nor the power, or indeed the obligation, of those courts and tribunals to request a preliminary ruling from the Court of Justice and the power of the Court to reply.”⁵⁰x Opinion 1/09, para. 77; see also AG Bot’s Opinion, para. 69.

  Nevertheless, the Advocate General does not identify any problems here either; instead he assumes that the fact that the CETA Tribunal is outside national jurisdiction perfectly corresponds to the lack of direct effect of the CETA.⁵¹x Id. paras. 94, 167 et seq. This approach fails to understand that demanding respect for the jurisdiction of national courts, as is submitted here, has nothing to do with them being able to interpret the CETA, but with their (and the CJEU’s) jurisdiction to interpret EU measures that are relevant in the dispute before the CETA ICS.

  3.4. Conclusion

  With its rather uncritical and incomplete consideration of CETA provisions on the jurisdiction of the CETA Tribunal, the CJEU allows the EU to continue to participate in investment protection mechanisms establishing investment courts having exclusive definitive jurisdiction,⁵²x See in this respect Opinion 1/17, para. 135. and to advance the project of a multilateral investment court, provided that the jurisdiction of the investment courts is limited to the interpretation and application of the respective treaty, and not EU law. The latter requirement must always be explicitly agreed upon by the EU and its treaty partners because of Article 42(2) ICSID (according to which the national law of the Contracting States is the applicable law for investment arbitration⁵³x In general international law, this is quite different, see PCIJ, Reports 1926, A, p. 19.).⁵⁴x Opinion 1/17, para. 117; see also AG Bot’s Opinion, paras. 77 et seq. The CJEU was obviously keen to keep the EU’s external policy leeway open with regard to establishing international courts, even though the EU only enjoys a shared competence with regard to investment courts.⁵⁵x Opinion of 16 May 2017, Opinion 2/15 pursuant to 218(11) TFEU, ECLI:EU:C:2017:376, paras. 292 et seq.; Judgment of 5 December 2017, Case C-600/14, Germany v. Council, ECLI:EU:C:2017:935, paras. 66 et seq., where the CJEU caused some uncertainty as to the nature of EU external competences (exclusive or still shared?) in the case of shared internal competences. This attitude of the CJEU is in clear contrast to the extremely critical, self-referential examination of the possible effects of arbitration mechanisms under investment protection law on EU law as delivered in Achmea,⁵⁶x Case C-284/16, Achmea, paras. 50 et seq. and to the fundamental rights jurisprudence of the ECtHR on the protection of fundamental rights in the EU following the accession of the EU to the ECHR in the Accession to the ECHR Opinion.⁵⁷x Opinion 2/13, paras. 187 et seq. For criticism of the different approaches see also Panos Koutrakos, ‘The Anatomy of Autonomy: Themes and Perspectives on an Elusive Principle’, in ECB (ed.), Building Bridges: Central Banking Law in an Interconnected World, 2019, pp. 96 and 98. (Koutrakos 2019b). While the CJEU was rather too critical at that time, now, in the CETA Opinion, it is too uncritical, confining itself to a purely formal, text-based analysis without examining the effects of the CETA ICS in greater depth.⁵⁸x See also Leonelli 2020, pp. 44, 48 et seq.

• 4. Safeguarding Democratic Decision-Making Processes, or Political Determinations of Levels of Protection?

  4.1. A New Autonomy Postulate

  For the CJEU, respecting the autonomy of the EU includes, for the first time, the stipulation that the jurisdiction of international courts must not prevent EU institutions from functioning in accordance with the EU constitutional framework.⁵⁹x Opinion 1/17, para. 119. Basically, one can agree with this postulate, since the protection of one’s own constitutional foundations may also be maintained when international agreements are concluded;⁶⁰x See Article 46 of the Vienna Convention on the Law of Treaties, which permits the invocation of a manifest violation of national law of fundamental importance. this must be granted to the EU just as much as to a nation state.⁶¹x Sauer 2019, p. 932.
  However, in the CETA Opinion, the CJEU goes far beyond safeguarding constitutional rules or even the substance of the constitution. The CJEU in this respect took issue with the competence of the CETA Tribunal to (also) review EU measures of general application. This is because it could lead to a situation in which the Tribunal makes the level of protection of a public interest as defined by the EU legislator the subject of review, instead of limiting itself to assessing whether the treatment of an investor meets the CETA investment protection standards.⁶²x Opinion 1/17, paras. 143 and 148. Although the CJEU had clarified before that the CETA Tribunal may neither annul an EU act nor demand its conformity with the investment protection standards, but is limited to a mere award of compensation for damages,⁶³x Id. para. 144. the CJEU recognizes a problem in that the EU would be forced to stop seeking this level of protection by repeated⁶⁴x This wording has led to considerations that a one-time award of damages could not yet be a problem. What speaks against such an understanding of the CJEU, however, is the fact that in the following paragraphs of the opinion the frequency of awards does not play a role in the CJEU’s argument. The CJEU formulates its conclusions on the lack of jurisdiction of the CETA Tribunal to assess levels of protection in absolute terms. awards.⁶⁵x Opinion 1/17, para. 149. The CJEU holds that the Union’s autonomous functioning within its constitutional framework would be impaired if examinations by international courts of a level of protection set by the EU were to lead to the EU having to amend or repeal a rule.⁶⁶x Id. para. 150. The justification given for this refers to the fact that it is only the CJEU (and no other court) that is competent to review the level of protection against the EU’s primary law requirements.⁶⁷x Id. para. 151. With this argumentation, the CJEU does not leave it at the purely formal consideration, according to which the CETA Tribunal may not judge on the level of protection or generally on the legality of an EU measure. Instead, the CJEU includes the effects of judgments on claims for compensation on the EU legislator into its consideration, in the sense of a ‘regulatory chill’. The CJEU assesses the de facto pressure for change on the legislator as a threat to the functioning of the EU constitutional order.
  One may wonder why the CETA Tribunal should have reason to deal with the level of protection defined by an EU act, as its competence is limited to the examination of the investment protection standards enshrined in CETA Chapter 8, Sections C and D. Its task is (only) to examine whether the contested measure constitutes unfair treatment or unlawful expropriation or discrimination. However, CETA contains some provisions which uphold the national right to regulate: the exception to the prohibition of discrimination according to Article 28.3(2) CETA, the affirmation of national regulatory sovereignty in Article 8.9 CETA, point 1 and 2 of the General Interpretative Instrument, and No. 3 of Annex 8-A on the definition of the notion of indirect expropriation. These provisions stipulate – generally speaking – that it does not constitute a violation of investment protection standards if a Contracting Party takes measures which pursue legitimate objectives; in some cases, the additional requirement of the necessity or of the non-discriminatory character of the measure must be respected. These rules serve as a justification for otherwise incompatible measures or as a guideline for the interpretation of investment protection standards. They must therefore be observed and taken into account by the CETA Tribunal when applying and interpreting the investment protection standards. From these rules, the CJEU, “reading those provisions together”,⁶⁸x Id. paras. 152, 154-155, 157. draws the conclusion that

  “the discretionary powers of the CETA Tribunal and Appellate Tribunal do not extend to permitting them to call into question the level of protection of public interest determined by the Union following a democratic process.”⁶⁹x Id. para. 156.

  The CJEU also held that “the CETA Tribunal has no jurisdiction to declare incompatible with the CETA the level of protection of a public interest established […] and, on that basis, to order the Union to pay damages.”⁷⁰x Id. para. 153. In conclusion, the CJEU states that the contracting parties have limited the scope of the investment protection standards, taking care to exclude the power of the CETA ICS to challenge democratic decisions on levels of protection.⁷¹x Id. para. 160: “Those tribunals have no jurisdiction to call into question the choices democratically made within a Party relating to, inter alia, the level of protection of public order or public safety, the protection of public morals, the protection of health and life of humans and animals, the preservation of food safety, protection of plants and the environment, welfare at work, product safety, consumer protection or, equally, fundamental rights.”

  4.2. Critical Analysis of the CJEU’s Argumentation

  It is important to present this sequence of the CJEU’s reasoning in such detail because the statements raise a number of questions and also invite misunderstandings, such as the misunderstanding that the CJEU has rejected any external control over EU acts. First of all, it should be noted that the CJEU determines the scope of jurisdiction of the CETA Tribunal on the basis of the relevant provisions in the CETA. Thus, it mainly interprets an international treaty and restricts the jurisdiction of the CETA Tribunal to examine contested EU measures in one respect, namely with regard to the EU’s determination of levels of protection. However, the CJEU dresses this in a language of protection of constitutional institutions. If the CJEU, in accordance with the heading to paras. 137 et seq. of the CETA Opinion, had been concerned with the operation of the constitutional organs of the EU in accordance with the EU constitutional framework, it would have had to identify threats to their functioning. However, the fact that an international court examines a national measure cannot, in itself, immediately and without further ado be perceived as a threat to the constitutional functioning of the EU legislator. To identify such a danger requires further and deeper argumentation and analyses, which the CJEU does not provide.
  The legislator is not automatically endangered in its functioning if one of its decisions is reviewed by a court (even an international one) established with its consent. The restriction of national leeway through international legal ties is not a new process, nor does it meet with democratic objections a priori. Entering into binding treaty obligations is an expression of democratic self-determination. The fact that international obligations are more difficult to reverse than those of a national law, does not put this into question.⁷²x For further details see Stoll et al. 2017, pp. 118 et seq. Therefore, any influence on national (or EU) laws that results from an international agreement that has been approved by the competent parliament can, in principle, be considered legitimate. The same applies to an external judicial control of legislation, which the legislator has agreed to. Binding the own sovereignty (and the autonomy of Union law is a variety of EU ‘sovereignty’⁷³x See Christina Eckes, ‘The Autonomy of the EU Legal Order’, Europe and the World: A Law Review, Vol. 4, Issue 1, 2020, pp. 8 et seq.) – including that of the legislator – is a necessary, almost phenotypical consequence of entering into international treaties. The ability and willingness to do so is a prerequisite for the conclusion of international treaties and for the participation in contemporary international law-making, which is characterized by a proliferation of international courts, not least in an effort to promote the rule of law in international relations. The democratically legitimate legislator may certainly confine the autonomy of its legal system, as long as there are no constitutional obstacles to this. The functioning of the legislator is not affected by this, in contrast to what the CJEU insinuates in paras 148 et seq. of the CETA Opinion – although the CJEU shares the starting point that the legislator is capable of harmonizing rules by entering into a treaty.⁷⁴x Cf. Opinion 1/17, para. 148, “without prejudice to a situation where the Parties have agreed, within the framework of the CETA, to harmonise their legislation.”
  WTO law, for example, contains numerous disciplines that limit the national design of foreign trade law and which were subject to an (until recently) effective WTO jurisprudence. Should this now be inadmissible? If the CJEU’s statement on safeguarding legislative protection levels against international courts’ jurisdiction (or: against restrictions under international law generally), derived from the protection of institutional democratic processes, were to be understood in such a general way, it would likely undermine any competence of the EU to enter into international legal obligations, and would also have to negate the leeway of the legislator to make the decision to restrict its autonomy. Certainly, the legislator’s capacity for self-restraint through entering into international treaties is subject to constitutional limits. However, the CJEU remains silent on these in its opinion. According to the interpretation of the CJEU, the legislature has in any case not subjected itself to external control as far as the levels of protection of public interests are concerned.
  With regard to the CETA, the CJEU is concerned with the protection of certain legislative decisions. It is not about the protection against threats to the constitutional institutions and their decision-making processes, but about safeguarding decisions taken regarding the level of protection. The two must be separated. Why and to what extent the CJEU wants to protect secondary legislation on levels of protection as an expression of the EU’s autonomy against international obligations is not really clear. For justifying safeguarding the level of protection adopted by the EU legislator against international obligations, the CJEU merely refers to the functioning of the EU institutions according to the EU constitutional framework.
  One could therefore consider taking as a justification the protection of democracy in the EU, which assigns essential decisions solely to the legislator;⁷⁵x Cf. Article 290(2) sentence 2 TFEU. however, this is not the approach of the CJEU. For the CJEU, the protection of e.g. occupational welfare, food safety, environmental protection or plant protection, is just as important as that of fundamental rights;⁷⁶x Opinion 1/17, para. 160. for a constitutional argumentation, this is a startling, even treacherous equalization. It is reasonable to assume that the CJEU was actually concerned with safeguarding its own ultimate responsibility for monitoring EU legislation with regard to the definition of protection levels.⁷⁷x On EU autonomy as a rhetorical shield for the protection of the competences of the CJEU see Bruno de Witte, ‘European Union Law: How Autonomous Is Its Legal Order?’, Zeitschrift für Öffentliches Recht, Vol. 65, Issue 1, 2010, p. 150. Namely, the CJEU assigns to itself the power to examine whether the level of protection of public interests ‘inter alia’ complies with EU primary law.⁷⁸x Opinion 1/17, para. 151. The wording ‘inter alia’ leaves room for the possibility that international law, insofar as it forms part of the acquis communautaire, is also one of the yardsticks of review to be examined exclusively by the CJEU. Interestingly, the CJEU again bases this on Article 19 TFEU,⁷⁹x Id. although it had previously stated that this provision does not take precedence over the jurisdiction of international courts (see above).
  Article 216(2) TFEU is also not a suitable justification, to the contrary. As agreements concluded by the EU take precedence over secondary law, according to the case-law of the CJEU,⁸⁰x See also Article 216(2) TFEU. the protection of secondary law against obligations resulting from international agreements is in need of justification. The yardstick for controlling the legality of international agreements of the EU is primary law, not secondary law, nor the level of protection laid down therein. Secondary law cannot therefore be used as a yardstick for assessing the legality of international treaties (to be) entered into by the EU.⁸¹x See also Article 218(11) TFEU.
  Even though the reasoning of the CJEU is not convincing, it is nevertheless correct to confine the scope of the jurisdiction of the CETA Tribunal because of the protection of the right to regulate in the CETA. Those who allow the control of legislative decisions by international courts in principle and with good reason (for what is the role of international courts other than to control and contain national measures?) will not immediately affirm that every international court would have this jurisdiction at all, or in full. The CJEU in its CETA Opinion derived from an interpretation of CETA provisions that the CETA Tribunal does not have this competence only insofar as the determination of the level of protection is concerned.⁸²x Lenaerts 2019, “emphasise[d] that what the Court is protecting here is not EU measures of general application as such. Nothing in Chapter 8 of the CETA suggests that measures of that kind are ‘immune’ from review before the CETA tribunals. On the contrary, as the Court expressly confirms, such measures may give rise to the award of monetary compensation under the ISDS mechanism when, for example, they amount to discrimination or arbitrary treatment affecting a Canadian investor. That is not incompatible with EU primary law […] What the Court is protecting instead is the essence of the democratic process leading to the adoption of EU norms protecting public interests, a process which forms part of the EU constitutional framework.” Indeed, investment protection standards do not directly deal with disciplining national regulation. Indirect effects nevertheless arise. Investment protection provisions providing for an obligation to pay compensation in the case of indirect expropriation or unfair treatment of foreign investors, lead to claims for compensation before an arbitral tribunal, because the arbitral tribunal may, in individual cases, consider a particular national provision to amount to an impermissible expropriation. A national decision laying down a specific reconciliation or balancing of interests (e.g. between economic interests and environmental protection) may be deemed as giving rise to compensation under the investment protection standards. Therefore, this national act in the end is at least partially inadmissible in view of the investment treaty obligations, if measured against the international legal standards in the individual case. However, the provisions in CETA which uphold the right to regulate, aim to limit or completely exclude precisely such findings of the CETA ICS. The emphasis on the right to regulate in the CETA provisions mentioned above demonstrates that the investment protection standards and their enforcement by the CETA ICS should not restrict national regulation in this respect. Therefore, it is consistent that the CJEU undertakes this protection of national regulations, to which the parties to the agreement attached such great importance, enforcing its observance by limiting the scope of jurisdiction of the CETA Tribunal.
  The CJEU’s reasoning, however, is not very consistent.⁸³x See also the criticism by Koutrakos 2019b, p. 100. The CJEU denies the CETA Tribunal the power to examine the compatibility of the level of protection set by the legislator with the CETA obligations. In this respect, it apparently restricts the scope of the Tribunal’s jurisdiction. However, the wording of the above-mentioned CETA provisions on the protection of the right to regulate much rather represents a justification provision or an interpretation guideline limiting the Tribunal’s control power. These provisions are not formulated as carve-outs from the Tribunal’s jurisdiction; instead, they represent justifications or govern the standard of review to be applied by the CETA Tribunal.⁸⁴x AG Bot’s Opinion, paras. 133 et seq., only considers them duties to be taken into account. Furthermore, it is not entirely clear whether the CJEU always excludes the jurisdiction to award compensation for damages when the level of protection is at stake. The exclusion of compensation awards is mentioned or at least alluded to in paras. 149, 153 and 159, but not in paras. 156 and 160, where the CJEU does not infer a prohibition against compensation awards. Does this mean that the CETA Tribunal may nevertheless award damages if and because it accepts the level of protection set by the legislator, recognizing it as a fact, i.e. does not call it into question, but nevertheless finds in its assessment that it (and not other provisions of an EU legal act⁸⁵x In this respect, the competence of the CETA Tribunal to award damages should be upheld, according to Lenaerts 2019.) amounts to an indirect expropriation in the individual case before the Tribunal? In any case, para. 159 of the CETA Opinion excludes the jurisdiction of the CETA Tribunal to award compensation for unfair treatment. Or does the exclusion of compensation awards on the basis of para. 152 (to which para. 153 refers) only apply in the context of invoking Article 28.3(2) CETA, but not in the case of the other above-mentioned CETA provisions which provide justification for the EU to exercise its right to regulate, or which prescribe a specific interpretation of the investment protection standards? Is the CETA Tribunal also barred from examining the restrictive conditions of these CETA provisions (no arbitrary or unjustifiable discrimination, no disguised restriction to trade), insofar as this would entail assessing the level of protection of the EU? Is there not a certain contradiction between paras. 153 and 156, as the former excludes the power of holding the level of protection incompatible with the CETA, whereas the latter prohibits already calling the level of protection into question, i.e. prohibiting any examination at all? Actually, the CETA Tribunal is only competent to examine a level of protection with a view to its compatibility with CETA provisions, i.e. to examine whether this EU determination of protection level means or in effect leads to discrimination, unfair treatment or unlawful indirect expropriation (which the CJEU prohibits under paras. 153 and 159). In any case, an investment tribunal is never competent to call into question a national level of protection of public interests, so that the CJEU’s prohibition in para. 156 actually goes without saying (except in the context of examining the necessity of a measure within the meaning of Article 28.3(2) CETA and No. 3 of Annex 8-A). The difference between paras. 153 and 156 with regard to the award of compensations could also be resolved if one granted the power to award compensations to the CETA Tribunal at the level of international law, but would disregard such a decision in the EU internally because of the priority of EU law. However, the CJEU has not in the least hinted at such a disconnection between the binding effect (only) under international law of a judgment of the CETA Tribunal and its lack of internal enforceability.⁸⁶x Such differentiation is also not obsolete due to the lack of direct effect of CETA. If the CJEU were to imply such a disconnection, it would have disregarded or even denied the largely unconditional enforceability of arbitral awards under the CETA pursuant to Article 54 ICSID Convention. There is no trace of this in the opinion. Therefore, it appears quite obvious that the CJEU, by excluding compensation awards, not only rejects their internal effect, but a priori denies the CETA Tribunal jurisdiction in this respect already at the level of international law. Furthermore, this exclusion of jurisdiction applies to all exceptions or affirmations of the right to regulate in CETA.⁸⁷x See also Catharine Titi, Opinion 1/17 and the Future of Investment Dispute Settlement: Implications for the Design of a Multilateral Investment Court, 2020, p. 17, at https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3530875.
  In sum, the CJEU in its CETA Opinion postulates that no compensation may be awarded if this would undermine the determination of a level of protection by the EU legislator. These findings of the CJEU are valid for the EU, but the CJEU missed the issue of how these findings can become binding on the level of international law. The CJEU does not address the issue of how the limitation to the jurisdiction of the CETA ICS that the Court has drawn in its opinion can become binding under international law.⁸⁸x See also Sauer 2019, p. 928. Making the findings of the CJEU effective at the level of international law requires a corresponding declaration of interpretation by the CETA parties or a reservation at the time of ratification by the EU. Furthermore, uncertainties remain, such as the question why the CETA Tribunal should not be able to examine the restrictive conditions in the exception clauses of the CETA (necessity, non-discrimination), since the parties to the CETA have qualified and hence limited the protection of the right to regulate in this respect. The conceptualizations of the CJEU seem to classify these provisions as self-judging, i.e. the competence to assess their requirements appears to lie solely with the contracting parties.⁸⁹x See Titi 2020, p. 18. There is, however, no trace in the CETA of a comprehensive self-judging nature of provisions on the right to regulate. Nor is there any hint in the CETA at why the respect for the legislator’s decision-making leeway should only apply to the level of protection and not to other aspects of the relevant legislation as well. The CETA provisions mentioned above protect the freedom of the legislator not solely in respect of determining levels of protection.

  4.3. Drawbacks on Other Courts?

  The above analysis of the opinion demonstrates that the emphasis on the protection of national regulatory sovereignty and the results of political processes with regard to the determination of the levels of protection, which the CJEU emphasizes in Opinion 1/17, apply to the investment protection mechanism in the CETA and the pertinent provisions in CETA. They did play a pivotal role in the CJEU’s reasoning. Hence, the ruling must not be regarded as a general guideline for, and limitation to all types of obligations of the EU under international law or for any existing international court. The latter question was actually not at stake. Therefore, one may not infer from broadly drafted statements of the CJEU’s CETA Opinion, such as in para. 150, that the CJEU rejects a control of legislative determinations of levels of protection from the outset and for all international courts. The CJEU simply did not comment on this.⁹⁰x Opinion 1/17, para. 119. The operative part of the CJEU’s opinion expressly applies to the CETA. It can only be applied to comparable constellations. The statements of the CJEU must be understood against this background: they were made in the field of international investment protection law and arbitral jurisdiction in this context. They refer to a court established by an international treaty that particularly emphasizes national regulatory sovereignty. The jurisdiction of the CETA Tribunal should find its limit in the regulatory sovereignty of the Contracting States with regard to the determination of protection levels. Respect for the right to regulate can indeed be seen to be quite deeply enshrined in the CETA regulations.⁹¹x On the protection of the ‘Right to Regulate’ in CETA, see Catharine Titi, ‘Right to Regulate’, in Makane Moïse Mbengue & Stefanie Schacherer (eds.), Foreign Investment under the Comprehensive Economic and Trade Agreement (CETA), Springer International Publishing, Heidelberg, 2019, pp. 159 et seq.; Schill 2018, pp. 70 et seq. A comparison to WTO law may be useful here. WTO law – upholding, similarly to the CETA, the freedom of regulation in various provisions⁹²x Wolfgang Weiss, WTO Law and Domestic Regulation, Beck Hart Nomos, Munich 2020, p. 8 et seq. – contains specific obligations which impose limits on the WTO parties in their foreign trade and internal economic regulation. It thereby restricts domestic legislative leeway in this respect, also with regard to the determination of protection levels (suffice it to recall the requirement for scientific assessments insofar in the SPS Agreement). However, this is not the case with investment protection law. The latter does not directly affect national regulatory leeway. Investment protection courts can only award compensation, but they cannot issue statements which oblige states to alter or amend their domestic regulations. The difference between the CETA Tribunal and the WTO jurisdiction is also emphasized by the CJEU in the CETA Opinion. Yet in this respect, the Court only emphasizes the greater flexibility in the implementation of WTO dispute settlement decisions⁹³x Opinion 1/17, para. 146. and therefore remains limited in its awareness of the differences between the CETA and WTO law. This is because the Court does not take into account the clearly different regulatory structure, objectives and subject matter of WTO law compared to investment protection law and the differences in the judicial system (protection of individual investors by awarding compensations in investment law versus state action under WTO law leading to obligations to amend domestic law). Although the parallelism between the developments of WTO law and investment protection law is now often emphasized, this does not allow for drawing premature conclusions. With the conclusion of the CETA investment protection law, the EU legislator in its treaty-making capacity does not make use of its constitutional competence to restrain the autonomy of the EU legal order by setting certain limits (this may be different in other chapters of CETA, where the parties might have agreed on harmonization of their legislation, which the CJEU explicitly refers to⁹⁴x Opinion 1/17, para. 148: “However, without prejudice to a situation where the Parties have agreed, within the framework of the CETA, to harmonise their legislation.”), whereas it did so by become a founding party of the WTO. This is a pivotal difference to WTO law.
  However, the attempt made here to confine the scope of the statements made by the CJEU in the CETA Opinion by emphasizing their context and analyzing the argumentation is subject to considerable doubts. Koen Lenaerts has described the constitutional protection of the level of protection established by the EU legislator as “a major contribution to what I often describe as the EU’s functional constitution, that is to say a Union founded upon democracy, justice and rights.”⁹⁵x Lenaerts 2019, italicization in the original. It can thus be expected that, at least in the view of the President of the Court, the CJEU’s commitment to safeguarding legislature’s prerogative to set the protection level as a barrier to the jurisdiction of international courts or even beyond, to the external treaty making capacity of the EU in general, is meant as a fundamental, universally valid imperative of EU constitutional law. For such a sweeping conclusion, however, this limitation is too poorly justified in the CETA Opinion. The reasoning and argumentation of the CJEU does not explain why and to what extent the determination on the level of protection is supposed to be the “essence of the democratic process”⁹⁶x Id. which is to be shielded from obligations under international law. But further questions also remain open: what does safeguarding the level of protection mean in concrete terms? Which provisions of a legislative act, within which a level of protection is established, are covered by constitutional protection?
  If safeguarding the level of protection were to be a constitutional obligation universally applicable to all sorts of external treaties, even if only vis-à-vis international courts, this could have considerable consequences for the EU’s capacity to submit to international dispute settlement, e.g. within the WTO. The WTO dispute settlement bodies would consequently be barred from taking decisions on general obligations under WTO law which imply an assessment of the EU’s level of protection. The only exception from this prohibition would apply to those WTO rules in which WTO members had more or less explicitly agreed to harmonize their legislation, as the CJEU has alluded to.⁹⁷x Opinion 1/17, para. 148. It is not clear what this would mean in practice for each individual provision in WTO law.
  Would the exception of harmonization already apply to general limitations of domestic legislative action, as is actually consistently the case with WTO disciplines?⁹⁸x See also Article XVI:4 of the WTO Agreement, according to which each member undertakes to bring its laws into conformity with its obligations under WTO law. Or would it only apply to explicit positive harmonization or standardization obligations in WTO law, which is only rarely found in WTO law? Such instances of explicit alignment would be the disciplines under Article VI:4 GATS, but also the requirements of scientific risk assessment, as required by Article 5 SPS Agreement for the determination of an appropriate level of sanitary or phytosanitary protection. WTO law is largely limited to negative integration, which restrains the domestic legislator’s room for maneuver by virtue of rather general obligations such as non-discrimination, or the obligation to reason domestic legislation against stipulations of necessity or scientific assessments, without requiring any harmonization of domestic regulations. Nevertheless, when applied by the WTO dispute settlement bodies, WTO law definitively has an impact on domestic determinations of levels of protection of policy interests. This can be clearly observed when analyzing the practice regarding the necessity test in the exception provisions laid down in Article XX GATT, Article XIV GATS, or in Article 2.2 TBT Agreement and Article 2.2 SPS Agreement.⁹⁹x For more detail on their effects on domestic regulatory autonomy see Wolfgang Weiss, WTO Law and Domestic Regulation, pp. 246 et seq. and 289 et seq. If one sticks to Koen Lenaerts’ generalizing view, which does not appear to be supported by the wording of the CJEU’s CETA Opinion, the legality of the jurisdictional competence of the WTO dispute settlement system under EU constitutional law would be subject to considerable doubt. In the CETA Opinion the CJEU makes an obiter dictum with regard to the WTO dispute resolution’s flexibility¹⁰⁰x Opinion 1/17, para. 146. whose relevance for determining the scope or applicability of the constitutional limitation is not clear. Maybe the CJEU wanted to stress the differences to indicate that its reasoning does not apply to the WTO jurisdiction. In light of Koen Lenaerts’ statement, a general carve-out for the WTO dispute settlement might be at least partially invalid. Although the implementation of WTO panel/Appellate Body reports always allows for a considerable degree of flexibility, this flexibility cannot be maintained in the long term against the will of a complainant who was victorious against the EU. Furthermore, if the CJEU’s statements applied to a WTO report against the EU in which the panel finds a level of protection under EU law to be incompatible with WTO obligation, such a report would overstep the jurisdiction of the international body. This is because the WTO panel would already have been incompetent to settle such dispute. Furthermore, a closer look at the factual effects of the WTO panel reports shows that they are implemented in the vast majority of cases, including the EU (with the exception of the hormone-treated beef conflict with the US, with which an alternative solution has been agreed upon, at least for the time being) so that they have an established impact on the EU legislator’s choices.
  The far-reaching consequences for the EU’s capacity to conclude treaties and its participation in the WTO call for reconsidering the CJEU’s commitment to safeguarding the level of protection defined by EU organs, if it is truly meant as a general statement. The conclusions the CJEU in the CETA Opinion derives from EU autonomy with regard to safeguarding levels of protection can only be welcomed insofar as it makes a clear case for appealing to the WTO judiciary to pay greater respect to a level of protection laid down by a democratic legislator and to limit its standard of review in this respect. But to postulate this as an EU limitation to establishing jurisdictional competences of international courts in general goes far beyond this.

  4.4. Conclusion

  Even though one might agree with the result of the CJEU’s assessment of conformity of the CETA ICS with EU primary law in respect of concerns raised regarding the domestic legislator’s leeway, there still remain the considerable deficiencies of the opinion’s reasoning. While the CJEU actually engaged in simple treaty interpretation of CETA provisions, it came up with an exaggerated dictum of applying constitutionally termed limitations to the EU’s international treaty making power by allegedly protecting the operation of the EU legislator in accordance with the EU constitutional framework. This resulted in an absolute protection of legislative determination of levels of protection, which was then transformed into a general constitutional limit through Koen Lenaerts’ subsequent comments. The CJEU’s argumentation is not convincing, even though one might excuse it by the necessities of the opinion procedure (which requires the CJEU to develop arguments based on primary law) as well as the intense political dispute over the CETA investment protection. The CJEU could have restrained itself to a mere statement that an impairment of the legislator’s leeway in public policies by the CETA investment protection is not to be feared, simply by referring to the confines to the adjudicative powers of the CETA Tribunal resulting from the guarantees of the right to regulate enshrined in the CETA. Furthermore, instead of decreeing an absolute limitation of the CETA ICS jurisdiction, it would have been sufficient to postulate a considerably reduced standard of review. The CJEU has made completely unnecessary statements here, failing to give sufficient reasons for them, and it did so without considering how the Court’s decree can be made binding under international law (its jurisdiction is binding only for the EU). In addition, further questions remain unanswered: Are the CETA provisions which protect the right to regulate a necessary condition for the compatibility of international jurisdiction with primary EU law,¹⁰¹x In this sense see Sauer 2019, p. 935. or do they represent mere sufficient conditions for finding their compatibility with EU primary law? What is the legal situation if in other EU BITs the provisions on the right to regulate are formulated differently? To what extent does the exclusion of compensations in the CETA, as inferred by the CJEU, apply to EU regulations that determine a level of protection, i.e. which provision of an EU legislative act would be covered by the exclusion of compensation?

• 5. Conclusion

  The CETA Opinion of the CJEU has instigated further development of EU law on external relations in constitutional parlance. On the one hand, the significance of fundamental rights was strengthened and the external autonomy of EU law has been expanded. On the other hand, the CJEU’s examination of the scope of the jurisdiction of the CETA Investment Tribunal remains incomplete and surficial. The Court’s obvious and, in the light of the values of Articles 3.5 and 21(2)(b) TEU, constitutionally mandated¹⁰²x AG Bot’s Opinion, paras. 176 and 178; de Witte 2014, p. 45. However, this does not imply a constitutional obligation for the EU to submit to any kind of international jurisdiction without limits. effort to expand the EU’s leeway to enter into international treaties establishing international courts, is undermined by an overly constitutional stipulation derived from EU external autonomy. The latter prescribes absolute protection for legislatively set protection levels for policy interests against international adjudication. This stipulation, which considerably confines the EU’s capacity to contribute to international courts is poorly reasoned and hardly justifiable, as far as it amounts to protection of specific political results and decisions of the legislator instead of the legislative decision-making process itself. A sound alternative would be to protect the legislature’s ability to shape the law, and to identify the constitutional limitations to the legislator’s capacity to bind itself.¹⁰³x Developing this further was beyond the scope of this paper. For more on this see Wolfgang Weiss, forthcoming in Europarecht, Vol. 55, Issue 6, 2020. The protection of the functioning of democratic processes is likely to remain a permanent postulate of EU autonomy by the CJEU against obligations flowing from international treaties. Nevertheless, it deserves a careful development of its concrete stipulations and a clear justification derived from the principle of democracy. Furthermore, making such limitations to international adjudication effective at the international level requires a corresponding declaration by the parties or a reservation at the time of ratification on the side of the EU.

Noten

• 1 See Opinion of 30 April 2019, Opinion 1/17 pursuant to Article 218(11) TFEU, CETA, ECLI:EU:C:2019:341 para. 108; Christian Riffel, ‘The CETA Opinion of the European Court of Justice and Its Implications – Not That Selfish After All’, Journal of International Economic Law, Vol. 22, Issue 3, 2019, p. 504.

• 2 Susan D. Frank, ‘The Legitimacy Crisis in Investment Treaty Arbitration: Privatizing Public International Law Through Inconsistent Decisions’, Fordham Law Review, Vol. 73, Issue 4, 2005, pp. 1521-1625; Charles Brower & Stephan Schill, ‘Is Arbitration a Threat or a Boon to the Legitimacy of International Investment Law?’, Chicago Journal of International Law, Vol. 9, 2009, pp. 471-498.

• 3 For this discussion in Germany see Claus-Dieter Classen, ‘Der EuGH und die Schiedsgerichtsbarkeit in Investitionsschutzabkommen’, Europarecht, Vol. 47, Issue 6, 2012, pp. 611-627; Claus-Dieter Classen, ‘Die Unterwerfung Demokratischer Hoheitsgewalt Unter eine Schiedsgerichtsbarkeit’, Europäische Zeitschrift für Wirtschaftsrecht, Vol. 25, Issue 16, 2014, pp. 611-616; Claus-Dieter Classen, ‘Die Unterwerfung unter Völkerrechtliche (Schieds-)Gerichte: Kein Verfassungsverstoss!’, Zeitschrift für Europarechtliche Studien, Vol. 19, Issue 4, 2016, pp. 449-458; Steffen Hindelang, ‘Repellent Forces: The CJEU and Investor-State Dispute Settlement’, Archiv des Völkerrechts, Vol. 53, Issue 1, 2015, pp. 68-89; Christoph Ohler, ‘Die Vereinbarkeit von Investor-Staat-Schiedsverfahren mit Deutschem und Europäischem Verfassungsrecht’, Juristenzeitung, Vol. 70, Issue 7, 2015, pp. 337-346; Stephan Schill, ‘Investor-Staat Schiedsverfahren Sind Verfassungskonform’, Recht und Politik, Vol. 51, Issue 1, 2015, p. 11; Stephan Schill, ‘Investitionsschutz in EU Freihandelsabkommen: Erosion Gesetzgeberischer Gestaltungsmacht?’, Heidelberg Journal of International Law, Vol. 78, Issue 1, 2018, pp. 33-92; Christian Tietje, ‘Investitionsschutzgerichtsbarkeit in CETA und anderen Freihandelsabkommen der EU: Völkerrecht als Verfassungsverstoß?’, Zeitschrift für Europarechtliche Studien, Vol. 19, Issue 4, 2016, pp. 421-432.

• 4 Or ‘regulatory chill’, Peter-Tobias Stoll et al., Investitionsschutz und Verfassung, Mohr Siebeck, Tübingen, 2017, p. 115.

• 5 For nuanced view see Kyla Tienhaara, ‘Regulatory Chill and the Threat of Arbitration: A View From Political Science’, in Chester Brown & Kate Miles (eds.), Evolution in Investment Treaty Law and Arbitration, Cambridge University Press, Cambridge, 2011, p. 606.

• 6 Markus Krajewski, ‘Umweltschutz und Internationales Investitionsschutzrecht am Beispiel der Vattenfall-Klagen und des Transatlantischen Handels- und Investitionsabkommens (TTIP)’, Zeitschrift für Umweltrecht, Vol. 25, Issue 7-8, 2014, pp. 396-403.

• 7 Judgment of 26 April 2017, Case C-142/16, Commission v. Germany, ECLI:EU:C:2017:301.

• 8 For the latter see e.g. Katrin Groh, ‘Endstation Karlsruhe! – Schiedsgerichtsbarkeit in Freihandelsverträgen’, Zeitschrift für Europarechtliche Studien, Vol. 19, Issue 4, 2016, p. 442.

• 9 For the argumentation in the pending constitutional complaint see at https://www.mehr-demokratie.de/fileadmin/pdf/2016-08-30_CETA-Klage.pdf. The German Federal Constitutional Court (BVerfG) did not comment on this in its decisions on interim relief (BVerfG, Beschl. v. 13.10. 2016 – 2 be 3.16 – BVerfGE 143, 65-101; BVerfG, Beschl. v. 07.10.2016 – 2 BvR 1444.16 – BVerfGE 144, 1-17).

• 10 See at https://diplomatie.belgium.be/de/newsroom/minister_reynders_reicht_antrag_auf_gutachten_zum_ceta_abkommen_ein.

• 11 Opinion of 30 April 2019, Opinion 1/17 pursuant to Article 218(11) TFEU, ECLI:EU:C:2019:341 - CETA.

• 12 Judgment of 6 March 2018, Case C-284/16, Achmea, ECLI:EU:C:2018:158. Advocate General Wathelet, however, had no objections to the compatibility of intra-EU arbitration mechanisms with EU law, see Opinion of Advocate General Wathelet delivered on 19 September 2017, Case C-284/16, Achmea, ECLI:EU:C:2017:699. He even regarded such arbitral tribunals as having the right to refer a case to the CJEU under Article 267 TFEU.

• 13 See Panos Koutrakos, ‘More on Autonomy – Opinion 1/17 (CETA)’, European Law Review, Issue 3, 2019, pp. 293-294. (Koutrakos 2019a).

• 14 See Marc Bungenberg & Catharine Titi, ‘CETA Opinion – Setting Conditions for the Future of ISDS’, EJILTalk!, 5 June 2019, at www.ejiltalk.org/ceta-opinion-setting-conditions-for-the-future-of-isds/.

• 15 This applies both to the EEA Court in its initial form, to the Unified Patent Court and the ECtHR, see Opinion of 14 December 1991, Opinion 1/91 pursuant to Article 228(1) EC, ECLI:EU:C:1991:490; Opinion of 8 March 2011, Opinion 1/09 pursuant to Article 300(6) EC, ECLI:EU:C:2011:123; Opinion of 18 December 2014, Opinion 2/13 pursuant to Article 218(11) TFEU, ECLI:EU:C:2014:2454.

• 16 Koen Lenaerts, Modernising Trade Whilst Safeguarding the EU Constitutional Framework: An Insight Into the Balanced Approach of Gutachten 1/17, 2019, p. 16, at https://diplomatie.belgium.be/sites/default/files/downloads/presentation_lenaerts_Gutachten_1_17.pdf.

• 17 According to Riffel 2019, p. 519, as a consequence of the CETA Opinion, a necessity test in the narrower sense under WTO law (as required by Article XX GATT, Article XIV GATS) must now be regarded as in violation of EU law. Similarly, the criticism of Patricia Sarah Stöbener de Mora & Stephan Wernicke, ‘Riskante Vorgaben für Investitionsschutz und Freihandel – Das CETA-Gutachten des EuGH’, Europäische Zeitschrift für Wirtschaftsrecht, Vol. 30, 2019, p. 973, which criticize a contradiction between the leeway parliaments have in a democracy with regard to their legislation, on the one hand, and their capacity to enter into international legal obligations, on the other hand. Such statements neglect the differences between WTO dispute settlement and CETA ICS. More on this below.

• 18 On this externalization see Heiko Sauer, ‘Europarechtliche Schranken Internationaler Gerichte’, Juristenzeitung, Vol. 74, Issue 19, 2019, p. 928 et seq.

• 19 Opinion 1/17, paras. 221 and 237.

• 20 See Judgment of 5 February 1963, Case C-26/62, Van Gend en Loos, ECLI:EU:C:1963:1, para. 12; Judgment of 15 July 1964, Case C-6/64, Costa v. E.N.E.L., ECLI:EU:C:1964:66. For more on the external autonomy see Loic Azoulai, ‘Structural Principles in EU Law: Internal and External’, in Marise Cremona (ed.), Structural Principles in EU External Relations Law, Hart Publishing, Oxford, 2018, p. 31.

• 21 Opinion 2/13, para. 158.

• 22 Opinion 1/91, para. 35.

• 23 Id. para. 46.

• 24 For the quotes see Opinion 1/17, paras. 109-110.

• 25 The CJEU emphasizes its particular importance for safeguarding these special characteristics, Opinion 1/17, para. 111.

• 26 Also the function of the national courts as Union courts is one of these essential elements of the constitutional order of the EU, see Opinion 1/09, paras. 68 et seq.; Opinion 2/13, para. 175. This is the corollary of the CJEU’s competence to secure its exclusive jurisdiction for preliminary rulings under Article 267 TFEU and thus again serves the CJEU. The role of the national courts was wrongly ignored in the CETA Opinion; see below.

• 27 See Koutrakos 2019a. For criticism of the self-referential attitude of the CJEU in previous decisions on international courts, which in effect denies legal dialogue, see Bruno de Witte, ‘A Selfish Court? The Court of Justice and the Design of International Dispute Settlement Beyond the EU’, in Marise Cremona & Anne Thies (eds.), The European Court of Justice and External Relations Law, Hart Publishing, Oxford, 2014, p. 46; Bruno de Witte, ‘The Relative Autonomy of the European Union’s Fundamental Rights Regime’, Nordic Journal of International Law, Vol. 88, Issue 1, 2019, p. 71; Jed Odermatt, ‘The Principle of Autonomy: An Adolescent Disease of EU External Relations Law?’, in Marise Cremona (ed.), Structural Principles in EU External Relations Law, Hart Publishing, Oxford, 2018, p. 316; Steve Peers, ‘The EU’s Accession to the ECHR: The Dream Becomes a Nightmare’, German Law Journal, Vol. 16, Issue 1, 2015, pp. 213-222; Riffel 2019, p. 506; Eleanor Spaventa, ‘A Very Fearful Court?’, Maastricht Journal of European and Comparative Law, Vol. 22, Issue 1, 2015, p. 47.

• 28 Opinion 1/91, para. 40.

• 29 Id. paras. 49 et seq.; Opinion 1/09, paras. 73 et seq.

• 30 See Judgment of 30 April 1974, Case C-181/73, Haegeman, ECLI:EU:C:1974:41, paras. 5 et seq.

• 31 Opinion 1/17, para. 116.

• 32 Riffel 2019, p. 513.

• 33 CJEU, Opinion 1/17, para. 117.

• 34 Id. paras. 118 et seq.

• 35 Id. paras. 120-126.

• 36 Giulia Claudia Leonelli, ‘CETA and the External Autonomy of the EU Legal Order: Risk Regulation as a Test’, Legal Issues of Economic Integration, Vol. 47, Issue 1, 2020, p. 50.

• 37 Opinion 1/17, para. 131.

• 38 See Hindelang 2015, pp. 76 et seq.; Leonelli 2020, pp. 55 et seq. The assessment required by Article 8.31(2) CETA as to whether the CETA Tribunal follows the prevailing interpretation of national law, is less a question of fact than of interpretation.

• 39 Opinion of Advocate General Bot delivered on 29 January 2019, Opinion 1/17, ECLI:EU:C:2019:72, paras. 136 and 142 (AG Bot’s Opinion). The problem that the CETA Tribunal then determines the substance of the EU act in a decisive way, since this determination is binding and unalterable in the investment protection trial, is also ignored by Advocate General Bot, see para. 143.

• 40 Opinion 1/17, para. 132.

• 41 Cf. Article 8.21(5-7) CETA.

• 42 See e.g. Article 1 of the Protocol on Mutual Recognition of the Results of Conformity Assessment regarding the notion of in-house body, or Annex I on Reservations for Existing Measures in which the EU in several instances refers to specific EU legislation. The same applies to Annex II regarding reservations for future measures.

• 43 So explicitly Opinion 1/17, para. 134.

• 44 Opinion 1/09, para. 80. Admittedly, the jurisdiction of the Patent Court to interpret relevant Union law was intended to be exclusive (see Riffel 2019, p. 512), which is not the case with the CETA Tribunal.

• 45 See Opinion 2/13, paras. 236 et seq. The CJEU raised concerns even though the Draft Accession Agreement to the ECHR provided for a preliminary referral mechanism to the CJEU, which is completely absent in CETA.

• 46 See Article 8.22(1)(f) and (g) CETA.

• 47 See para. 24 of the CJEU’s Recommendations to National Courts and Tribunals in Relation to the Initiation of Preliminary Ruling Proceedings.

• 48 See Article 8.31(2) CETA.

• 49 For a case scenario insofar see Leonelli 2020, pp. 59-61.

• 50 Opinion 1/09, para. 77; see also AG Bot’s Opinion, para. 69.

• 51 Id. paras. 94, 167 et seq.

• 52 See in this respect Opinion 1/17, para. 135.

• 53 In general international law, this is quite different, see PCIJ, Reports 1926, A, p. 19.

• 54 Opinion 1/17, para. 117; see also AG Bot’s Opinion, paras. 77 et seq.

• 55 Opinion of 16 May 2017, Opinion 2/15 pursuant to 218(11) TFEU, ECLI:EU:C:2017:376, paras. 292 et seq.; Judgment of 5 December 2017, Case C-600/14, Germany v. Council, ECLI:EU:C:2017:935, paras. 66 et seq., where the CJEU caused some uncertainty as to the nature of EU external competences (exclusive or still shared?) in the case of shared internal competences.

• 56 Case C-284/16, Achmea, paras. 50 et seq.

• 57 Opinion 2/13, paras. 187 et seq. For criticism of the different approaches see also Panos Koutrakos, ‘The Anatomy of Autonomy: Themes and Perspectives on an Elusive Principle’, in ECB (ed.), Building Bridges: Central Banking Law in an Interconnected World, 2019, pp. 96 and 98. (Koutrakos 2019b).

• 58 See also Leonelli 2020, pp. 44, 48 et seq.

• 59 Opinion 1/17, para. 119.

• 60 See Article 46 of the Vienna Convention on the Law of Treaties, which permits the invocation of a manifest violation of national law of fundamental importance.

• 61 Sauer 2019, p. 932.

• 62 Opinion 1/17, paras. 143 and 148.

• 63 Id. para. 144.

• 64 This wording has led to considerations that a one-time award of damages could not yet be a problem. What speaks against such an understanding of the CJEU, however, is the fact that in the following paragraphs of the opinion the frequency of awards does not play a role in the CJEU’s argument. The CJEU formulates its conclusions on the lack of jurisdiction of the CETA Tribunal to assess levels of protection in absolute terms.

• 65 Opinion 1/17, para. 149.

• 66 Id. para. 150.

• 67 Id. para. 151.

• 68 Id. paras. 152, 154-155, 157.

• 69 Id. para. 156.

• 70 Id. para. 153.

• 71 Id. para. 160: “Those tribunals have no jurisdiction to call into question the choices democratically made within a Party relating to, inter alia, the level of protection of public order or public safety, the protection of public morals, the protection of health and life of humans and animals, the preservation of food safety, protection of plants and the environment, welfare at work, product safety, consumer protection or, equally, fundamental rights.”

• 72 For further details see Stoll et al. 2017, pp. 118 et seq.

• 73 See Christina Eckes, ‘The Autonomy of the EU Legal Order’, Europe and the World: A Law Review, Vol. 4, Issue 1, 2020, pp. 8 et seq.

• 74 Cf. Opinion 1/17, para. 148, “without prejudice to a situation where the Parties have agreed, within the framework of the CETA, to harmonise their legislation.”

• 75 Cf. Article 290(2) sentence 2 TFEU.

• 76 Opinion 1/17, para. 160.

• 77 On EU autonomy as a rhetorical shield for the protection of the competences of the CJEU see Bruno de Witte, ‘European Union Law: How Autonomous Is Its Legal Order?’, Zeitschrift für Öffentliches Recht, Vol. 65, Issue 1, 2010, p. 150.

• 78 Opinion 1/17, para. 151.

• 79 Id.

• 80 See also Article 216(2) TFEU.

• 81 See also Article 218(11) TFEU.

• 82 Lenaerts 2019, “emphasise[d] that what the Court is protecting here is not EU measures of general application as such. Nothing in Chapter 8 of the CETA suggests that measures of that kind are ‘immune’ from review before the CETA tribunals. On the contrary, as the Court expressly confirms, such measures may give rise to the award of monetary compensation under the ISDS mechanism when, for example, they amount to discrimination or arbitrary treatment affecting a Canadian investor. That is not incompatible with EU primary law […] What the Court is protecting instead is the essence of the democratic process leading to the adoption of EU norms protecting public interests, a process which forms part of the EU constitutional framework.”

• 83 See also the criticism by Koutrakos 2019b, p. 100.

• 84 AG Bot’s Opinion, paras. 133 et seq., only considers them duties to be taken into account.

• 85 In this respect, the competence of the CETA Tribunal to award damages should be upheld, according to Lenaerts 2019.

• 86 Such differentiation is also not obsolete due to the lack of direct effect of CETA.

• 87 See also Catharine Titi, Opinion 1/17 and the Future of Investment Dispute Settlement: Implications for the Design of a Multilateral Investment Court, 2020, p. 17, at https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3530875.

• 88 See also Sauer 2019, p. 928.

• 89 See Titi 2020, p. 18.

• 90 Opinion 1/17, para. 119.

• 91 On the protection of the ‘Right to Regulate’ in CETA, see Catharine Titi, ‘Right to Regulate’, in Makane Moïse Mbengue & Stefanie Schacherer (eds.), Foreign Investment under the Comprehensive Economic and Trade Agreement (CETA), Springer International Publishing, Heidelberg, 2019, pp. 159 et seq.; Schill 2018, pp. 70 et seq.

• 92 Wolfgang Weiss, WTO Law and Domestic Regulation, Beck Hart Nomos, Munich 2020, p. 8 et seq.

• 93 Opinion 1/17, para. 146.

• 94 Opinion 1/17, para. 148: “However, without prejudice to a situation where the Parties have agreed, within the framework of the CETA, to harmonise their legislation.”

• 95 Lenaerts 2019, italicization in the original.

• 96 Id.

• 97 Opinion 1/17, para. 148.

• 98 See also Article XVI:4 of the WTO Agreement, according to which each member undertakes to bring its laws into conformity with its obligations under WTO law.

• 99 For more detail on their effects on domestic regulatory autonomy see Wolfgang Weiss, WTO Law and Domestic Regulation, pp. 246 et seq. and 289 et seq.

• 100 Opinion 1/17, para. 146.

• 101 In this sense see Sauer 2019, p. 935.

• 102 AG Bot’s Opinion, paras. 176 and 178; de Witte 2014, p. 45. However, this does not imply a constitutional obligation for the EU to submit to any kind of international jurisdiction without limits.

• 103 Developing this further was beyond the scope of this paper. For more on this see Wolfgang Weiss, forthcoming in Europarecht, Vol. 55, Issue 6, 2020.