CETA ISDS Mechanism Compatible with EU Law: What Implications?
This article has jointly been co-authored by Isabelle Van Damme and Quentin Declève
On 30 April 2019, the Court of Justice of the European Union (CJEU) decided in Opinion 1/17 that the chapter on investor-State dispute settlement (ISDS) in the Comprehensive Economic and Trade Agreement between Canada, of one part, and the European Union and its Member States, of the other part (CETA) is compatible with EU primary law. On 29 January 2019, Advocate General Bot had already reached the same conclusion (for an analysis, see here).
Opinion 1/17 removes a significant obstacle to the ratification of CETA by the EU Member States and the ratification of investment protection agreements with, for example, Singapore and Vietnam, which contain similar chapters on ISDS. The Opinion also significantly boosts the European Union negotiating position in the ongoing United Nations Commission on International Trade Law (UNCITRAL) negotiations on ISDS reform. At the same time, the Opinion might, to some extent, tie the hands of the European Union in negotiating in that forum.
CETA is a mixed agreement and therefore requires ratification both at EU level and by all EU Member States. Following complications in its ratification process, Belgium sought an Opinion of the CJEU on whether the investment court system (ICS) in Chapter Eight, Section F, of CETA, regarding the “Resolution of investment disputes between investors and States” is compatible with EU law, in particular with the autonomy of the EU legal order and fundamental rights.
The ICS in CETA is a new model for resolving disputes between foreign investors and States (or the European Union). The European Union has designed that model with the objective of addressing concerns that the traditional form of ISDS does not guarantee fundamental rights and values relating to, notably, the independence of arbitrators, legitimacy, access to courts, transparency, and fails to result in a coherent body of case-law regarding the interpretation and application of investment protection standards. Under the ICS model, investor-State disputes arising under CETA will be heard by a permanent tribunal composed of independent and publicly appointed members of a first instance tribunal and, in case of an appeal, an appellate tribunal. The so-called CETA Tribunal, composed of 15 members, will hear cases in divisions of three members appointed at random. The ICS model also introduces several other significant procedural innovations (mostly based on the procedures governing WTO dispute resolution) that signal a fundamental departure from the current ISDS mechanism used for resolving disputes under existing investment agreements. Whilst the European Union is insisting on the inclusion of this ICS model in new investment protection agreements which it is now negotiating, it also advocates the establishment of a multilateral investment court (MIC), which builds on the ICS model, in the ongoing negotiations in UNCITRAL Working Group III. Those negotiations started in November 2017. Having finalised a work plan in early April 2019, the UNCITRAL Working Group III will now start discussing various models for reform, including the MIC. In a paper dated 18 January 2019, the European Union explained how it envisages the design and operation of a MIC. Significantly, the European Union, the EU Member States and Canada have already included in CETA a commitment that “[u]pon establishment of such a multilateral mechanism, the CETA Joint Committee shall adopt a decision providing that investment disputes under this Section will be decided pursuant to the multilateral mechanism and make appropriate transitional arrangements” (Article 8.29 of CETA). In Opinion 1/17, the CJEU took note of that fact.
This client alert discusses the main elements of the Opinion and identifies the key takeaways from Opinion 1/17. In particular, it discusses the implications of Opinion 1/17 for: (i) the European Union’s position as a party to international agreements establishing international courts and tribunals; (ii) the UNCITRAL negotiations on the MIC; (iii) intra-EU bilateral investment agreements post-Achmea; and (iv) existing bilateral investment agreements between Member States and third States.
The ICS model under CETA respects the autonomy of the EU legal order and the jurisdiction of the CJEU
As a matter of EU law, it is settled that the European Union may, in principle, become a party to an international agreement providing for the establishment of a court or tribunal having competence to interpret that agreement and deliver decisions that bind the European Union as long as doing so would not produce adverse effects on the autonomy of the EU legal order. The autonomy of the EU legal order refers to the fact that the European Union has its own legal system that is autonomous and independent from that of the Member States and that may not be undermined by, in particular, the European Union’s external actions. One important feature of that autonomy is that the CJEU enjoys exclusive jurisdiction to interpret EU law and to decide on the validity of acts of the EU institutions. For that reason, the CJEU has previously found that the European Union may not become a party to, for example, the European Court of Human Rights (Opinion 2/13) or the European and Community Patents Court (Opinion 1/09), or that agreements between Member States may be contrary to EU law (for example, Case C-284/16 Achmea).
The central question put to the CJEU was therefore whether a mechanism through which the resolution of disputes between investors and States (and the European Union) is subjected to judicial adjudication by a CETA Tribunal and Appellate Tribunal (and possibly a MIC) has an adverse effect on the autonomy of the EU legal order, meaning in essence, the unique constitutional framework of the European Union which includes the judicial system aimed at ensuring that EU law is interpreted consistently and uniformly (see paras 110 and 111). The answer to that question depends on whether, pursuant to the international agreement at issue, the tribunal may or must interpret EU law.
The CJEU started its analysis by emphasizing that the ICS model under CETA established a judicial system separate from that of Canada, the EU Member States and the European Union. It explicitly recognised that its jurisdiction to interpret and apply international agreements (which form part of EU law) “does not take precedence over either the jurisdiction of the courts and tribunals of the non-Member States with which those agreements were concluded or that of the international courts or tribunals that are established by such agreements” (para. 116). In addition, the CJEU also underscored the importance of the power of the European Union to enter into international agreements that confer powers on an international court or tribunal to interpret the agreement without being bound by interpretations adopted by the courts of the parties because of, on one hand, the reciprocal character of international agreements such as CETA and, on the other, the need for the European Union to maintain powers in international relations (para. 117). As long as such international courts do not have the power to interpret or apply EU law other than the provisions of the agreement themselves or to make awards that might prevent the EU institutions from operating in accordance with the EU constitutional framework, EU law does not preclude the establishment of those international courts (para. 118).
In light of those considerations, the CJEU next examined the text of CETA.
First, the CJEU found that tribunals under CETA do not have any power to interpret or apply EU law other than, pursuant to Article 8.31.1 of CETA, the power to interpret or apply the provisions of the agreement having regard to the rules and principles of international law applicable between the Parties. Furthermore, it was important that those tribunals, pursuant to Article 8.31.2 of CETA, do not have jurisdiction to decide on the legality of a measure, alleged to constitute a breach of CETA, under the domestic law of a Party to CETA. Both provisions guaranteed that, unlike the case in, for example, Opinion 1/09 (or Case C-284/16 Achmea), the (appellate) tribunals under CETA would not have any power to interpret and apply EU law (paras 122 and 133).
Showing a distinct awareness that its judgment in Achmea had resulted in considerable uncertainty as to the limits of the principle of the autonomy of the EU legal order, the CJEU also underscored that Achmea concerned an agreement between two Member States which are bound by the principle of mutual trust requiring them “to consider, other than in exceptional circumstances, that all the other Member States comply with EU law, including fundamental rights, such as the right to an effective remedy before an independent tribunal laid down in Article 47 of the Charter [of Fundamental Rights of the European Union]” (para. 128). However, that principle does not apply in relations between the European Union and third States (para. 129).
Another concern raised was whether a tribunal under CETA, when considering whether the European Union had acted in accordance with its obligations under CETA, would nonetheless need to interpret EU law in establishing the facts at issue. In that regard, the CJEU recognised the well-established principle under international law that, in the adjudication of disputes relating to the interpretation and application of international law, the domestic law of a Party is considered as a question of fact. That principle is expressly confirmed in Article 8.31.2 of CETA (para. 130) which adds that the tribunal would need to accept the prevailing interpretation given to domestic law by national courts or tribunals and that any interpretation of domestic law by the tribunal would not bind national courts. In that context, the CJEU distinguished between the interpretation of EU law as such and the examination of the effect of an EU measure challenged by an investor (para. 131). The CJEU also accepted that the Appellate Tribunal’s power to identify manifest errors in the appreciation of facts and thus in the appreciation of the interpretation of domestic law does not confer power to interpret domestic law (para. 133).
Another context in which international courts and tribunals might be called upon to interpret EU law is that where it needs to decide whether the European Union and/or the Member States is to be the respondent. That matter arguably depends on EU law given that the Treaties lay down the vertical allocation of competences. In that regard, the CJEU underlined that Article 8.21 of CETA precludes tribunals from determining whether the European Union or a Member State should be the defendant in a dispute involving a claim made by a Canadian investor. That determination remained within the powers of the European Union, including the CJEU which enjoys exclusive power to decide on the allocation of powers between the European Union and the Member States (para. 132). That feature distinguished CETA also from the draft agreement on the accession of the European Union to the European Court of Human Rights at issue in Opinion 2/13.
Considering those features, the CJEU accepted that there was no need for CETA to make “provision for the prior involvement of the Court that would permit or oblige that Tribunal or Appellate Tribunal to make a reference for a preliminary ruling to the Court” (para. 134) or to establish “any procedure for the re-examination of the award by a court of [the host State] or by the Court and without that investor being permitted […] to bring, during or on the conclusion of the procedure before those Tribunals, the same dispute before a court of that State or before the Court” (para. 135).
Second, the CJEU accepted that the awards of tribunals under CETA would not prevent the EU institutions from operating in accordance with the EU constitutional framework. Here, the focus of the CJEU’s analysis was on whether those tribunals could call into question, especially in assessing a defence advanced by the European Union in response to an alleged breach of the substantive investment protections under CETA, the European Union’s right to define the level of protection of public interests under EU law. The starting point of that analysis was that the jurisdiction of those tribunals might be particularly wide given the broad definitions of an investment and investor and the type of measure that may be challenged by an investor. The CJEU also considered that the tribunals are expressly precluded from, as is common in international adjudication, to annul the contested measure. At the same time, unlike the compliance obligations under World Trade Organization (WTO) law, the CJEU emphasised that tribunals under CETA may award damages to a private investor (paras 144 and 145).
Although all those features were found to be consistent with the protection of foreign investors (para. 147), the CJEU signalled that the autonomy of the EU legal order would nonetheless preclude an agreement providing for tribunals that might call into question the level of protection of a public interest leading to the adoption of the measure (paras 148 and 149). The review of the compatibility of the level of protection of public interests established by EU legislation with EU law is a prerogative of the CJEU (para. 151). However, CETA did not undermine that prerogative, given that, taking into account multiple express clauses in CETA confirming the inherent right to regulate, the tribunals have no jurisdiction to declare the level of protection of a public interest established by EU measures incompatible with CETA (paras 152 to 156). In other words, CETA guaranteed the right to regulate economic activity in the public interest (para. 155).
The ICS model under CETA respects the principle of equality
A separate set of doubts raised in Belgium’s request for an Opinion related to whether granting a Canadian investor a remedy against the European Union that is not available to a European investor constitutes a discrimination based on nationality.
The CJEU repeated that the principle of non-discrimination under the first paragraph of Article 18 of the Treaty on the Functioning of the European Union (TFEU) and Article 21(2) of the Charter is not intended to protect nationals of third States. In other words, that principle does not apply in case of a difference in treatment between nationals of Member States and third States (para. 170). By contrast, the scope of the principle of “everyone is equal before the law” under Article 20 of the Charter is not similarly limited. The CJEU found that it applies to all situations governed by EU law and thus also to situations “falling within the scope of an international agreement entered into by the European Union” (para. 171). Even though that principle does not require the European Union, in its external relations, to accord equal treatment to different third States, the CJEU nonetheless examined whether the ICS model under CETA treats differently, within the European Union itself, EU investors and investors of third States.
The CJEU accepted that there was a difference in treatment between, on one hand, enterprises and natural persons of Member States investing within the European Union and subject to EU law, and, on the other hand, Canadian enterprises and natural persons investing within the same commercial or industrial sector of the EU internal market (para. 179). Only the latter may resort to the ICS under CETA. However, both categories were not comparable given that the latter are foreign investors, but the former are not (para. 181). The CJEU did not elaborate on that distinction. Furthermore, damages awarded to an enterprise established within the European Union, and owned or controlled by a Canadian investor, are granted to an investment of that Canadian investor (para. 183).
The CJEU’s conclusion that Canadian investors and investors of EU Member States within the European Union are not comparable was also not affected by the fact that the tribunals may issue an award stating that a fine imposed by the European Commission or by a competition authority of a Member State on a Canadian investor, because of an infringement of the main competition principles found in Articles 101 and 102 TFEU, breaches CETA (para. 184). According to the CJEU, CETA offered sufficient guarantees to ensure that such an award would only be possible where the decision imposing a fine would violate the fair and equitable treatment obligations under CETA or the fundamental right of property. At the same time, such an award would be “unimaginable” where the Commission or the competition authorities of the Member States have “correctly applied” competition rules (para. 185). In addition, an EU investor could resort to EU remedies for obtaining the annulment of the fine. In other words, both an EU investor and Canadian investor have remedies available to them and therefore there is no situation of unequal treatment. Overall, the CJEU’s reasoning is primarily based on the assumption that EU competition law as such always respects the fair and equitable treatment obligations under CETA or the fundamental right of property.
The ICS model under CETA respects the requirement that EU (competition) law must be effective
Next, the CJEU examined whether awards issued by the ICS would undermine the effectiveness of EU competition law. The specific question addressed by the CJEU was whether the fact that the ICS might issue an award that could, theoretically, cancel out the effect of a competition law fine imposed by the EU Commission (or by a competition authority of a Member State) impedes the full application of EU competition rules. Where a decision imposing a fine would violate the fair and equitable treatment obligations under CETA or the fundamental right of property, the CJEU considered that the fact that an award could cancel out the fine did not create the risk of undermining EU competition law because EU law itself allows the annulment of a competition law fine vitiated by a defect similar to those that could be identified by the ICS (para. 187).
The ICS model under CETA guarantees the right of access to an independent tribunal
In the remaining part of its Opinion, the CJEU considered whether the design of the ICS complied with fundamental rights and, in particular, the right to an accessible and independent tribunal.
The CJEU emphasised that international courts and tribunals under international agreements to which the European Union is a party, such as CETA, must be accessible and independent (para. 191). The CJEU linked those values also to the objective of free and fair trade pursued by CETA (para. 200). What matters, as a matter of EU law, is that the safeguards under EU (fundamental rights) law are guaranteed. The fact that the third State with which the European Union concluded the agreement is not bound by those safeguards is immaterial (para. 192). Indirectly, the CJEU signalled that the same safeguards must be respected in relation to traditional arbitration mechanisms in relation to investments (para. 193).
In examining the features of the ICS model under CETA, the CJEU recognised that that mechanism, while based on traditional investment arbitration mechanisms, also showed features relating to the composition of the tribunals and the treatment of cases that are distinct (paras 193 and 194). In particular, those distinct features included: the establishment of a permanent tribunal established by law and applying rules of law, the tribunals will exercise their functions wholly autonomously, the decisions of the tribunals will be final and binding, the composition of a division in a random and unpredictable manner, the availability of an appeal mechanism, and the compulsory character of the jurisdiction of the tribunals (paras 195 to 198). Those features showed that the ICS Chapter under CETA establishes tribunals that will effectively exercise judicial functions (para. 197), reflecting also the intention of the parties, expressed in a Joint Interpretative Instrument, that “CETA moves decisively away from the traditional approach of investment dispute resolution and establishes independent, impartial and permanent investment Tribunals, inspired by the principles of public judicial systems” (para. 195).
The CJEU’s assessment of whether the ICS model under CETA is accessible focused primarily on the financial burden for investors, including natural persons and small and medium-sized enterprises, when using that mechanism. As a general principle, the CJEU accepted that their access to the system could be limited provided that any restrictions (including those relating to court costs) are proportionate, pursue a legitimate aim and do not adversely affect the very essence of the right of access (para. 201). The financial burden of using the ICS comprised the costs of legal representation and assistance and the cost of the proceedings (paras 209 and 210). The CJEU held that “in the absence of rules designed to ensure that the CETA Tribunal and Appellate Tribunal are financially accessible to natural persons and small and medium-sized enterprises, the ISDS mechanism may, in practice, be accessible only to investors who have available to them significant financial resources” (para. 213). The CJEU therefore continued to examine whether CETA offered guarantees that a body of rules ensuring the level of accessibility required by EU law would be put into place as soon as those tribunals are established (para. 214). It found that, by means of Statement No 36, the Commission and the Council had committed to implement rapidly and adequately Article 8.39.6 of CETA which provides that “[t]he CETA Joint Committee shall consider supplemental rules aimed at reducing the financial burden on claimants who are natural persons or small and medium-sized enterprises” (para. 218). It further underscored that the conclusion of CETA by the Council “is envisaged subject to the premiss that the financial accessibility of the CETA Tribunal and Appellate Tribunal for all EU investors will be ensured” (para. 221). As a result of those commitments, the CJEU was satisfied that sufficient guarantees were in place so that access to tribunals under CETA would be accessible (para. 219).
Finally, in assessing whether the ICS model under CETA is compatible with the requirement of external independence, the CJEU highlighted the following features as being relevant to the external independence of the ICS: fixed term appointments for members of the tribunals, the need for specific expertise, remuneration commensurate with the importance of the members’ duties, the availability of protections against the removal of members (paras 223 to 226). The fact that the CETA Joint Committee – comprising representatives of both Parties and deciding by mutual consent – enjoyed certain powers with respect to the appointment and removal of members, the remuneration of members and the adjustment of the number of members was not an obstacle (paras 227 to 230). Indirectly, the CJEU endorsed that the ICS model under CETA could evolve towards “the gradual establishment of a court composed of Members who will be employed full-time” (para. 231). At the same time, the CJEU emphasised that the consent of the European Union in the decision-making process in the CETA Joint Committee (and thus the decision of the CETA Joint Committee themselves) would need to respect EU primary law (para. 235). It followed that those decisions could have no retroactive effect and no direct effect on pending cases (even if CETA itself did not provide for that limitation) (para. 237).
The CJEU also accepted that the ICS model under CETA satisfied the requirement of internal independence, with an emphasis on the need for impartiality, the maintenance of an equal distance from the parties to the dispute and the absence of any personal interest of the members in the outcome of the proceedings. The features of the ICS model that were deemed relevant in that respect were: the random and unpredictable composition of a division, the requirement that members be impartial and independent from the initiation of a dispute until its termination, the composition of the remuneration (monthly retainer fee and fees calculated based on the days of work devoted to a dispute) and the requirement that members may not be affiliated with any government and may not take instructions from any organisation or government with regard to matters related to the dispute (paras 238 to 244).
Key Takeways from Opinion 1/17
Implications for the European Union’s position as a party to international agreements establishing international courts and tribunals
First, the CJEU’s Opinion eliminates to a considerable extent doubts as to whether the European Union may become a party to international agreements subjecting the European Union to the jurisdiction of an international court regarding the resolution of disputes on the interpretation and application of those agreements. In that regard, the CJEU’s recognition of the importance of reciprocity in international relations and of the fact that the European Union’s external relations may be seriously hampered if it cannot agree to be subject to the jurisdiction of an international court is welcome albeit self-evident.
Provided that such tribunals may not interpret or apply EU law, as part of the applicable law in resolving the dispute, as part of its assessment of the facts or as part of its powers to identify the respondent in a dispute, the autonomy of the EU legal order is safeguarded and there is no need to envisage that, for example, the international court may request a preliminary ruling from the CJEU. However, where such conditions are not satisfied because, for example, EU law forms part of the applicable law or is incorporated in the Parties’ obligations under the agreement (see, for example, Association Agreement between the European Union and its Member States, of the one part, and Ukraine, of the other part, discussed here), it remains undecided whether the inclusion of a preliminary ruling procedure in the international agreement is either necessary or sufficient to avoid adverse effects on the autonomy of the EU legal order.
Second, CETA and other recent investment protection agreements negotiated by the European Union are unique in that they establish a detailed procedural framework within which ICS tribunals are to operate. Examples include clauses relating to the applicable law, the standard of review in assessing domestic law as a fact or the recognition of the right to regulate. Some international courts and tribunals established by an agreement to which the European Union is a party, notably the WTO dispute settlement system, do not operate within the confines of such a procedural framework. Instead, their position as to, for example, the treatment of domestic law, the definition of the applicable law, the identification of a respondent or the standard of review in assessing the exercise of a party’s right to regulate (subject to conditions laid down in the agreement) is established by means of case-law. Assessing their compatibility with the autonomy of the EU legal order must take that factor into account. In practice, the assessment by an international court of, for example, domestic law or the conditions under which a party may affect in protection of the public interest might be less straightforward than envisaged in CETA or by the CJEU. At the same time, as reform of such international courts and tribunals, especially the WTO dispute settlement system, is underway, the ICS chapter under CETA might, as a result of Opinion 1/17, inform or even cross-fertilise the negotiating position of the European Union in those fora.
Third, as a result of the specific design and operation of the ICS under CETA distinguishing ICS from ISDS, the CJEU’s conclusion that that model guarantees the right of access to an independent court does not resolve whether traditional ISDS mechanisms, which are provided for in older agreements (including the Energy Charter Treaty), are compatible with the need to safeguard EU fundamental rights and values. The final part of Opinion 1/17 could be read as setting a “floor”. Thus, putting aside whether those older investment agreements adversely affect the autonomy of the EU legal order and especially the jurisdiction of the CJEU, Opinion 1/17 calls into doubt whether traditional ISDS mechanisms are compatible with EU fundamental rights.
Implications for the UNCITRAL negotiations on the MIC
Opinion 1/17 encourages the European Union to further pursue its agenda of reforming traditional ISDS by establishing a MIC. In different parts of the Opinion, the CJEU expressly refers to the evolution from traditional ISDS towards a judicial system for resolving disputes between foreign investors and States. The CJEU even expressly included the MIC in stating that EU law does not preclude the European Union from agreeing to the establishment of international courts and tribunals which have no power to interpret or apply provisions of EU law or to make awards that might prevent the EU institutions from operating in accordance with the EU constitutional framework. In addition, the CJEU appeared to hint at the possibility that traditional ISDS may fall short of the requirements which EU primary law imposes on the European Union in negotiating and concluding international agreements establishing courts and tribunals.
At the same time, the CJEU’s detailed assessment of specific provisions in CETA suggests that the absence of similar provisions in any agreement establishing a MIC or otherwise reforming ISDS might result in obstacles for the European Union to become a party to such an agreement. In other words, Opinion 1/17 appears to tie to some extent the hands of the European Union in negotiating in that forum.
Implications for intra-EU bilateral investment agreements post-Achmea
The CJEU emphasised that the Achmea analysis must be read as applying specifically to agreements concluded between EU Member States which are bound by the principle of mutual trust. That principle is specific to EU law and informs the entire mechanism upon which EU remedies are built. It finds no application in relations between the European Union (and the Member States), on one hand, and a third State, on the other. The CJEU’s insistence on the intra-EU aspect and the principle of mutual trust could be read as signalling that it shares the position of the Commission that the Achmea analysis applies also to the Energy Charter Treaty with respect to cross-EU investments, given that that treaty applies between EU Member States.
Implications for existing bilateral investment agreements between Member States and third States
Despite the focus of the Achmea judgment on intra-EU bilateral investment agreements and of Opinion 1/17 on EU investment agreements with third States, both decisions also affect existing bilateral investment agreements between Member States and third States. Insofar as the latter contain applicable law clauses that include the domestic law of the parties, such clauses will need to be amended.
Given that any future MIC would only have jurisdiction regarding claims arising out of a pre-existing investment agreement if both State parties to that agreement have ratified the multilateral treaty establishing the MIC, the negotiations in UNCITRAL Working Group III will likely need to address whether that same multilateral treaty (or a related one) may be used to amend clauses on the applicable law and possibly other provisions in existing bilateral investment agreement.