The recent, widely-reported ‘legal scrub’ of the Canada-EU Comprehensive Economic and Trade Agreement (CETA) has drawn attention for its endorsement of a radical shift away from the model of investor-state dispute settlement that has prevailed in investment agreements to date. The new text indicates that Canada has agreed to the EU’s proposals on an investment court system, with a permanent roster of arbitrators appointed by Canada and the EU, rather than ad hoc tribunals whose members are appointed by the disputing parties themselves. In another innovation, CETA will also include an appeals mechanism, which will have power to review the merits of first-instance rulings, going beyond the limited grounds for annulment of awards in the existing ICSID system.
Alongside these revolutions, the new CETA text also contains another change from the earlier text. Under the heading of ‘Applicable law and interpretation’, Article 8.31(2) of the new text provides:
The Tribunal shall not have jurisdiction to determine the legality of a measure, alleged to constitute a breach of this Agreement, under the domestic law of the disputing Party. For greater certainty, in determining the consistency of a measure with this Agreement, the Tribunal may consider, as appropriate, the domestic law of the disputing Party as a matter of fact. In doing so, the Tribunal shall follow the prevailing interpretation given to the domestic law by the courts or authorities of that Party and any meaning given to domestic law by the Tribunal shall not be binding upon the courts or the authorities of that Party.
Although the provision is new in CETA, it has also recently appeared in the EU-Vietnam FTA and in similar language in the EU’s November 2015 TTIP proposals. While this might suggest that the provision is a recent invention of the EU, its inspiration in CETA could equally have come from Canada, which included a similar provision in its 2008 FTA with Colombia. In fact, Colombia itself appears to have first spearheaded the provision, including language on domestic law broadly similar to the provision’s first sentence in its 2007 Model BIT and in agreements signed as far back as 2006 with Japan, the UK, India, Belgium, China, Peru and Switzerland.
In the CETA formulation, tribunals are apparently to be barred from determining whether host state measures are legal under domestic law. But there are instances in which an investment tribunal may need to decide precisely this question. Cases where the termination of a contract is in issue, such as Malicorp v Egypt, might require a determination of whether the contract was lawfully terminated under domestic law, in order to confirm whether any rights susceptible of expropriation persist. A tribunal might need to confirm whether the state complied with domestic legal procedures when expropriating the investor, in order to decide whether the expropriation was done ‘under due process of law’, as Article 8.12(1)(b) requires. Examination of a local court judgment’s consistency with domestic law might form part of a determination of whether the local court engaged in a clear and malicious misapplication of domestic law, amounting to a denial of justice and a breach of CETA Article 8.10(2)(a).
One response to this is that CETA may treat such domestic law analysis as merely incidental to a principal finding of breach of international law. On this view, what is prohibited by Article 8.31(2) is a principal finding by the tribunal that the respondent breached domestic law; CETA does not prohibit such incidental findings of domestic illegality en route to a principal determination. But if this is so, Article 8.31(2) simply replicates Article 8.18(5), which already provides that the tribunal has no jurisdiction to decide (ie, make principal determinations on) anything other than claimed breaches of CETA’s substantive investment protections.
A second response is that Article 8.31(2) goes on to clarify that the tribunal may, nevertheless, determine domestic legality as a question of fact. (CETA’s treatment of domestic law as fact is reiterated in Article 8.28, which grants the new Appellate Tribunal power to review an award based on ‘manifest errors in the appreciation of the facts, including the appreciation of relevant domestic law’.) Certainly, this appears to align with the PCIJ’s oft-repeated position that, ‘from the standpoint of International Law […], national laws are merely facts’ (Certain German Interests in Polish Upper Silesia, at 19). But, as Jenks recognised even in 1938, it would be ‘a mistake to attach undue importance’ to this position. Investment tribunals (and other international courts and tribunals) have often interpreted and applied domestic law when necessary, and it is not clear that treating this domestic law as fact has made or would make much difference to the tribunal’s reasoning process. After all, a party seeking to rely on a point of customary international law must prove that point by reference to sufficient evidence (of state practice and opinio juris), and yet custom is clearly law, not fact. As Ketcheson observes (at 57), investment tribunals have been known to hear expert witness evidence on questions of international law (for example, on the availability of restitution in von Pezold v Zimbabwe, and on denial of justice in Chevron v Ecuador), and yet the presence of expert witnesses did not remove the legal character of the tribunal’s determination on those points. In deciding a question of domestic law, tribunals must typically apply that law to underlying facts; it is not entirely natural to treat this process instead as an instance of applying facts to facts.
Thus, to paraphrase the SPP v Egypt tribunal (Decision on Jurisdiction, ):
‘[CETA’s] contention that […] municipal law should be treated as a “fact” is not helpful. [When disputing] parties are in fundamental disagreement as to what [a provision of domestic law] means […] the Tribunal therefore must interpret [that provision] and determine its legal effect’.
Given this, the expressed bar on determination of domestic legality is likely to be of little consequence.
CETA also provides that, when considering a question of domestic law (as fact), the tribunal ‘shall follow the prevailing interpretation given to the domestic law by the courts or authorities of that Party’ (emphasis added). This language appears to be mandatory, with no discretion granted to the tribunal. However, situations may exceptionally arise in which the investor might allege that the courts or authorities of the host state are not to be trusted. Under influence from the executive, local courts may have issued a self-serving ruling that, for instance, a contract was lawfully terminated. The tribunal will nevertheless be bound to defer to the problematic local ruling, and therefore bound to find that it has no jurisdiction because the investment disappeared when the contract terminated.
The CETA parties might be confident that such a situation will never arise under the prevailing standards of good governance in Canada and the EU. But CETA (and TTIP, which also contains the clause) have been described as ‘gold standard’ agreements, which may become a model for new investment treaties involving states with more concerning records of governance. A better view on the issue is the one already acknowledged by investor-state tribunals (eg Helnan v Egypt at ), WTO panels (eg US – Section 301 at [7.19]), the ECHR (eg Kononov v Latvia at ), the PCIJ (eg Brazilian Loans at 124) and the ICJ (eg Diallo at ): while international adjudicators must pay the utmost regard to domestic interpretations of local law, exceptional circumstances may call for departure from this.
Conversely, the provision also declares that ‘any meaning given to domestic law by the Tribunal shall not be binding upon the courts or authorities of that Party’. This probably does no more than restate the general position. For one thing, investment tribunals sometimes need to consider the domestic law of a state other than the respondent state (eg the consideration of US and Canadian law in the EuroGas v Slovakia pleadings). CETA does not envisage this possibility, but it will be unavoidable in some circumstances, given the range of laws potentially relevant to a cross-border investment operation. A finding on a particular state’s domestic law could undoubtedly not bind that state where it is not party to the dispute (or even party to the relevant investment treaty). Even where it is the respondent state’s law in issue, it has never been suggested that a tribunal’s findings on that law would bind respondent state courts, whether those findings are treated as fact or law (a point also recognised by Jenks, at 71). To the contrary, international adjudicators frequently recall that national courts remain the authoritative interpreters of their own law. While Canadian and EU authorities are bound to comply with and enforce awards of the CETA tribunal under Article 8.41, it is only the award’s dispositif that is ‘binding between the disputing parties and in respect of that particular case’. The meaning given to international law by the tribunal cannot be directly binding on other international or national courts; arbitral awards are not the primary sources of international law. Equally, the meaning given to domestic law cannot be binding either, absent a national rule that ascribes a place to international arbitral decisions in the sources of national law.
Canada may have used something similar to CETA’s domestic law clause first, but its appearance in CETA is more likely intended to satisfy the other contracting party, the EU, that the tribunal will not interfere with the ECJ’s claimed monopoly on interpretation of EU law (as domestic law in cases against the EU or Member States). However, the existence of this monopoly has already been doubted by the EURAM v Slovakia (at ) and Eureko v Slovakia (at ) tribunals, and the clause’s language may simply run along similar lines to the common insistence of NAFTA tribunals that they are not ‘fourth instances’, sitting in appeal over host state decisions on domestic law (eg ADF Group v USA at ). Ultimately, while any effort to encourage investment arbitrators to take greater care with issues of domestic law is to be welcomed, CETA’s domestic law clause may have more political than legal import – as perhaps given away by its preface of ‘for greater certainty’.