The Path to Reform of ISDS: What Role for National Courts?

Abstract

This chapter analyzes the possible role which national courts could play in the main reform scenarios which States are currently considering, i.e., if investment arbitration is (i) reformed through targeted adjustments, (ii) supplemented with an appellate mechanism, (iii) replaced by a multilateral investment court, or (iv) supplanted by inter-State mechanisms. Each of these reform scenarios raises specific challenges when it comes to their articulation with domestic courts.

4.1 Introduction

Chapter 3 has reviewed the areas of interaction between investment arbitration tribunals and national courts in the current framework. This Chap. 4 examines how the role of national courts may change under each of the main scenarios for reform of the investor-State dispute settlement system that States are presently considering. The design options and legal challenges associated with these reform proposals have been analyzed elsewhere,¹ as has the potential of each of them to address the alleged concerns with the existing system.² With a view to taking that analysis further, this chapter seeks to outline what (if any) role domestic courts may play within each of the reform proposals.

To that end, we will consider four reform scenarios which envisage maintaining or establishing a mechanism for the resolution of disputes concerning an investment on the international plane, namely:

• a. Improvement of the current investor-State arbitration system (“investment arbitration improved”) (infra at Sect. 4.2);

• b. Addition of an AM to the current investment arbitration regime (infra at Sect. 4.3);

• c. Introduction of a MIC (with or without a built-in appeal) (infra at Sect. 4.4); and

• d. Replacement of the current system with SSDS (infra at Sect. 4.5).

The options just mentioned are the main ones advanced in the discussions around reform of the investor-State dispute settlement system³ and reflect the principal alternatives available for the design of dispute settlement systems. They also represent the broad spectrum of positions and views expressed in recent State practice and in the debate surrounding investment arbitration.

Reforming investor-State dispute settlement with an eye to considering the appropriate role that domestic courts may or should play within these different reform scenarios requires examining primarily the models of jurisdictional coordination between national and international fora and the role of national courts in support and control of these international fora (if any). The current framework of interaction between national and international tribunals, discussed supra at Sects. 3.2 and 3.3, may thus serve as a useful starting point to delineate relations between national and international tribunals for the reform efforts. In the following analysis, however, it will be seen that many of the existing rules examined earlier in this study would require adaptation to the new institutional settings envisaged by some of the reform options.⁴ Beyond those already examined, two other “jurisdiction-regulating” models (to use Yuval Shany’s terminology)⁵ and their potential applicability to the investment framework will be reviewed, i.e. (i) preliminary rulings from domestic courts to international tribunals; and (ii) complementarity between domestic and international courts (infra at Sect. 4.6).

In addition to the four reform options enumerated in para. 176 above which entail an international remedy (whether investor-State or State-to-State), a fifth possible reform outcome is sometimes advocated by certain stakeholders, i.e. the replacement of the current system with domestic courts only, for all or some categories of disputes concerning investments. This fifth reform scenario will also be considered (infra at Sect. 4.7).

Before delving into each of these reform options, two observations are in order. First, in terms of system design, not only can international mechanisms be combined with national courts in various ways (which is the subject of this chapter), international mechanisms may also be combined with other international mechanisms. For instance, States could design dispute settlement mechanisms in which investment arbitration is combined with SSDS for certain questions of interpretation of the IIA.⁶ This latter type of combination is not systematically examined as it falls outside the scope of this study. Second, the five reform scenarios that have been chosen for discussion in this chapter are limited to dispute settlement mechanisms that lead to a binding decision. This means that this study does not examine methods such as mediation, conciliation, ombudsman, etc. This limitation in no way implies any judgment on the usefulness of these alternative mechanisms, which may well deserve to be the subject of a further study. It should also be noted that in any event non-binding methods of dispute resolution are usually combined with one of the binding options discussed here.

Finally, in terms of instruments for effecting the changes that are discussed in the following sections, depending on the reform option chosen these may involve (i) the amendment of IIAs; (ii) the negotiation of an opt-in Mauritius Convention-type plurilateral treaty able to effect changes en bloc for a number of treaties; and/or (iii) the negotiation of statutes or constitutive treaties for the establishment of new international bodies (the AM Statute or MIC Statute).⁷

4.2 Investment Arbitration “Improved”

One option that is currently being considered is the reform of investment arbitration through “incremental” changes or improvements, i.e. reform short of creating new standing bodies. “Improving” investment arbitration in this manner may for instance include effecting changes in respect of the appointment of and rules of conduct for arbitrators. This may include providing for appointment predominantly by arbitral institutions or effected jointly by disputing parties, creating a roster-system, adopting ethical rules, reinforcing procedures for the dismissal of frivolous claims, enhancing transparency of proceedings and the like. These more “limited” changes will in and of themselves not affect the relationship between the “improved” investor-State arbitral mechanism and the domestic courts. Thus, the various modes of interaction between national courts and international tribunals described supra in Chap. 3 would continue to apply.

The panoply of solutions on the coordination between investment tribunals and domestic courts described previously can, however, guide States wishing to recalibrate the coordination of investment arbitration and domestic courts in a different way from what they have done so far, in line with their policy preferences.⁸ States may for instance introduce, remove, or re-modulate exhaustion of local remedies requirements, domestic litigation requirements, fork-in-the-road clauses or waiver clauses in their existing and future IIAs. As mentioned previously, certain more recent treaties already formulate fork-in-the-road and waiver clauses with a view to remedying certain shortcomings arising from what have been viewed as excessively formalistic applications of the so-called triple identity test.⁹

4.3 Investment Arbitration + Appeal

This reform option envisages the creation of an AM for awards rendered in investor-State arbitration proceedings.¹⁰ The creation of an appeal layer does not in and of itself have an effect on the relationship between the first-instance arbitral jurisdiction and domestic courts, which absent specific rules would continue to be governed by the framework provided under the specific IIA. Thus, for instance, if an IIA provides for an 18-month domestic litigation requirement as a pre-condition to accessing the arbitral tribunal, the addition of an appeal will in and of itself not affect such requirement which will continue to exist as far as the first level of jurisdiction is concerned.

Nevertheless, the introduction of an AM may significantly affect the role of domestic courts in controlling the arbitration, especially at the annulment and enforcement stages. These aspects will need to be carefully examined if and when an appellate mechanism for investor-State arbitral awards is established. The legal issues to be considered in this context are significant and require taking into account the distinctions between ICSID and non-ICSID arbitrations, which are subject to different legal regimes.¹¹ This holds especially true if States were to establish one, single, stand-alone AM with appellate jurisdiction over both ICSID and non-ICSID awards.¹² The following discussion assumes (i) the creation of such a stand-alone AM (as opposed to multiple treaty-specific AMs), and (ii), as far as ICSID awards are concerned, the permissibility of an inter se modification of the ICSID Convention pursuant to Article 41 of the VCLT, on which these authors have taken an affirmative view.¹³

In designing an AM for investor-State arbitral awards, one threshold question requiring consideration is whether the new AM should be subject to a national lex arbitri (like non-ICSID investment arbitrations) or be de-localized and governed only by international law (like ICSID arbitrations).¹⁴ One possibility would be that the procedural law applicable to the AM proceedings would be the same as the procedural law governing the first-instance proceedings. Thus, to give an example, proceedings in an UNCITRAL investment arbitration seated in Switzerland would be subject to Chapter 12 of the PILA for both the first instance and appeal proceedings¹⁵; whereas in an ICSID arbitration subject to appeal, both the first instance and appeal proceedings would be governed by international law. This “dual track” would entail two types of legal regimes applicable to appeal proceedings, including for awards rendered under the same IIA, if the IIA provides for a choice between ICSID and non-ICSID arbitration. A different possibility to be explored would be to have a completely de-localized AM procedure subject only to international law for all types of appeal awards. The legal consequences of these choices are important because, as seen above in Sect. 3.3, national courts potentially play a different role in an arbitration governed by a national lex arbitri as opposed to an “a-national” arbitration.

The treaty establishing the AM (the AM Statute) should thus regulate these matters to avoid uncertainties, in particular as far as annulment and enforcement are concerned, which are the two areas in which court intervention will be most relevant.

Starting with the relationship between a potential AM and annulment, these authors have taken the view that the prospective AM should substitute rather than be combined with any annulment-type review present under national law or the ICSID Convention.¹⁶ In other words, appeal and annulment remedies appear mutually exclusive. This is in part because grounds for appeal are normally broader than (and thus already include) the usual grounds for annulment.¹⁷ Furthermore, providing for the possibility of annulment of appeal awards would de facto create a three-tier dispute settlement system, which would go against the objective of efficiency in terms of time and costs.¹⁸

With this assumption in mind (i.e. that no annulment remedies will be provided for appeal awards), for non-ICSID arbitrations, the AM Statute should exclude any role of domestic courts for the purposes of annulment of appeal awards, i.e. the AM Statute should provide for a waiver of judicial review in respect of awards rendered by the AM, in order to avoid a duplication of remedies.¹⁹ Because not all domestic laws would necessarily recognize such a waiver as a valid agreement to exclude the right to seek annulment before their courts, Contracting Parties should consider passing legislation to this effect. In that context, it should also be provided that the arbitration (including the appeals phase, should it not be de-localized for all types of proceedings) must be seated in a State that is a party to the AM Statute. Otherwise, in circumstances where the seat is situated in a third State, there is a risk that such State would not recognize the waiver of judicial review as valid. With regard to ICSID awards, the AM Statute should similarly exclude any annulment of ICSID awards under Article 52 of the ICSID Convention.

With regard to enforcement, the question arises as to the effects of adding an AM layer on the enforcement of an award which has been subject to appeal under the AM.²⁰ As a preliminary remark, it is important to note that, as is the case with annulment, any specific enforcement regime set out in the AM Statute will only bind the Contracting Parties to the Statute.²¹

With regard to enforcement in their territories, Contracting Parties may opt for one single enforcement regime for all appeal awards (whether rendered in non-ICSID or ICSID arbitrations). Alternatively, they could put in place distinct enforcement regimes depending on whether the first-instance award is an ICSID or a non-ICSID award. The former could for instance be enforced pursuant to an Article 54-type rule, whereas the latter would be subject to the NYC regime. It should be noted that if States opt for a dual enforcement regime, depending on the nature of the underlying arbitration, there could potentially be two different enforcement regimes applicable to appeal awards rendered under the same IIA, where the IIA provides for an option between ICSID and non-ICSID arbitrations. This, in practice, would mean that the scope of review of domestic courts at enforcement would be broader for certain awards (Article V of the NYC) than for others (ICSID Convention Article 54-type provision).

With regard to enforcement in third States, an award subject to an appeal or the appeal award itself²² would be enforceable under the NYC,²³ because the addition of an appellate layer does not change the nature of the arbitration process.²⁴ This is true, of course, for a non-ICSID award that is subject to appeal. With regard to ICSID awards subject to an appeal, non-parties to the inter se modification would not be bound by the special enforcement regime that were to be established along the lines of Article 54 of the ICSID Convention. Rather, they would be in a situation similar to that of non-ICSID Contracting Parties in respect of an ICSID award. In other words, they would have to enforce the ICSID award in accordance with the NYC.²⁵ This matter can also be viewed from a different angle. Suppose the AM Statute were to provide a waiver of the grounds for refusal of enforcement (which provision would bind both the State Contracting Parties and the investor which accepts to arbitrate under the treaty). What effect would such a waiver have on third States? It would seem that it would be for each (third) State to determine to what extent a waiver of the grounds for refusal of enforcement included in the AM Statute would be valid in their own legal system. That being so, it is doubtful that the waiver would be effective. Even if specific language is used, as is required under some legislation, it would remain that “the grounds for refusal of enforcement in paragraph 2 of Article V of the New York Convention are legally not capable of being waived or contracted out of”.²⁶

In sum, with regard to the role of domestic courts in a reform scenario providing an AM for investor-State arbitral awards, the domestic courts’ role at annulment is susceptible to being significantly curtailed when compared with their role under the existing regime vis-à-vis non-ICSID arbitrations. This is a natural consequence of the addition of a second layer of review which makes the courts’ supervisory role largely unnecessary. In enforcement matters, domestic courts are likely to keep a role in third countries not parties to the AM Statute, as well as in Contracting Parties depending on the enforcement regime set out in the Statute.

4.4 Multilateral Investment Court

If a MIC is created, how is the role of domestic court going to change?²⁷ Once again, based on the categories of jurisdictional coordination reviewed in chapter 3.2 above, States may envisage designing a role for domestic courts in the sequential or alternative modes previously examined. As an example of a regulation of the interplay between domestic courts and a standing investment court, one can look to the constitutive instrument of the Arab Investment Court, which provides for a fork-in-the-road clause.²⁸

In terms of treaty drafting and taking into account that consent to the jurisdiction of the MIC would be contained in a separate instrument (a future IIA or an existing one for which an opt-in has been exercised),²⁹ there seem to be two possibilities.

First, the MIC Statute may simply defer to any jurisdictional requirements contained in the underlying IIA over which it has jurisdiction. Thus, for instance, where an IIA between States A and B provides for an 18-month litigation requirement, such requirement would continue to apply to proceedings to be brought before the MIC. By contrast, an IIA between States C and D without any such requirement would allow a qualifying investor to directly access the MIC without having to first resort to domestic courts. Under this approach, the MIC Statute would simply defer to any jurisdictional or admissibility requirement (including those aimed at regulating the interplay with domestic courts, including exhaustion, litigation, fork-in-the-road, or waiver clauses) contained in the underlying IIA.

An alternative approach would be to include within the MIC additional jurisdictional or admissibility requirements to be met in addition to those governing under the IIA. Thus, the MIC Statute could contemplate any of the sequential or alternative methods discussed above, for instance a fork-in-the-road clause, that would apply to any proceedings brought before the Court. There would be nothing unusual in this approach; it is adopted by the ICSID Convention, which provides for autonomous jurisdictional requirements (e.g., definition of “investor”, nationality restrictions, etc.) which must be met in addition to the conditions set out in the relevant IIA. Further, if the MIC Statute wishes to provide greater flexibility to Contracting States (with a view to reaching a wider consensus), it could leave it to the Contracting States to opt into some but not all of these requirements (like Article 26 of the ICSID Convention for exhaustion of local remedies). For instance, the MIC Statute could provide that no exhaustion of local remedies is required before the Court, unless a Contracting State indicates otherwise when acceding or ratifying the treaty.

The role of courts in support and control of the proceedings may considerably change as a result of the transition from arbitration to a standing body, depending on how the MIC is conceived. If, as is likely, the MIC is a self-contained court, governed solely by public international law,³⁰ the role of domestic courts would be much curtailed:

• The standing nature of the Court would render any role for domestic courts in the appointment of adjudicators unnecessary.

• With regard to challenges to adjudicators, which are likely to be much less frequent than for arbitrators, the MIC Statute could confer the power to decide on disqualification to an external authority, for instance the PCA Secretary-General, the ICSID Secretary-General, or the ICJ President, or reserve this function to an internal body of the Court.³¹

• In respect of the faculty to seek provisional remedies from domestic courts which may exist under national laws of procedure, the MIC Statute could provide for a solution akin to the one found in the ICSID Convention context, which excludes any such role (unless otherwise provided by the parties).³²

• The MIC Statute is also likely to provide for its own system of review of first-instance decisions either in the form of an annulment or of an appeal.³³

• A residual role will remain for domestic courts at the enforcement stage, an issue which these authors analyzed in the First CIDS Report.³⁴

4.5 Replacing the Existing System with State-to-State Dispute Settlement

An option that is sometimes discussed is to allow only the home State to enforce the IIA obligations on behalf of their investors through SSDS.³⁵ This reform option would essentially entail a return to the pre-investment arbitration system of diplomatic protection and resemble the situation under the FCN treaties pre-dating modern BITs. The potential drawbacks of this option have been highlighted in Chap. 2.³⁶

What effect would a reform proposal aimed at strengthening SSDS have on the role of domestic courts? As previously seen, under customary international law, where the home State brings a claim for injury to one of its nationals against another State (as opposed to a claim for direct injury to itself), the national must first have exhausted all local remedies.³⁷ In Italy v. Cuba, one of the few inter-State cases under a BIT,³⁸ the tribunal confirmed that the exhaustion rule was a prerequisite for Italy’s diplomatic protection claim under the BIT.³⁹

Thus, a reform option centered around the enhanced prominence of SSDS would entail a greater role for domestic courts. In their treaties, States could, however, waive exhaustion of local remedies as pre-condition to SSDS, or they could conceive of SSDS and domestic courts as alternative fora. The Brazilian model Cooperation and Investment Facilitation Agreements (CIFA), which does not provide for investor-State arbitration, but opts for a framework involving an Ombudsman, a Joint Committee of the Treaty Parties, and SSDS, specifies that if an investor has obtained a domestic court judgment with res judicata effect, resort to SSDS under the treaty is foreclosed; if the domestic court litigation is pending, the investor’s waiver of domestic court proceedings is a pre-condition to the home State’s commencement of SSDS proceedings.⁴⁰

4.6 Two Alternative Models: Preliminary Rulings and Complementarity

Beyond the means of coordination to which IIAs typically resort (exhaustion of local remedies, domestic litigation requirements short of exhaustion, fork-in-the-road, and waiver clauses), it may be instructive to address two coordination modes, or “jurisdiction-regulating” norms, existing in other areas of public international law, namely preliminary rulings and complementarity.⁴¹ Could these be transposed to the investment treaty realm?

A preliminary ruling proceeding is a procedure by which a court refers a decision on a specific issue arising in pending proceedings to another court, normally seeking the interpretation of a legal norm from the other court. The proceedings before the court seeking the ruling are typically suspended pending the determination by the other court, which will usually bind the court requesting it. That court will then incorporate the content of the ruling into its overall resolution of the dispute.

The most well-known example is the preliminary ruling procedure pursuant to Article 267 of the Treaty on the Functioning of the European Union (ex Article 234 of the Treaty establishing the European Community), whereby a court of a Member State of the European Union may, and in certain instances must, request the CJEU to give a ruling on the interpretation of a question of EU law that arises in an action pending before the Member State court and is unsettled.⁴² In the context of the EU, the preliminary ruling procedure was needed to foster the unity of the EU legal order in spite of the decentralized interpretation and application at the national level. It has worked as a powerful tool to ensure the uniform application of EU law and thereby the preservation of the legal unity of the Union.⁴³

The transposition of a preliminary ruling mechanism to investment arbitration (i.e. could an investment tribunal seek a preliminary ruling on an unsettled issue of investment law from a permanent body) has been examined elsewhere.⁴⁴ Here, in line with the aim of this study, the purpose is to inquire whether it could be contemplated that a national court seized of an issue of international investment law could seek a preliminary ruling from one of the four international dispute settlement systems envisaged above (investment arbitration, AM, MIC, SSDS).⁴⁵ Or, put differently, assuming a dispute arising out of an IIA is brought before a national court, could that court refer an unsettled question of interpretation of that IIA to an international dispute settlement body? In accordance with the preliminary ruling logic, the international body from which the ruling is requested would not dispose of the dispute pending before the national forum. It would merely authoritatively determine a discrete issue of international investment law to assist the domestic court in resolving the dispute. At the same time, that practice would work toward the “uniformization” of international investment law.

While in theory conceivable, the transposition of this model in the investment law setting at issue appears difficult, if only because the governing law before the domestic court would not necessarily be international law. Indeed, as was seen above, in certain legal systems, domestic courts cannot apply IIAs directly as a result of constitutional limitations.⁴⁶ Hence, in such a situation, a preliminary ruling mechanism would serve no purpose.

An alternative model for the interplay between domestic and international jurisdiction is the principle of complementarity enshrined in the Rome Statute of the International Criminal Court.⁴⁷ Under this principle, States have the first responsibility and right to prosecute the most serious crimes of international concern. The International Criminal Court may only exercise jurisdiction where the national legal system “is unwilling or unable genuinely to carry out the investigation or prosecution” (Article 17 of the Rome Statute).⁴⁸ Transposed to investment disputes, the international settlement mechanism would exercise a sort of “jurisdiction of last resort”; it would do so in the absence of a credible judicial alternative at the national level, i.e. when the domestic courts are “unwilling or unable” to adjudicate the dispute. Complementarity may be regarded as having the advantage of strengthening the capabilities of domestic courts while at the same time preserving international remedies for cases where justice cannot be rendered at the national level. This said, the International Criminal Court is built upon a complex interaction between the State Parties, the U.N. Security Council, the Prosecutor, and the other organs of the Court. These actors or their equivalents are largely absent in the adjudication of investment disputes. As a result, the implementation of a jurisdiction of last resort does not appear practically feasible and would at best be fraught with uncertainty.

In conclusion, while the preliminary rulings and the complementarity models make for an interesting coordination of domestic and international jurisdictions, their implant within the investment framework appears either inapposite (preliminary rulings) or not easily implementable (complementarity).

4.7 Replacing the Existing System with Domestic Courts

Finally, under the most “radical” reform option advanced by certain stakeholders, domestic courts should become the exclusive forum for the settlement of investment disputes. As discussed in previous parts of this study, a wholesale return to only domestic remedies should be considered with caution in particular for States where the courts’ impartiality and the rule of law might be open to question.

States wishing to pursue this option would need to either (i) amend their IIAs to eliminate investor-State arbitration and SSDS clauses; or (ii) terminate their IIAs entirely. Under the first option, domestic courts may apply IIAs, if so allowed under their legal system. Under the second option, investment disputes would simply be adjudicated by reference to domestic legal standards.

Recent State practice provides some examples of the greater role domestic remedies could play in the adjudication of disputes between States and foreign investors.

South Africa, for instance, which has adopted a policy against investor-State dispute settlement, has “recently reviewed all of its IIAs and terminated most of them”.⁴⁹ The investment protection regime is now established under domestic law, in particular, the Protection of Investment Act No. 22 of 2015.⁵⁰ According to the Act, foreign investors are entitled to the same treatment as that afforded to South African investors in like circumstances, save for certain exceptions.⁵¹ The “domestication” of the investment regime not only applies to substantive protection, it also extends to procedural remedies. The Act does not provide for investor-State arbitration, but only for recourse to domestic courts.⁵² The Government may, however, consent to inter-State arbitration with the investor’s home State on a case-by-case basis, subject to the exhaustion of local remedies.⁵³

Some States have followed a different approach, which has been referred to as “selective judicialisation”.⁵⁴ Under this approach, the treaty excludes from the jurisdiction of the international tribunal certain “sensitive areas” which are reserved for domestic courts. For instance, the 2015 Indian Model BIT excludes from the scope of the treaty any measure by a local government (as well as any law or measure regarding taxation), compulsory licenses granted in relation to intellectual property, government procurement, subsidies or grants and services supplied in the exercise of governmental authority.⁵⁵ As a consequence, disputes arising out of measures of this nature cannot be resolved through the settlement mechanisms foreseen in the treaty and must be brought before national courts.⁵⁶