Authors: Anderson Mōri & Tomotsune
Introduction
In investment treaty arbitration, a tribunal is not bound by prior decisions. But the reality is that prior investor-state awards are frequently cited and relied on by parties in their arguments and by tribunals in their awards. In this way, prior awards frequently operate as ‘precedents’ in investment treaty arbitration, as ‘persuasive authorities’ typically do in common law or judicial precedents in civil law systems. This practice is a central fixture in investor-state arbitration, where it is far more prevalent than in commercial arbitration.
This is in part because many of the underlying treaties provide the same or similar rights to investors – of, for instance, fair and equitable treatment (FET) or full protection and security – and, therefore, there is a narrower and more consistent set of claim bases and issues in investor-state disputes than in commercial matters.
Moreover, these rights arise out of public international law rather than any national law. As there is no code or national court jurisprudence to interpret or define the contours of these rights, it may be the natural inclination of tribunals to treat the body of awards issued by other tribunals reviewing similar issues as a guiding legal corpus. That is, tribunals in investor treaty arbitrations may sometimes be motivated to conform to past precedent as a means of developing norms and predictability in a system that otherwise lacks them. This latter view is somewhat controversial, however, given that the mandate of tribunals is not to improve the system but clearly limited to deciding the specific dispute before them.
Perhaps most importantly, this functional system of precedents in investment treaty arbitration is uniquely possible because awards are made publicly available for the most part, in contrast to commercial arbitration, where confidentiality tends to be the rule.
These factors all contribute to the unique system of precedent in investment treaty arbitration. Yet the central premise of international arbitration must be borne in mind: that each tribunal has a mandate to decide the matter before it and no other, and tribunals are entirely free to disregard the rulings of prior tribunals that have previously ruled on the same legal questions.6 A strict view of this mandate might suggest that, in principle, tribunals should not be influenced by wider systemic considerations.
Nonetheless, the practice of heeding precedents does exist in the investment treaty arbitration system. Like case law that gradually evolves and changes in common law courts despite the doctrine of stare decisis because of how courts adopt, limit or modify the legal principles when applying precedents, the ramification of awards treated as precedents in investment treaty arbitration also depends on the way that subsequent tribunals interpret and apply them.
Therefore, the nature and function of arbitral precedents in investment treaty arbitration (or in arbitration in general) is always a subject of debate. In this chapter, this topic is covered by first elaborating on the lack of a formal system of precedents, then exploring the well-established tendency to cite and rely on arbitral precedents and decisions in recent years reflecting on the role of precedents in investment treaty arbitration.
Lack of a Formal System of Precedents
Article 38(1)(d) of the Statute of the International Court of Justice provides that ‘judicial decisions and the teachings of the most highly qualified publicists of the various nations, as subsidiary means for the determination of rules of law’, which is often taken as support for citing precedents in the context of public international law. However, it would be mistaken to understand this as establishing a principle of stare decisis in public international law, because the same Statute makes clear in Article 59 that ‘[t]he decision of the Court has no binding force except between the parties and in respect of that particular case’.
Similarly, the International Centre for Settlement of Investment Disputes Convention (ICSID) Convention (the ICSID Convention or the Convention), which is a foundational document in the system of investment treaty arbitration, provides no basis for creation of a body of case law – or any substantive law – in investment treaty arbitration. The Convention, by Article 48(3), requires only that ‘[t]he award shall deal with every question submitted to the Tribunal, and shall state the reasons upon which it is based’. There is no further articulation on what those reasons should be. Nor is a failure to apply any legal principle or to follow any particular law a ground for annulment of the award. Article 52(1) of the Convention provides grounds for annulment only if it is shown that:
- the tribunal was not properly constituted;
- the tribunal has manifestly exceeded its powers;
- there was corruption on the part of a member of the tribunal;
- there has been a serious departure from a fundamental rule of procedure; or
- the award has failed to state the reasons on which it is based.
None of these grounds concern the substantive merits of the ‘reasons’ stated by the tribunal.
Notably, Article 53 of the Convention provides that the award ‘shall be binding on the parties’ only, which is often read as implying that ICSID awards have no precedential value in relation to future ICSID awards. Accordingly, the ICSID system does not provide for a formal system of precedents.
Although ICSID is the most prominent regime for investment treaty arbitration, sometimes arbitration occurs outside the ICSID framework. The most notable alternatives are ad hoc arbitration, most often under United Nations Commission on International Trade Law (UNCITRAL) arbitration rules, and institutional arbitration administered by the same institutions as international commercial arbitration. Compared with ICSID arbitration, these alternative regimes are even more fragmented and can be less transparent, because they will depend on the varying operation of individual ad hoc tribunals and arbitration institutions across the world.
In 2014, UNCITRAL developed its Rules on Transparency in Treaty-Based Investor-State Arbitration (the UNCITRAL Transparency Rules). Whether wider adoption of these rules will bring changes to this situation remains to be seen.
Aside from the lack of support from the ICSID Convention itself, the realities of investment treaty arbitration also preclude the establishment of a formal system of stare decisis for several reasons. First, there is no centralised superior tribunal or appellate court in investment treaty arbitration. For example, although the ICSID Secretariat has the power to conduct limited review on the awards in ICSID arbitrations, the grounds of annulment are limited to procedural issues. Second, the arbitrators are not members of a standing body. Like those in private commercial arbitration, they are appointed case by case. Unlike judges, who belong to a judicial institution that may be legally required to follow precedents, arbitrators have no obligation to consider or refer to other arbitral decisions. Third, although the attempts to follow precedents contribute to consistency – and, arguably, legitimacy to an extent – a formal system of stare decisis would necessarily need to address the subtler questions, such as the vertical hierarchy of the precedents, the scope of the precedential effect of a decision, and how and when a subsequent tribunal can override a precedent. These cannot be implemented with certainty, short of a new international treaty. States carefully negotiate their own trade treaties and, although they often incorporate investor protections, including arbitration rights, states may be reluctant to have those treaties subject to a global framework of binding precedent arising from a constellation of independent treaty-based disputes and that may evolve in unexpected ways. Therefore, any framework treaty for stare decisis is likely to suffer from extensive carve-outs and exceptions, and perhaps be unsustainable in the end.
In sum, there is currently no stare decisis doctrine in investment treaty arbitration and it is unlikely that any such formal system will come into place without a major overhaul of the current framework of international investment treaty arbitration.
Actual Reliance on Arbitral Precedents
Despite the lack of a formal system of arbitral precedents, one would be misguided to believe that arbitral precedents do not have a role, or to assume they are unimportant. On the contrary, it is now an accepted reality that the use of arbitral precedents is deeply ingrained in the contemporary practice of investment treaty arbitration.
One important decision that is frequently commented on in this regard, and highlights the role of arbitral precedents, is Burlington Resources Inc v. Ecuador, in which the members of the tribunal were split on the issue. The majority noted:
As stated in the Decision on Jurisdiction, the Tribunal considers that it is not bound by previous decisions. Nevertheless, the majority considers that it must pay due regard to earlier decisions of international courts and tribunals. It believes that, subject to compelling contrary grounds, it has a duty to adopt solutions established in a series of consistent cases. It further believes that, subject to the specifics of a given treaty and of the circumstances of the actual case, it has a duty to seek to contribute to the harmonious development of investment law, and thereby to meet the legitimate expectations of the community of States and investors towards the certainty of the rule of law.
This view is widely shared by arbitrators in the field, but it must be noted that it is not universal. In the same award, one arbitrator dissented, noting that she ‘does not analyze the arbitrator’s role in the same manner, as she considers it her duty to decide each case on its own merits, independently of any apparent jurisprudential trend’.
It is interesting to note that the dissenting arbitrator in Burlington Resources, in a parallel case, Occidental Petroleum Corp v. Republic of Ecuador, criticised the majority’s decision based on a comparison with a precedent ICSID case; at the same time, she noted that her ‘critique of the majority’s position, however, is not based on an error of law or an excess of power, but on a different appreciation of the factual situation, which is at the discretion of the Tribunal’. Therefore, although experienced arbitrators do not all, or always, agree regarding the precise value of precedents in investment treaty arbitrations, in practice, reliance on and consideration of prior arbitral precedents in investment treaty disputes is a well-established practice.
Some commentators have sought to develop a theoretical framework for how norms and rules evolve with precedents in investment treaty arbitration. Emergence of these norms and rules often revolve around what Professor Jean Ho identified as two ‘anchors’, or common themes, in arbitral awards. One anchor consists of certain ‘core standards of treatment’, such as basic concepts in international law for protection law, such as protection of aliens and FET, including basic proportionality and due process concepts, while the other ‘anchor’ is the substantive legal rules on the protection of alien property based on concepts of expropriation.
In addition, arbitrators may be motivated by an interest in conforming to apparent international standards. Compared with the tribunals in private commercial arbitration, tribunals in investment treaty arbitration may face greater public pressure to produce well-reasoned decisions that are well-situated in the world of investment treaty ‘jurisprudence’. The reason for this is that the institution of investment treaty arbitration itself is under constant public scrutiny and criticism as it is sometimes perceived as undermining the sovereign rights of governments to make decisions or of the democratic process. In addition, the arbitrators, unlike judges, are not publicly accountable. Arbitrators, therefore, may feel compelled to present their decisions as consistent with, and supported by, a broader framework of jurisprudence that is known to both states and investors. By doing so, they may be seeking to justify investment treaty arbitration as comporting with the rule of law. To achieve that, decision-making by tribunals needs to take on law-like features, for which consistency is important. Following precedents in a way that leads to a generally applicable principle can be claimed to advance these goals. Therefore, to the extent that precedents are available, it is very difficult for arbitrators to refuse to consider or address them. As a result, requirements to publish awards, or even just the legal reasoning part of awards, such as Article 48(4) of the ICSID Convention or Article 3(1) of the UNCITRAL Transparency Rules, do contribute to the formation of an informal system of precedents, even though they do not make the precedents binding.
Accordingly, despite the apparent lack of a stare decisis doctrine in investment treaty arbitration, the practice of citation to and recognition of precedents is now established in practice and is unlikely to go away. Indeed, it has become natural for tribunals to view prior arbitral awards as having precedential value, and it is now difficult to identify any cases where the tribunal simply refused to even consider precedents. In fact, tribunals frequently devote a large amount of energy towards analysing, adopting or distinguishing multiple precedents submitted by the parties.
Decisions reflecting on the role of precedents
In its 2020 decision in RWE v. Kingdom of Spain, in interpreting the FET standard under Article 10(1) of the Energy Charter Treaty (ECT) prohibiting ‘unreasonable or discriminatory measures’, the tribunal discussed the standard of threshold in terms of establishing ‘unreasonableness’ under this provision, and noted that the threshold ‘is a high one, that is frequently assimilated with arbitrariness’. In support of this view, the tribunal cited EDF (Services) Ltd. v Romania and Electrabel SA v. Hungary. The EDF case was actually interpreting a different treaty, the bilateral investment treaty between the United Kingdom and Romania, but the RWE tribunal was apparently persuaded by the detailed articulation of the standard for unreasonable and discriminatory conduct adopted in EDF. Nevertheless, the RWE tribunal stated that, while ‘the Tribunal considers [the proportionality standard in EDF] to be appropriate to describe what is required to show unreasonableness within Article 10(1) ECT’, it also commented ‘that the consideration of whether an act is unreasonable under Article 10(1) may engage an issue of disproportionality’ based on the Electrabel tribunal’s decision interpreting the same provision of ECT. Therefore, the fact that a precedent interpreted a different treaty did not prevent the tribunal from considering the precedent as not belonging to the same body of precedents, especially if it is properly supplemented by precedents on the more specific application in the particular case. However, it is interesting to observe that the tribunal readily accepted that the concept of ‘unreasonable or discriminatory conduct’ engenders a general body of case law applicable to any treaty, in principle.
Another example that sheds light on how investment treaty arbitration tribunals navigate arbitral precedents can be found in OperaFund Eco-Invest et al. v. Kingdom of Spain. In dealing with jurisdictional objections, with the respondent arguing that the tribunal was without jurisdiction because the dispute was an ‘intra-EU’ dispute, the tribunal specifically invited parties to comment on several arbitral precedents concerning the same issue, by asking them to (1) address the application of ‘margin of appreciation enjoyed by national regulatory agencies when dealing with public policy determinations’ articulated in another ICSID decision and (2) explain, in a chart, ‘the common denominators and main differences of the factual and legal background’ between the parties’ own case (OperaFund) and five arbitral precedents identified during the proceeding. Further, the tribunal invited the parties to submit comments in their post-hearing briefs to another arbitral decision that had been issued in the meantime.
Having considered the parties’ comments, the tribunal stated in its award that ‘there is no need to “re-invent the wheel” and start a new examination of all the details regarding the intra-EU objection’ and that the tribunal ‘agrees with all the recent conclusions of other tribunals to the effect that . . . the intra-EU objection is not justified’. Nevertheless, the tribunal briefly explained the substantive reasons for its decision. The OperaFund tribunal has gone to considerable lengths to expound the precise scope and implication of arbitral precedents.
Another recent elaboration of the role of precedents can be found in the 2021 ICSID decision in BayWa re Renewable Energy GmbH et al. v. Kingdom of Spain, in which the tribunal, faced again with a case under the ECT, stated:
As a general matter, investment tribunals (like other international tribunals) are not bound by a strict doctrine of precedent, but are charged to make their own appreciations based on the evidence and argument presented to them. On the other hand, in practice tribunals regularly cite previous, publicly available awards and pay careful attention to them. In the Tribunal’s view, concordant decisions on the interpretation and application of the ECT are entitled to respect, especially if they rise to the level of a jurisprudence constante. On the other hand, where they diverge, a later tribunal has no choice but to form its own view of the relevant law and its application to the facts. This the Tribunal has done.
The BayWa tribunal’s references to ‘jurisprudence constante’ and ‘concordant decisions on the interpretation and application of the ECT’, along with the OperaFund tribunal’s statement that it would not ‘reinvent the wheel’ in the fact of multiple precedents on the same issue are all very intriguing because they reflect the typical approach of common law courts towards persuasive authorities. For example, in such courts, where all (or an overwhelming majority of) other courts have reached a certain conclusion, or if prior decisions addressed the exact same statutory language or fact patterns, even if those decisions are not binding, the persuasive weight of those authorities would be viewed as particularly high.
It should be noted that even when investment treaty tribunals have expressly refused to follow certain arbitral precedents, that is nonetheless an acknowledgment of the perceived pertinence of precedents in arbitration, because the precedents would require no distinguishing or explanation at all for departure from them if they were per se irrelevant.
Of course, the system of precedents in investment treaty arbitrations is still evolving, and there are always voices sounding caution. As the tribunal in Aaron C Berkowitz et al. v. Republic of Costa Rica stated:
The jurisdictional aspects of this case are heavily fact-specific. Although interpretations of law, notably of CAFTA [Central America Free Trade Agreement] Article 10.1.3 and 10.18.1, are necessary, the Tribunal’s assessment ultimately turns on appreciations of fact. The Tribunal thus cautions any reading of this Award that would give it wider ‘precedential’ effects.
Therefore, the Aaron C Berkowitz tribunal is expressing the view that, even if a subsequent case presents a similar fact pattern, it should not be taken for granted that the jurisdictional determination is the same, because each tribunal’s ‘appreciations’ of the facts may be different.
Further, the ISCID Secretariat, in a decision on an application for annulment, Malaysian Historical Salvors SDN BHD v. The Government of Malaysia, also cautioned against applying precedents mechanically. The issue here involved the definition of ‘investment’, for which the award cites certain criteria outlined in Salini Costruttori SpA and Italstrade SpA v. Kingdom of Morocco, which are ‘widely accepted as a starting point of an ICSID Tribunal’s analysis on this point’. Citing Biwater Gauff (Tanzania) Limited v. United Republic of Tanzania,44 the ISCID committee endorsed the view that:
There is no basis for a rote, or overly strict, application of the five Salini criteria in every case. These criteria are not fixed or mandatory as a matter of law. They do not appear in the ICSID Convention. Further, the Salini Test itself is problematic if, as some tribunals have found, the ‘typical characteristics’ of an investment as identified in that decision are elevated into a fixed and inflexible test, and if transactions are to be presumed excluded from the ICSID Convention unless each of the five criteria are satisfied. This risks the arbitrary exclusion of certain types of transaction from the scope of the Convention. It also leads to a definition that may contradict individual agreements (as here), as well as a developing consensus in parts of the world as to the meaning of ‘investment’ (as expressed, e.g., in bilateral investment treaties). If very substantial numbers of BITs across the world express the definition of ‘investment’ more broadly than the Salini Test, and if this constitutes any type of international consensus, it is difficult to see why the ICSID Convention ought to be read more narrowly. [Therefore,] a more flexible and pragmatic approach to the meaning of ‘investment’ is appropriate, which considers the features identified in Salini, but along with all the circumstances of the case, including the nature of the instrument containing the relevant consent to ICSID.
In sum, it is common for tribunals to take arbitral precedents very seriously and make good, or even extra, efforts to follow precedents, as the tribunal in OperaFund did. Tribunals are particularly likely to do so in cases like OperaFund, where prior decisions decide the exact, purely legal questions. One argument in favour of such consistency is that failure to follow those precedents would categorically deprive aggrieved parties of the forum of investment treaty arbitration in its entirety, where prior tribunals allowed it, resulting in substantial injustice. On the other hand, as the Aaron C Berkowitz tribunal and the committee in Malaysia Historical Salvors has suggested, application of precedents should always be done with caution.
Conclusions Although there is no formal system of stare decisis in investment treaty arbitration, nor is there likely to be any in the foreseeable future, arbitral precedents are nonetheless central to investment treaty arbitration today and the treatment and interpretation of precedential awards continue to evolve. Even though it is possible that the precise attitude or approach to utilising precedents will differ depending on the specific tribunal, it is very unlikely that a tribunal will refuse to consider precedents entirely. On the contrary, because investment treaty arbitrators are often experienced practitioners or highly accomplished scholars, their professional life and reputation depend on their ability to provide well-reasoned awards, which often turn on their ability to distil and apply workable legal principles while weeding out the chaff from the precedents. Political or public pressure may also add to the incentives for tribunals to justify decisions that are sometimes seen by critics as working against the democratically elected administrators and legislators or decisions of sovereign states. Further, greater uniformity in treaty drafting, increased access, transparency and ease to conduct research on arbitral precedents have also contributed to the continued expansion and development of precedents in investment agreement arbitration.
