A New Test Toward Consistency in International Investment Arbitration

Abstract

The issue of inconsistency in international investment arbitration has become more evident with the proliferation of investment disputes. This article analyzes the value as well as the desired extent of consistency and coherence for investment arbitration, and further proposes a new test that would enhance consistency and foster the development of investment law. By matching the current status of investment arbitration with three identified conditions for the application of stare decisis or jurisprudence constante, this article concludes that strictly applying the doctrines to achieve consistency is neither feasible nor favorable for the investment arbitration regime. Further, while the public nature of this dispute-resolution mechanism requires consistency and coherence, existing attempts and approaches in this regard have significant limitations. As such, this article proposes a soft precedents test to reach a balance between consistency and flexibility. The test requires the tribunals to consider previous cases in good faith when facing the same or similar issues discussed therein, and the core consideration in whether to agree with previous cases is to develop international investment law. It is suggested that this proposed test would help to clarify, enrich and develop the content of the law within the investment arbitration community.

I. INTRODUCTION

With the recent rapid development of international investment law and the proliferation of investment arbitrations, inconsistency among the tribunals’ rulings become more and more of a focus for practitioners and scholars.

Recent decades witnessed a rapid rise in international investment arbitrations. There has been a surge in the number of bilateral and multilateral investment treaties (BITs and MITs). Today, about 2500 such treaties are in existence,¹ with their respective dispute resolution clause linked to various institutions and mechanisms, notably the International Centre for Settlement of Investment Disputes (ICSID) that settles investor-State disputes. Tribunals are composed on a case-by-case basis according to parties’ autonomy, and the arbitrators may well be—and often are—different in every case. It is therefore not a surprise that they sometimes hold different positions on the same issues, even in similar situations. Domestic laws often achieve consistency by implementing stare decisis or jurisprudence constante doctrines. However, unlike the domestic legal systems, the dispute resolution system for international investment law lacks a hierarchy or a uniform adjudicatory body for appeal. Therefore, inconsistent final awards or orders are likely to remain as they are, confusing the parties and frustrating the purposes of the international investment regime.

This problem is especially obvious in proceedings where the same or related disputes are submitted concurrently to more than one forum (tribunal or court) for resolution.² This is exemplified in a series of cases against Argentina involving the US–Argentina BIT: CMS v Argentina (‘CMS’), LG&E v Argentina (‘LG&E’), Enron v Argentina (‘Enron’) and Sempra v Argentina (‘Sempra’).³ In those cases, the tribunals held differently on the ‘necessity’ defense raised by Argentina for the same circumstance, ie the Argentina economic crisis, which further led to different outcomes in the final awards. The Tribunals in CMS and Enron both held that Argentina had alternative measures that would have been less detrimental to investors to address the crisis, and thus its measures did not satisfy the requirements of customary international law set forth in Article 25 of the Draft Articles on Responsibility of States for Internationally Wrongful Acts (ARSIWA). In contrast, the Tribunal in LG&E opined that Argentina was in a state of emergency and thus should not be held accountable for the loss caused by its measures to investors during this period. In the subsequent Sempra case, the Tribunal noted that CMS, Enron and LG&E all revolved around the Argentine economic crisis and some particular issues. Nevertheless, while the Sempra Tribunal concluded that the conclusions of previous cases varied mainly due to different factual backgrounds, it did not delve into these differences. The Tribunal merely determined that Argentina’s economic crisis did not jeopardize the nation’s existence or independence and Argentina couldn’t fulfill all the conditions of Article 25 of ARSIWA, and thus rejected its defense.⁴

These cases reflected an obvious lack of consistency since they reached different conclusions with regard to similar facts and laws. In Sempra, although the Tribunal briefly discussed previous cases, it did not provide a detailed explanation for its deviation from them. Such inconsistency among different tribunals will reduce the foreseeability of international investment law, which in turn frustrates the purpose of international investment arbitration.

This article will look into the issue of consistency in the context of international investment arbitration and propose a new legal test concerning prior cases suitable for the current status of investor-State dispute settlement (ISDS). The purpose of the test is to enhance consistency within the investment arbitration regime while leaving space for its inherent requirement for a certain degree of flexibility.

Section II first addresses the significance of consistency for a legal system and introduces stare decisis and jurisprudence constante doctrines which have been widely adopted to achieve consistency. It then analyzes the applicability of the doctrines in international investment arbitration on the basis of the three conditions identified for the doctrines to work. Section III suggests that there is a need for a feasible approach to enhance consistency in international investment law, as a certain degree of consistency is desirable in international investment arbitration while existing proposals to achieve consistency are not sufficient to achieve this goal. In light of this, Section IV proposes a new soft test of precedents in the specific context of investment arbitration, which requires the arbitrators to consider prior cases in good faith. The article further envisages several application details of the test and analyzes why this test is most feasible and advantageous to the development of international investment law.

II. CONSISTENCY AND DOCTRINES CONCERNING PRIOR CASES

Consistency means ‘a logical coherence among things or a uniformity of successive results’.⁵ In legal systems, consistency is considered one of the most important attributes of law which contributes to, if not endows, the legitimacy of the laws. It is said that ‘[l]egal theory tells us that the rule of law is only the rule of law if it is consistently applied so as to be predictable’.⁶

A. Stare Decisis and Jurisprudence Constante

In the experience of most legal systems, consistency of law is mainly achieved by following prior rulings. As pointed out, adherence to legal decisions ‘is imperative if the law is to fulfil one of its primary functions, ie the maintenance of security and stability’.⁷

Common law jurisdictions achieve consistency mainly by stare decisis. The US Supreme Court described the rationale of the doctrine as ‘promot[ing] the evenhanded, predictable, and consistent development of legal principles, foster[ing] reliance on judicial decisions, and contribut[ing] to the actual and perceived integrity of the judicial process’.⁸ In addition, in many common law jurisdictions, the party submissions are also presumptively public.⁹ This makes it easier for everyone to have notice of the cases that are likely to establish a precedent, and to weigh in on those cases as appropriate.

For civil law jurisdictions, consistency is primarily achieved by codified legal text that is public to all citizens. However, cases also play an important role under the doctrine of jurisprudence constante.¹⁰ This doctrine is described as a weakened formulation of stare decisis: the longer a series of identical decisions in identical cases based on the same legal rule by the courts becomes, the more determinative the decision found in the series becomes for future cases.¹¹ In China, for example, the Supreme People’s Court will publish series of Guiding Cases from time to time in order to summarize judicial experiences and ensure the consistent application of law.¹²

When it comes to international law, the extent of consistency achieved by following prior rulings varies in different detailed areas.¹³

One extreme is international commercial arbitration, where arbitrators seldom refer to previous awards. It has been noted that when the United Nations Convention on Contracts for the International Sale of Goods (CISG) applies, although it is a uniform set of rules and there are databases of CISG cases, the arbitrators still refrain from making references to prior cases.¹⁴ With a closer look, one may find that when there are citations to jurisprudence, they mostly concern jurisdiction and procedure; and when deciding substantive issues, arbitrators in international commercial arbitration rarely consult prior awards.¹⁵ A possible explanation might be that the arbitrators in those cases are often dealing with a one-off contract, as opposed to the same set of substantive issues that comes up in different cases.

Another extreme is public international law adjudicated by the International Court of Justice (ICJ or the Court) and the law of the World Trade Organization (WTO).

Article 59 of the ICJ Statute provides that ‘[t]he decision of the Court has no binding force except between the parties and in respect of that particular case’.¹⁶ Therefore, there are no binding precedents in the ICJ. However, the ICJ tribunals maintain a strong reliance on jurisprudence, which is said to reflect jurisprudence constante to a large extent.¹⁷ Article 38 of the ICJ Statute stipulates that judicial decisions are one of the ‘subsidiary means for the determination of rules of law’.¹⁸ In practice, past decisions are frequently cited. As the Court stated in Land and Maritime Boundary, ‘[I]t is not a question of holding [the parties in the instant case] to decisions reached by the court in previous cases. The real question is whether in this case, there is cause not to follow the reasoning and conclusions of earlier cases.’¹⁹

Similar for the WTO regime, while there is no express doctrine of binding precedent in the Understanding on Rules and Procedures Governing the Settlement of Disputes (‘DSU’), the Appellate Body in US—Stainless Steel (Mexico) held that ‘[e]nsuring “security and predictability” in the dispute settlement system, as contemplated in Article 3.2 of the DSU, implies that, absent cogent reasons, an adjudicatory body will resolve the same legal question in the same way in a subsequent case’.²⁰ Thus, it seems to be a de facto jurisprudence constante within the WTO regime.²¹

In conclusion, while following prior rulings is crucial to the consistency of a legal regime, the application of relevant doctrines in international law regimes differs in each field due to its respective features. In some fields, the doctrines may not work as well as in other fields due to a lack of relevant conditions.

B. Lack of Basis for Adoption of Stare Decisis or Jurisprudence Constante in International Investment Arbitration

In terms of the international investment law regime, unlike the ICJ or WTO regimes, there is a lack of basis for strict adherence to stare decisis or jurisprudence constante. This article will discuss the conditions for the application of the doctrines and whether they are or will be met in international investment arbitration.

This article identifies three main conditions necessary for the doctrines to work: (i) uniformity and hierarchy of adjudicating bodies; (ii) mature stage of legal development; and (iii) transparency and accessibility.

(i) Uniformity and hierarchy of adjudicating bodies

The first crucial condition for both doctrines to work is the presence of unified and hierarchical adjudicating bodies within a legal system. In domestic systems that have binding precedents, the government has established the system and oversees it, as the judiciary is a part of the government. This provides an institutional safeguard for the application of the doctrines. When the adjudicating bodies are consistent and there is a strict hierarchy among them, it is straightforward for the adjudicators to ascertain which cases to follow, especially when there are conflicts among them. Otherwise, different adjudicating bodies might come up with contradictory decisions, and the binding or reference effect of one court’s decisions on other courts will be dubious.

The history of judicial reform in England sets a good example for demonstrating how important a consistent and well-structured court system is. Before the enactment of the Judicature Acts in 1873, the English court structure was complex and disorganized. There were two separate court systems acting in parallel, namely the common law courts and the courts of equity. The jurisdictions of the two court systems were unclear and overlapped, and there was no unified central judicial authority which could definitely say what the law was.²² In the nineteenth century, the Judicature Acts were passed to completely reorganize the court structure by merging the two systems into a unified system of courts with a formal hierarchical structure.²³ It is said that this reform made possible ‘for the first time a coherent application of the strict doctrine of stare decisis’.²⁴ A strict stare decisis—where a decision on a particular point of law must be followed by courts at the same level and lower within a judicial hierarchy—came about only gradually after this reform.²⁵

Similarly, for legal regimes where the doctrine of jurisprudence constante is deemed to work,²⁶ this first condition is also met. Take public international law as an example: the ICJ judges are a fixed group of people who sit for an extended period of time. As a result, the authority of the ICJ jurisprudence would not be difficult to build considering the monopoly status of the court in relevant fields and the consistent pool of judges. In the same vein, the WTO dispute resolution system features a stable and authoritative Appellate Body. Although the panels are different in each case, questions concerning the application of the WTO law, if doubted by the parties, will eventually come to the standing Appellate Body. This hierarchy structure and the permanent nature of the adjudicating body allow for the possibility of a consistent interpretation and application of the laws.

Nevertheless, this condition is not and might never be satisfied in the context of investment law. In investment arbitration cases, parties are free to nominate the arbitrators under the principle of party autonomy. Given that different cases will involve different industry sectors and parties from different countries, parties in a particular case will generally select arbitrators by taking into account the circumstances of the case. Accordingly, tribunal members will not and must not be limited to only a few individuals, and the mainstream view now is to promote diversity in the composition of the arbitration panels. Therefore, the first condition is naturally not met, and it is neither practicable nor desirable to establish a unified adjudicating body to hear all the investment arbitration cases.²⁷

Moreover, investment tribunals have no hierarchy or ranking of seniority. There is neither an appeal nor a standing review body for investment arbitration cases. Although the awards of ICSID tribunals might be annulled, the ICSID annulment committees are ad hoc instead of standing in nature, and the narrow grounds of annulment limit their remit to procedural issues, precluding a review of any substantive errors of law. As is illustrated in the Argentina cases, opinions of one tribunal may largely differ from another. Such absence of hierarchy among tribunals naturally hinders adherence to previous cases—you simply do not know whose opinions to follow in the face of divergence. Although there are reform proposals such as establishing a standing body of review, such reform may not be feasible and is contrary to the basic idea of party autonomy. Section III.B. will analyze this issue in further detail.

In a nutshell, the impossibility of meeting the first condition is one of the fundamental reasons why stare decisis or jurisprudence constante cannot be rigidly applied in the field of investment arbitration.

(ii) Mature stage of legal development

The second condition for the doctrines to work is that a legal system has reached the mature stage, the failure of which becomes another fundamental reason that renders them inviable in international investment arbitration.

The key feature of a mature legal system is that a comprehensive body of legal rules has been developed over time. The importance of this condition can be evidenced by the English legal system where the development of stare decisis has experienced a long process, and it was not until the nineteenth century that the doctrine became entrenched as a matter of judicial practice.²⁸ At the early stage of the English common law system, the absence of statutes and limited precedents left judges with little guidance for resolving diverse disputes. As a result, judges played a significant role in shaping the law itself. Gradually, judges have, through the issuance of reasoned decisions, filled the gaps in laws. As the legal system matured, comprehensive legal rules were distilled from a vast body of case law, and the primary role of judges when rendering judgments became interpretation of case laws or statutes, rather than reinventing the wheel.

As for jurisprudence constante which can be found in the WTO dispute settlement mechanism, it is evident that the WTO agreements formed a relatively well-developed legal regime covering the vast majority of detailed issues in international trade. As demonstrated by the WTO Analytical Index, the meaning of the applicable law concerning one issue is enriched by the discussion in relevant cases of the same set of legal texts and the reasoning about such texts in prior cases.

However, this is not the case in the investment law regime. International investment law is of relatively recent origin and is still in flux. Also, the investment arbitration regime is based on a dynamic treaty law system where bilateral and multilateral investment treaties have gone through an evolutionary process and are still under reform. In this light, there is no basis to have a stable series of cases to be followed. To rigidly adhere to prior rulings will also impede this developing legal regime from self-correction and enrichment. Instead, to fill in the gaps between the laws, adjudicators are encouraged to the take the initiative to cite and discuss previous cases in their explanation of the law. It has been pointed out that ‘[w]hen arbitrators apply a body of rules that is less developed and is still in the process of being formed, their role with respect to the establishment of predictable rules is much more important’.²⁹ Therefore, the current development status of international investment law fails to meet the second condition for adopting stare decisis or jurisprudence constante.

(iii) Transparency and accessibility

The third condition for stare decisis and jurisprudence constante to work effectively is the availability of transparent and searchable prior rulings, as well as other documents.

The ability to access previous decisions is essential for adjudicators, lawyers and academics. This accessibility enables adjudicators to make informed and consistent rulings by considering previous rulings for the same issues. It also ensures that legal practitioners can research and cite relevant cases when building their arguments and anticipating the potential outcomes. Moreover, academics would be able to conduct in-depth research and analysis on past decisions and contribute to the evolution and development of the laws.

To better effectuate adherence to the doctrines, domestic legal systems usually have a case reporting system, which records precisely what exact words were said in earlier cases. In England, the Incorporated Council of Law Reporting was founded in 1865 and started to publish legal precedents in the Law Reports.³⁰ A standardized case law reporting system ensures that prior rulings are well documented and therefore paves the way for the development of precedents or jurisprudence.

In addition to publishing previous decisions, the ability to access documents filed in the proceedings, especially the submissions, is also important. As mentioned in Section II.A., many domestic legal systems make the documents filed in proceedings public. The WTO dispute resolution mechanism also makes public the parties’ submissions or the summaries of submissions and allows third parties to participate.³¹ This enables non-parties to weigh in, eg by submitting amici curiae briefs. Otherwise, it might not be fair to have a case binding upon non-parties if they initially have no opportunity to participate.

In contrast, commercial arbitral awards and submissions are often confidential and not made publicly available. Due to the relatively strict confidentiality of commercial arbitrations, the tribunals and parties only have limited recourse to jurisprudence in international commercial arbitrations. This lack of transparency hinders the development of precedents in the realm of commercial arbitration.

In the sphere of investment arbitration, there has been an increasing number of decisions and relevant documents made available to the public. As an effort to promote transparency, the 2006 version of the ICSID Arbitration Rules required the ICSID Centre to publish promptly excerpts of tribunals’ legal reasoning.³² Further, Rule 62 of the amended ICSID Arbitration Rules (2022) provides that the ICSID Centre must publish every award with consent of the parties and there must be deemed consent if no written objection is made within 60 days after the dispatch of an award.³³ If such objection is made, the ICSID Secretariat will at least publish the excepts of the awards.³⁴

Through such disclosure of legal reasoning, ICSID tribunals are able to take account of previous rulings made by others to foster greater consistency in investment arbitration cases. As the tribunal in El Paso v Argentina commented:

It is nonetheless a reasonable assumption that international arbitral tribunals, notably those established within the ICSID system, will generally take account of the precedents established by other arbitration organs, especially those set by other international tribunals.³⁵

By comparison, the transparency requirement under the UNICITRAL Rules is even higher, where consent is given by virtue of the selection of the UNCITRAL Rules. In 2013, the UNCITRAL Rules on Transparency in Treaty-based Investor State Arbitration (‘Transparency Rules’) were introduced to increase transparency in investor-State arbitrations by requiring the publication of parties’ submissions, witness statements, expert reports, transcripts and tribunal decisions.³⁶ In 2015, the United Nations Convention on Transparency in Treaty-based Investor-State Arbitration (‘Transparency Convention’) further extends the application scope of the Transparency Rules to investor-State arbitration cases arising under the treaties concluded prior to the Transparency Rules.

As for documents filed in the proceedings, pursuant to Rule 64 of the amended ICSID Arbitration Rules (2022), the parties may agree to publish any written submission that they filed in the proceeding, including pleadings, witness statements, expert reports, legal authorities and exhibits.³⁷ Even if the parties do not agree, one party can make public its own submissions. The other party can request redactions on the submissions, and they may have recourse to the tribunal if disagreements arise.³⁸

The above rules and practices illustrate a continuing effort to promote transparency as well as a fairly satisfactory status of transparency already achieved in the field of investment arbitration. In this light, the third condition for applying the stare decisis or jurisprudence constante could be met, and the available previous rulings and documents should be valuable sources for the development of investment arbitration.

Nevertheless, considering the situation of the first two conditions, this article finds that strictly employing stare decisis or jurisprudence constante to achieve consistency in investment arbitration is not a feasible choice.

III. THE NEED FOR A FEASIBLE APPROACH TO ENHANCE CONSISTENCY AND COHERENCE IN INTERNATIONAL INVESTMENT LAW

While it has been argued that strict stare decisis or jurisprudence constante are not suitable for investment arbitration, an effort toward consistency is still desirable and even necessary in view of the public nature of international investment law. Considering that there have been a wide range of concerns over the lack of consistency and the deficiencies in existing attempts to tackle this problem, a more feasible approach to enhance consistency and coherence in international investment arbitration is still required.

A. The Necessity of Consistency in International Investment Arbitrations and its Extent

In 2007, the ICSID Tribunal in Saipem stated that:

The Tribunal considers that it is not bound by previous decisions. At the same time, it is of the opinion that it must pay due consideration to earlier decisions of international tribunals. It believes that, subject to compelling contrary grounds, it has a duty to adopt solutions established in a series of consistent cases. It also 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 certainty of the rule of law’.³⁹

Similar to this tribunal, many other ICSID tribunals expressly take into account previous awards when deciding the same issues regarding similar facts. In the Argentina cases, the Sempra Tribunal addressed the holdings in the previous CMS, Enron and LG&E cases. Although the Tribunal only slightly touched upon the reasons for the deficiency among the cases without an in-depth analysis, the fact that prior rulings were mentioned demonstrates that the Tribunal felt the need for consistency in international investment arbitrations, especially in proceedings emerging out of the same measure and circumstances.

Compared to international commercial arbitration, international investment arbitration has and should have a stronger requirement of consistency. The main reason lies in the fact that the former has more of a public nature. Unlike international commercial arbitration, which is between equal parties, international investment arbitration is commonly between investor and State: the laws applicable to the substance of the dispute and the agreement to arbitrate are both public international law,⁴⁰ and the subject matter often involves ‘questions about the scope and limits of the host State’s regulatory powers’⁴¹ rather than contractual rights and obligations in commercial cases.

More importantly, the public nature of ISDS may be best illustrated by the potential broad impact of specific cases. First, an investment arbitration might have an influence on all the investors from relevant States, as the holdings concerning whether a State measure is protected under relevant provisions of a treaty will affect the activeness and strategy for existing or future investors to exert their funds. Second, an investment arbitration may also make the host State willing to review or reregulate its standards of behavior on certain measures in the long term. In this sense, it has been pointed out that ‘investment arbitration shares the public nature of international law in creating global interests and global goods relevant for the international community as a whole’.⁴² Third, an investment arbitration award might further have a direct or indirect impact on the population of the host State, as the State may be held responsible for a substantial amount of compensation to the investor and/or consequently decide to change its policies. As such, ISDS not only operates in the sphere of public international law, but also involves public interest.

Therefore, inconsistency in investment arbitration has wider real-world consequences. Lack of coherence impairs the stability and predictability of the legal system, resulting in a reduction of credibility and popularity of the dispute resolution regime and the law itself. On this, some contend that decision-makers have a moral, or even legal, obligation to strive for consistency and predictability and thus to follow precedents.⁴³

Notably, in recent years, the lack of consistency and coherence of ISDS have already drawn a wide range of concerns and triggered some serious discussions. Since 2017, UNCITRAL entrusted Working Group III to work on the possible reform of ISDS, and it was reported that the ‘criticism of a lack of consistency and coherence was one of the reasons behind Commission’s decision to embark on work on possible ISDS reform’.⁴⁴

While discharging this mandate, Working Group III received several submissions from governments of State members which set out specific concerns over the adverse effect of inconsistency. The primary issue raised, which was also emphasized by the Secretariat of Working Group III, was that inconsistent awards can negatively affect the reliability and predictability, and eventually the credibility of the investment arbitration regime.⁴⁵ Another major criticism was that the cost and duration of ISDS could be increased in the absence of a coherent and predictable framework, as the parties would be encouraged to relitigate recurrent issues because ‘there is no guarantee that one ad hoc tribunal will follow an interpretation, however well-reasoned, of another ad hoc tribunal’.⁴⁶ In addition, there was the complaint that there are many broadly and loosely defined terms in ISDS such as ‘legitimate expectations of the investor’, and their meanings may ‘differ from one arbitral tribunal to another’.⁴⁷ UNCITRAL Working Group III is still working to investigate and find possible solution for the current status of inconsistency in ISDS.

On the other hand, Working Group III also noted that the lack of consistency might be a ‘logical result’ of the ISDS regime, mostly due to its fragmented nature. In this regard, the desirable extent of consistency of coherence was also considered, and it was pointed out that ‘predictability and correctness should be the objective rather than uniformity’.⁴⁸

In a word, international investment arbitration neither has a structural basis to completely adopt stare decisis or jurisprudence constante as in domestic courts or WTO, nor can it entirely abandon consistency. A balanced approach has to be found between the two extremes.

B. Problems in Existing Attempts to Achieve Consistency and Coherence

To achieve a balance between flexibility and consistency, several reform attempts have been proposed.

The most discussed solutions concern a standing adjudicatory body, be it a court of appeal or uniform courts. For example, it is proposed that an investment court system should be established pertaining to single investment treaties as exemplified in the Comprehensive Economic and Trade Agreement (CETA) between Canada and the European Union; according to this proposal, the ad hoc tribunals would be replaced by a fixed tribunal and an appellate tribunal previously appointed by the CETA Joint Committee.⁴⁹ Another proposal is to set up a multilateral investment court involving the biggest possible number of countries.⁵⁰ Similarly, there is a proposal for an appellate body to review awards issued by tribunals on the errors of law.⁵¹

These proposals aim at setting a structural foundation in international investment arbitration for the application of stare decisis, either by setting up a hierarchy or by concentrating the adjudicatory power into the hands of a permanent institution with a relatively stable composition. Such proposals would definitely bring international investment arbitrations closer to consistency. But probably too close.

From a theoretical point of view, the spirit of arbitration lies in the principle of party autonomy. Setting up permanent courts, be it an appeal court or not, will bring political concerns into the regime of arbitration. States might be able to influence the composition of the court, breaking the balance of the parties in investor-State disputes, which actually frustrates, instead of facilitates, the purpose of international investment law. From a practical point of view, it would be very difficult to seek States’ consensus on setting up a uniform court.⁵² In this regard, the WTO Appellate Body serve as a good example. While a hierarchy and standing body has already been set up and operated for decades, it can still be paralyzed by WTO member States when they are disagree.

Therefore, setting up permanent bodies to adjudicate or review investment disputes does not seem like a feasible way to achieve consistency in international investment disputes, at least at its current stage of development.

In this connection, UNCITRAL Working Group III is envisaging an appellate mechanism, whether standing or ad hoc, to enhance the consistency of decisions.⁵³ Nevertheless, during the recent 44th session held in January 2023, Working Group III identified multiple concerns over this proposal, including the additional costs and time the parties might incur, financing of the mechanism and the risk of further fragmentation. In the view of Working Group III, ‘more general discussions on the desirability, the purpose, and the possible structure of an appellate mechanism’ are still needed.⁵⁴ Further to the discussions of Working Group III, this article suggests that an appellate mechanism may not be a desirable approach to tackle the problem of inconsistency for three reasons. First, it is doubtful whether the mere existence of a potential appellate tribunal would urge the tribunal of the first instance to consider similar cases, as the appeal itself is still more of a case-wise mechanism focusing mainly on the fact and law of a specific case. Second, it has been pointed out that, since investment disputes are based on different BITs with different wordings, an appellate body may achieve coherence only to a moderate extent.⁵⁵ Third, as investment arbitration has already been criticized for the significance of duration and costs, a potential appeal proceeding may only increase investors’ reluctance to resort to this dispute resolution mechanism.

There are other attempts aiming at applying stare decisis or jurisprudence constante in international investment arbitrations. For example, to incorporate the doctrines into current arbitration rules.⁵⁶ While this proposal seems to incorporate precedents into the existing regime, avoiding any unfeasible systematic reform, questions remain to be answered on many practical aspects. For example: since there are contradictory awards in the current system, which line of cases should be counted as precedents, and more importantly, based on which reason can a tribunal depart from previous (contradictory) cases. Without a concrete answer to such questions, it is hard to say whether such a proposal would be feasible as a change in arbitration rules which would be welcomed by investors.

IV. A NEW SOFT PRECEDENTS TEST IN INTERNATIONAL INVESTMENT LAW

Considering that a higher level of consistency and coherence is desirable in international investment arbitration, and this mechanism itself is still developing and not yet completely mature, this article proposes a soft precedents test to enhance consistency within the international investment arbitration regime while preserving adequate room for a degree of flexibility, which, to a large extent, reflects and formalizes current practice.

The test requires that, when facing the same or similar issues discussed in previous cases, the tribunals should consider relevant rulings and reasonings therein when making their own rulings. While tribunals determine whether to follow previous cases, the guiding principle is to develop international investment law. Due to the fragmented and non-hierarchy features of international investment arbitrations, this loose obligation to consider does not require strict adherence to precedents. It only requires the tribunal to discuss whether to agree with previous cases in good faith. The scope of previous cases is confined to those brought up by the parties, while the tribunal can invite the parties to consider and comment on certain prior rulings on its own initiative. Further, it is proposed that such a test could be incorporated into the setting aside and annulment mechanism of the awards so as to incentivize tribunals to comply with the test.

A. A Loose Obligation to Consider

One distinct feature of the new soft test is that it only imposes on the tribunals a loose obligation to take into account previous rulings, ie to ‘consider’ them and decide whether to agree with them in the consideration of developing international investment law. The new soft test differs from stare decisis or jurisprudence constante by not imposing an obligation on tribunals to adhere to previous cases unless there are compelling grounds for not doing so, but the tribunal must explain why their decision to agree or disagree with prior cases develops international investment law.

The idea of taking previous cases into account comes from stare decisis and jurisprudence constante in domestic laws as mentioned. However, the obligation in the new soft test is much looser. Normally, the doctrines require courts to follow the reasonings in jurisprudence and set relatively high standards for deviating from previous cases. Take the doctrine of stare decisis in the United States as an example. The Supreme Court held that ‘when governing decisions are unworkable or are badly reasoned, “this Court has never felt constrained to follow precedent”’.⁵⁷ Such a threshold is high. In Brown v Board of Education, the Court gave a prolonged analysis of public education in the light of its full development after the precedent was made, before reversing the ‘separate but equal’ doctrine in Plessy v Ferguson.⁵⁸

In the same vein, WTO tribunals are also required to apply a high bar against overturning previous cases. The Appellate Body in US—Stainless Steel (Mexico) specified that ‘absent cogent reasons, an adjudicatory body will resolve the same legal question in the same way in a subsequent case’.⁵⁹ The Panel in China—Rare Earths further considered that ‘the expression “cogent reasons” may be understood as referring generally to a high threshold’.⁶⁰ When applying the test, the WTO panels took a very prudent stance. In China—Rare Earths, China cited US—Carbon Steel to argue that even though paragraph 11.3 of China’s Accession Protocol does not mention Article XX of the General Agreement on Tariffs and Trade 1994, it does not necessarily mean that the latter cannot be treated as an integral part of the former. This argument is inconsistent with the Appellate Body Report in China—Raw Materials, so the Panel carefully examined whether China’s reasonings and the citing of US—Carbon Steel constituted a cogent reason to depart from China—Raw Materials. After a detailed comparison of the similarities between US—Carbon Steel and China—Raw Materials concerning the problem they discussed, the method of reasoning and the conclusions they reached, the Panel concluded that there was no incompatibility, but actually striking similarities between the analysis in the two cases, and that China’s reliance on US—Carbon Steel was misplaced.⁶¹

In contrast, the soft test does not impose an obligation to adhere to prior rulings. It only entails an obligation to consider which should be construed in combination with the principle of good faith. While the principle of good faith is fundamental to more or less every legal system,⁶² its application in the WTO tribunals concerning Article XXI of the GATT 1994⁶³ can be a good reference for its meaning in the soft precedents test. Article XXI provides that: ‘Nothing in this Agreement shall be construed to prevent any contracting party from taking any action which it considers necessary for the protection of its essential security interests’ (emphasis added). The WTO panels held that such language requires—and solely requires—the parties to comply with the principle of good faith. The Panel considered that:

[Such principle] applies not only to the Member’s definition of the essential security interests, but also to their connection with the measures at issue, demanding that the measures at issue meet a minimum requirement of plausibility in relation to the proffered essential security interests, i.e. that they are not implausible as measures protective of these interests.⁶⁴

Similarly in the soft test, tribunals only have to meet a minimum requirement of plausibility on how their positions regarding previous cases develop international investment law. The plausibility is to be demonstrated in both their understanding of the development of international investment law and the connection between their rulings and such development. In other words, if there are contradictions in prior cases, the tribunal may choose to follow one line of reasoning and depart from the other, or find a middle road or create a new line of reasoning. When taking a position, the tribunal should explain its reasons in respect of why such a decision develops international investment law. Actually, parties to the CETA agree that tribunals should take into account some proceedings to solve the problem of coherence. Article 8.24 of CETA provides that:

Where a claim is brought pursuant to this Section and another international agreement and: (a) there is a potential for overlapping compensation; or (b) the other international claim could have a significant impact on the resolution of the claim brought pursuant to this Section, the Tribunal shall, as soon as possible after hearing the disputing parties, stay its proceedings or otherwise ensure that proceedings brought pursuant to another international agreement are taken into account in its decision, order or award.⁶⁵

The soft test is suitable for investment arbitrations because it caters to two main features of international investment law: (i) there is no and likely will not be a structural basis for a strict application of the stare decisis or jurisprudence constante; (ii) international investment law has not yet reached the stage of development that is mature enough for either doctrine. As mentioned above, when there is no hierarchy of tribunals, it is difficult to determine which case should be authoritative. As a result, there are unavoidably contradictory holdings. Moreover, when the legal regime is not yet well developed, there are gaps in the law that can only be filled after there is a sufficient number of significant cases that discuss and analyze the issue from different angles in various factual circumstances. While establishing a hierarchy structure may not be appropriate, and a legal regime is not perfected in one day, we can make progress and approach consistency through an effective accumulation of comments on previous cases. In the beginning, tribunals might diverge on new issues. With an increased number of cases concerning the same issue, all of which comment on previous cases with their own analysis and opinions, the discussions around such an issue will eventually be enough to address detailed points and form consensus so as to subsequently develop into newly emerged rules. This in turn helps the development of the international investment law regime.

B. The Main Consideration in the Review: ‘Development’

The inspiration to use the term ‘develop’ comes from one of the fundamental roles that precedents play, ie the development of the law to address changes.⁶⁶ As Lord Denning put it: ‘[I]f we never do anything which has not been done before, we shall never get anywhere. The law will stand still whilst the rest of the world goes on: and that will be bad for both.’⁶⁷ While following precedents ensures consistency, a departure from precedents welcomes changes and developments.

Admittedly, the meaning of ‘develop’ seems to be too vague in the beginning. The tribunal might construe ‘develop’ in a broad variety of ways and thus depart from prior cases easily. However, with more and more applications of the test, the connotation of the test will be enriched, and its meaning will become clearer over time.

This formulation of the soft test encourages discussions around ‘developing’ international investment law. It increases opportunities for various practitioners to express their opinions and finally form a consensus regarding the core values and recent changes in the legal regime as well as social reality. The soft test actually puts pressure on the parties and their counsel to study relevant previous rulings concerning their cases to analyze whether such rulings are worth following in regard to the development of international investment law. If they do not study relevant previous cases sufficiently, they may face the risk of being caught off guard by the counterparties who have done the research and developed their arguments according to the prior cases. Meanwhile, tribunals also have to pay attention to prior cases and reflect on what benefits the development of international investment law. Such a soft obligation to consider previous opinions prevents tribunals from talking to themselves and forces the arbitrators to ‘have conversations’ with each other on investment law issues. This would also trigger more research in academia which would accelerate the maturity of the international investment legal regime.

In the end, consistent rules in investment law will be formed which reflect the efforts of the entire investment arbitration community over the years. Such a legal system would be rooted in dispute resolution practice, which endows it not only with consistency and predictability as a legal system but also credibility and popularity among investors.

C. The Scope of Previous Cases

When implementing the soft precedents test, the first practical issue would be what constitutes previous rulings which should be considered.

In the domestic legal systems, take the US stare decisis as an example: vertical stare decisis means that decisions of higher courts take precedence over the decisions of lower courts; horizontal stare decisis holds that prior decisions made by courts at a particular appellate level (such as a federal court of appeals) should provide some precedent for cases heard by courts of the same appellate level. In the WTO tribunals, jurisprudence constante is only expressly applicable to Appellate Body Reports. It is not hard to find that the scope of ‘previous cases’ in the doctrines is always cases from higher or equivalent-level courts.

However, international investment arbitrations have a special feature that makes the scope of previous cases hard to define, which is the fragmented nature of various arbitration institutions including the significant number of ad hoc arbitrations. Such a structure of dispute resolution forums makes it difficult for any tribunal to address every single jurisprudence relevant to the dispute at hand. Requiring the tribunal to do such thorough research is also unrealistic since it would take significant time and effort which would be reflected in the financial burden of the parties, rendering the regime of investment arbitration less popular among investors.

As a result, the scope of previous cases that the tribunal should consider has to be confined. With its core value being party autonomy, it might be appropriate to only require the tribunals to consider prior cases brought up by the parties. In this way, the parties may cite previous rulings from the same or other arbitration institutions to support their arguments, and the tribunals only have to examine a limited number of cases that reflect the parties’ understandings of the law.

Meanwhile, according to their iura novit curia powers, it is submitted that tribunals should be given the power to cite previous cases on their own action and invite the parties to opine on those cases.⁶⁸ This is to prevent the risk of a party being negligently unaware or intentionally silent on certain important cases, which might result in insufficient discussion. It is not rare in international investment arbitrations that one party might face a scarcity of legal resources (especially in developing countries). Granting the tribunals the power to bring up cases would mitigate such inequality in legal resources. Moreover, since international investment law is still in the early stage of development, not all legal practitioners are well versed in the intricacies of various fields in international investment law. The arbitrators, who are generally experienced in their specific fields, can serve as good guides in finding the most relevant cases.

Nevertheless, it needs to be clarified that, although tribunals can bring the parties’ attention to relevant prior cases cited on the tribunals’ own initiative, they merely serve as reminders or suggestions from the tribunals. Utimately, it is the parties who decide which cases to rely on when arguing for their claims. Moreover, iura novit curia powers of investment arbitration tribunals have some limitations, and such powers should be applied within such scopes.⁶⁹

D. Legal Effect of not Complying with the Test

The soft precedents test might not make so much difference per se without a corresponding mechanism to incentivize tribunals to apply it substantially. In this regard, the setting aside and recognition and enforcement mechanisms based on the New York Convention, as well as the annulment procedure under the ICISD Convention might be good choices to operate in combination with the test.

For ICSID arbitration, the annulment mechanism provided in Article 52 of the ICSID Convention is the applicable way of remedy that could be sought by the parties. As can be seen from the text itself, grounds for annulment are limited without touching upon any substantive issue of law. Nevertheless, were the soft precedents test to be established as a principle in investment arbitration, it is proposed that the unjustified failure to apply such test may constitute a manifest excess of powers under Article 52(1)(b), and/or failure to state the reasons on which the award is based under Article 52(1)(e), and thus the arbitral award could be annulled.

In this regard, the annulment committee in Sempra offered notable inspiration, which held that the Award should be annulled on the ground of excess of power since the Tribunal failed to apply Article 11 of the US–Argentina BIT. In this sense, it is suggested that if an award reveals that the tribunal failed to adequately consider similar cases or prove that its departure from prior cases was to facilitate the development of international investment law, it could be considered an excess of power since it failed to decide the case with the law cited by the parties. Admittedly, ‘facilitate the development’ is a rather loose phrase. Since there is no uniform definition, it is considered that as long as a tribunal discusses the reason for departure in the view of the development of international investment law, it should probably not be considered an excess of power. This combination would not contradict the current concept of annulment because such an examination stays solely on the surface of the tribunal’s behavior and does not go so much into the errors of substantive law.

For non-ICSID arbitration, the applicable remedy is commonly regulated by Article V (recognition and enforcement) and Article VI (setting aside) of the New York Convention. In particular, Article V(1)(c) provides that courts may refuse to recognize or enforce an award if ‘it deals with a difference not contemplated by or not falling within the terms of the submission to arbitration, or it contains decisions on matters beyond the scope of the submission to arbitration’. In this scenario, the lack of consideration of prior rulings that the parties have opined on may imply the tribunal’s failure to make its rulings by deciding on specific controversies within certain issues that have already been covered by the cases in the parties’ submissions. The initiative of the tribunal to discuss and decide the whole issue ‘anew’ beyond the scope designated by the parties might then constitute the condition for setting aside as well as non-recognition and enforcement provided by Article V(1)(c).

V. CONCLUSION

Due to the public nature of international investment arbitration, consistency is required to enhance its credibility and legitimacy. Nevertheless, certain flexibility is also required in view of the fragmented nature of this mechanism and its current stage of development. Considering that strict stare decisis or jurisprudence constante are not suitable for international investment arbitration, and that existing attempts to achieve consistency have significant limitations, there is a need for an innovative approach to tackle the issue of inconsistency under the present situation of international investment law.

A soft precedents test is thus proposed to achieve the balance between the need for consistency and flexibility. The test requires tribunals to consider in good faith previous rulings and reasonings when facing the same or similar issues discussed in prior cases, and the core consideration in whether to agree with previous cases is to develop international investment law. It is believed that the application of such a test would encourage discussion on specific issues in tribunals, and an effective accumulation of remarks on prior cases could be achieved. Such discussion and accumulation would help clarify, enrich and develop the content of the law while enhancing the consistency and coherence of awards, which would in turn improve the legitimacy and credibility of the investment law regime.

Footnotes

Author notes