Challenging Domestic Judgments Through Investment Arbitration: Implications for the Forced Labour Litigation in Korea?

Abstract

This paper considers the international legal implications of certain civil lawsuits that recently resulted in Japanese companies being ordered to pay compensation to Korean victims of forced labour, focusing specifically on whether investor-state arbitration could provide a means of redress against judgments affecting those companies. After identifying the jurisdictional hurdles that those companies might face were they to challenge those judgments before an Arbitral Tribunal, this paper explores the most relevant treaty protection disciplines that could be relied upon in bringing such challenges, and discusses the remedies that a competent Arbitral Tribunal could prescribe were it to find that those judgments were not in conformity with Korea's international obligations. Building on existing jurisprudence, this paper shows that investment arbitration may provide a means for offsetting the adverse consequences of Korean forced labour litigation, but also highlights a number of difficulties that the Japanese companies would face in pursuing such an avenue.

In the autumn of 2018, the Supreme Court of Korea delivered final judgments in civil lawsuits brought against certain Japanese companies for damages suffered by Korean nationals on account of forced labour during World War II. In the first of these judgments, dated 30 October 2018, the Court upheld an earlier ruling against New Nippon Steel Corporation, ordering the payment of compensation to several Korean nationals who were found to have been forced to work at the company's Osaka Steel Mill between 1943 and 1945.¹ In two further judgments dated 29 November 2018, the Court issued similar orders in lawsuits that were brought against Mitsubishi Heavy Industries on similar grounds.² Following the Supreme Court's decisions, other pending lawsuits involving forced labour apparently picked up speed,³ while over a thousand other plaintiffs reportedly moved to file similar suits against the same, as well as a range of other Japanese companies, including Mitsubishi Materials, Nippon Coke & Engineering, Sumiseki Holdings, Nippon Mining & Metals, Nachi-Fujikoshi, Nishimatsu Construction, and Hitachi Zosen.⁴

The question of law that has been at the heart of these lawsuits, and that was also squarely at issue before the Supreme Court, was whether such claims were actually permitted by the 1965 Agreement Between Japan and the Republic of Korea Concerning the Settlement of Problems in Regard to Property and Claims and Economic Cooperation [the 1965 Agreement],⁵ one of the agreements through which the two states sought to normalize relations following Japan's colonial occupation of Korea. In Article 2 of this Agreement, the two states namely “confirm[ed] that [the] problem concerning property, rights and interests of the two Contracting Parties and their nationals (including juridical persons) and concerning claims between the Contracting Parties and their nationals … is settled completely and finally”. The Korean Supreme Court eventually interpreted Article 2 as not preventing plaintiffs from exercising their individual right to claim compensation for forced labour. According to the Court, the claims for compensation were not civil law claims for unpaid wages, but tort claims “premised on the inhumane and wrongful act of the Japanese corporation directly related to Japan's unlawful colonial rule of the Korean Peninsula and its war of aggression”.⁶ These claims, which “directly related to the illegality of colonial rule”, could “hardly be deemed to be subject to the application of the Claims Agreement”—at least in the absence of a “specific reference” to Japanese colonial rule in the 1965 Agreement—and were therefore not within the scope of claims that could be deemed to have been “completely and finally settled” within the meaning of Article 2.⁷

As one may have expected, the Supreme Court's decisions further strained the already complicated relations between Japan and Korea, long marred by historical and territorial disputes originating from the colonial era. Japan strongly protested against the judgments, viewing all claims for reparations, including individual claims, to have been definitely resolved by the 1965 Agreement. Japan considered the judgments not only “extremely regrettable and totally unacceptable”, but also took the view that they “clearly violate Article II of the Agreement”.⁸ On 20 May 2019, deeming that a disagreement had arisen concerning the interpretation and implementation of the 1965 Agreement, Japan formally requested the referral of the dispute to arbitration, as provided for under Article 3 of the 1965 Agreement.⁹ The request came after Japan had already proposed in November 2018 that the dispute be submitted to the International Court of Justice [ICJ].¹⁰ However, with Korea resisting Japan's proposals for third-party settlement, the dispute in the meantime moved to a different arena. In July 2019, Japan imposed export sanctions directed at the Korean semiconductor industry, to which Korea responded by filing a formal complaint with the World Trade Organization. A month later, Japan proceeded to remove Korea from its “white list” of countries that receive preferential trade treatment, to which Korea retaliated by removing Japan from its own “white list” of trusted trade partners.¹¹

In the absence of a clear jurisdictional basis for the dispute to be submitted to the ICJ, or Korea's acceptance of arbitration pursuant to the procedure provided for under the 1965 Agreement,¹² the correctness of the Korean courts’ interpretation of Article 2 of that Agreement appears unlikely to be soon reviewed through a state-to-state dispute settlement mechanism. But since particular assets have already been seized from some of the Japanese companies involved in the domestic suits, and since these assets are in the process of being liquidated in satisfaction of the domestic judgments,¹³ the question arises whether the companies themselves may seek recourse to some international jurisdiction to challenge the adverse judgments. Concretely, the question is whether the companies could avail themselves, for that purpose, of the mechanism of direct investor-state arbitration, such as the one provided for under Article 15(3) of the Korea-Japan bilateral investment treaty [BIT] (1992),¹⁴ or some other applicable BIT.

The question may be less hypothetical than it may seem. Over the past years, investment arbitration has already proven useful for contesting domestic judicial decisions considered to be injurious by foreign investors. One of the most notable examples is the US oil company Chevron's successful challenge of a judgment rendered by an Ecuadorian court in the Lago Agrio litigation. The company would originally have been required to pay almost US$10 billion in damages for environmental pollution, were it not for a treaty-based Arbitral Tribunal finding that the impugned judgment had been procured through corruption, and consequential ordering that its enforcement be suspended.¹⁵ Further, the example is not an unusual one. Prior to that, Chevron itself, as well as other investors, had readily resorted to investment treaty arbitration to seek redress for delays deriving from inaction of domestic judiciaries,¹⁶ domestic courts’ refusals to enforce commercial awards,¹⁷ or judicial interferences in ongoing commercial arbitrations.¹⁸

Building on this jurisprudence, the present paper proceeds to examine whether investor-state arbitration could provide Japanese companies with some form of redress against judgments rendered against them in the forced labour cases. The discussion begins with certain general considerations concerning the possibility of challenging the propriety of the Korean courts’ judgments before Arbitral Tribunals deciding investment disputes (Part I). It then proceeds to examine the most relevant treaty protection disciplines in which claims brought to such tribunals could be grounded (Part II). The discussion then moves to considering whether proceedings before those tribunals could also accommodate claims directly grounded in purported violations of the 1965 Agreement (Part III). The paper eventually touches on the consequences of a potential finding that the Supreme Court's judgments are not in conformity with Korea's international obligations, and examines the remedies that a competent Arbitral Tribunal could prescribe (Part IV), before it concludes with some final observations (Part V).

I. ADJUDICATION OF JUDICIAL RESPONSIBILITY WITHIN NARROW JURISDICTIONAL CONFINES

There is little doubt that a potential failure on the part of Korean courts to correctly interpret the scope of the waiver stipulated in Article 2 of the 1965 Agreement and the potentially improper application of the Agreement resulting from such interpretation would be capable of engaging the responsibility of Korea under international law. The fact that the organ interpreting and applying the 1965 Agreement was a judicial organ, operating independently from the Korean government, obviously plays no role in that respect. From the perspective of contemporary international law, which treats the state as a unitary actor, courts are no different from other state organs.¹⁹ Their conduct is attributable to the state and is capable of engaging its responsibility just like the conduct of any other organ.²⁰ In the jurisprudence of international courts and tribunals, one can find sufficient support to that effect, not only in pronouncements at the level of principle, but in the many cases where the conduct of the judiciary as such formed the predicate of an international claim for failing to conform to the state's obligations under customary international law,²¹ or under specific conventions.²²

In practice, state responsibility will often be engaged on account of the courts’ unwarranted exercise of adjudicative powers (such as when they exercise their jurisdiction beyond the bounds fixed by international law, or fail to accord immunities from such jurisdiction),²³ or their disrespect of obligations demanding specific procedural treatment in the context of the domestic adjudicative process (such as when they fail to provide a fair trial).²⁴ But these are not necessarily the only types of violations potentially associated with judicial conduct. Given that it is essentially the courts that determine the existence and scope of rights which find subsistence in domestic law (as well as establish in whom these rights might be vested), courts’ conduct is likely to engage any state obligation that has the object of regulating matters within the remit of domestic law. Typical in this respect would be obligations protecting proprietary rights and interests, such as those that can be found in the more traditional friendship, commerce, and navigation treaties, or nowadays predominantly in bilateral and other investment treaties. The obligations stipulated in Article 2 of the 1965 Agreement arguably fall into the same category, as they regulate the “problem concerning property, rights and interests of the two Contracting Parties and their nationals”. Indeed, that stipulations of such kind could be violated through the courts’ conduct is not inconceivable. In Decision No. 196 (1955), for example, the Italo-French Conciliation Commission notably established that France was responsible for judicial liquidations of certain property of Italian nationals in Tunisia because such measures contravened the provisions of Article 79 of the 1947 Treaty of Peace with Italy, which proscribed the seizure of that particular category of private property.²⁵

Still, in challenging the Korean courts’ judgments for their potential lack of conformity with the obligations under Article 2 of the 1965 Agreement, the Japanese companies would not be confronted with an easy task. Like other international adjudicatory bodies, investment tribunals are not tribunals of unqualified, general jurisdiction. They are creatures of the Parties’ consent, which provides the necessary basis for, but at the same time also sets limits to, their adjudicatory powers. Under some treaties, such powers extend over all disputes concerning an investment. Under others, however, they are limited to disputes concerning violations of the instrument in question. The latter is also the case with the Korea-Japan BIT, which permits covered investors to recover only losses or damage incurred “by reason of, or arising out of, an alleged breach of any right conferred” by the BIT (Article 15(1)). Given these limitations, it is hence not immediately obvious whether (or how) the alleged judicial violations of Korea's obligations under the 1965 Agreement could actually be presented for adjudication to an investment tribunal established pursuant to the Korea-Japan BIT.

The problem, to be sure, is not whether a tribunal established pursuant to an investment treaty could take cognizance of a host state's international obligations other than those arising under an investment treaty. Whether by reference to Article 31(3)(c) of the Vienna Convention on the Law of Treaties or otherwise,²⁶ there is nothing that prevents such tribunals from interpreting the provisions of an investment treaty by reference to other relevant rules of international law and taking such rules into account in the treaty's application. But it is one thing to consider the existence or scope of other obligations under international law, and arguably something else to determine if those same obligations have been complied with. In practice, the problem is typically dealt with as one of “incidental jurisdiction”—that is, as one concerning the extent to which an adjudicatory body can make pronouncements on issues that lie outside the scope of its subject-matter jurisdiction but that are indispensable for deciding issues that lie within its material scope of adjudication.²⁷

In general, international courts and tribunals have not looked unfavourably towards extending their jurisdiction to determining such incidental-yet-indispensable matters, provided that such matters did not concern the very object of the dispute. In Polish Upper Silesia, the Permanent Court of International Justice [PCIJ] famously posited that “the interpretation of other international agreements is indisputably within the competence of the Court if such interpretation must be regarded as incidental to a decision on a point in regard to which it has jurisdiction”.²⁸ In determining whether Poland violated the Upper Silesia Convention, the PCIJ then also effectively proceeded to incidentally determine whether Germany had not violated, among other instruments, the Treaty of Versailles.²⁹ More recently in the Chagos case, an UNCLOS Annex VII Tribunal repeated the PCIJ's stance, explaining that “[a]s a general matter”, where a dispute concerns the interpretation or application of a treaty, the jurisdiction of a court or tribunal established pursuant to such a treaty “extends to making such findings of fact or ancillary determinations of law as are necessary to resolve the dispute presented to it”.³⁰ The Tribunal did not therefore “categorically exclude” the possibility of ruling on a minor issue of territorial sovereignty if such issue were ancillary to a dispute concerning the interpretation or application of the Law of the Sea Convention [LOSC], over which it indisputably had jurisdiction.³¹ But the Tribunal nonetheless cautioned that, where the “real issue in the case” and the “object of the claim” did not relate to the interpretation or application of the jurisdiction-conferring treaty, an incidental connection between the dispute and some matter regulated by that treaty would be insufficient to bring the dispute, as a whole, within the ambit of a tribunal's jurisdiction.³²

Along the same lines, investment tribunals were not hesitant to exercise incidental jurisdiction over legal issues arising under external instruments where such issues were essential for determining claims founded on the investment treaty. In a number of cases, for example, investment tribunals confirmed that they enjoyed the competence not only to take note of and interpret the contracts on which those treaty claims were grounded,³³ but also to determine potential breaches of those contracts insofar as such breaches were relevant to ascertaining whether the respondent states violated their obligations under the relevant treaties.³⁴ But such practice has not only been pursued with regard to points of domestic law. As the succeeding analysis will demonstrate, many tribunals in fact considered themselves equally competent to determine incidental points of international law that were relevant to determining whether violations of the applicable investment treaty occurred. It is with this in mind that the next section now turns to considering whether the obligations under the 1965 Agreement could equally inform the determination of whether Korea has complied with the standards of treatment it was supposed to guarantee under the Korea-Japan BIT with its courts deciding the way they did the forced labour cases.

II. PURSUING THE INDIRECT ROUTE: CHALLENGING ADVERSE JUDGMENTS THROUGH INVESTMENT PROTECTION DISCIPLINES

The possibility thus examined implies pursuing an indirect route: the idea is one of using the prescribed investment treaty standards as a way to indirectly raise the question of Korean courts’ compliance with the provisions of the 1965 Agreement by presenting such a question as a necessary predicate of claims grounded in the Korea-Japan BIT itself.³⁵ This requires first identifying the investment treaty obligations that lend themselves to such use. In principle, a great number of investment treaty standards may well be capable of being engaged through court conduct. Investments protected by treaties such as the Korea-Japan BIT are grounded in proprietary rights governed by the host state's law and operate within the host state's legal framework. In being thereby subject to the adjudicatory jurisdiction of the host state's courts, they are also susceptible of being adversely affected by specific judicial measures. It is nonetheless the case that some standards of treatment are more likely to be violated through the intermediary of the courts than others.

In Article 3 of the Korea-Japan BIT, one can find thus a separate obligation concerning the host state's judicial system. This demands that Parties accord to each other's investors national- and most-favoured-nation treatment “with respect to access to the courts of justice and administrative tribunals and agencies in all degrees of jurisdiction, both in pursuit and in defense of such investors’ rights”. In view of the way it is formulated, the provision appears mainly to protect investors against one of the most elementary forms of denial of justice—i.e. that resulting from obstruction of access to courts. Though arguably capable of being violated through specific court conduct, however, the provision does not appear to be of much assistance to the Japanese companies. The Korean judicial system remained generally accessible to the Japanese companies to defend themselves in the lawsuits concerning forced labour claims, and did not seem to discriminate against them in their pursuit or defence of their rights.

Consideration must hence be given to those general standards of treatment which have in practice been more frequently resorted to with a view to challenging adverse judicial measures. The discussion shall focus first on an obligation that is most closely associated with the conduct of domestic courts—the obligation not to deny justice in judicial procedures, which is deemed to be subsumed under the fair and equitable treatment [FET] standard (Section A). The inquiry will then move to considering provisions that are geared towards providing security to the rights underpinning the investment, namely the protection of legitimate expectations under the FET standard (which aims to guarantee the integrity of the legal framework in which the investment is imbedded) (Section B); the expropriation clause (which prohibits uncompensated takings of the investment) (Section C); the umbrella clause (which aims at ensuring respect of commitments entered into by the host state in relation to the investment) (Section D); and the “effective means” clause (which aims to guarantee a system for asserting claims and enforcing rights with respect to investments) (Section E).

A. Non-compliance with the 1965 Agreement as a Breach of the Fair and Equitable Treatment Standard in the Form of Denial of Justice

Building on past arbitral practice, it would not be unusual if the companies affected were to challenge the adverse judicial measures on the ground of denial of justice. It is not only because of the latter that, historically, the responsibility of states has most often been engaged by their judicial organs. It is also through the lens of denial of justice that investment tribunals have most often proceeded to appraise the propriety of impugned judicial action. This practice is not surprising. The prohibition of denial of justice is deemed to form part of the minimum standard of treatment which, in addition to other requirements, demands from states the maintenance of a system of justice that treats foreigners fairly and impartially, and that generally affords adequate judicial protection to their rights.³⁶ But this same obligation is also considered to be subsumed under the FET standard guaranteed by investment treaties, so that in practice, conduct passing the threshold of the customary delict of denial of justice has generally been taken to constitute a violation of the FET standard as well.³⁷ Though the Korea-Japan BIT would not appear to allow injured investors to bring claims regarding violations of the minimum standard of treatment as such, the obligation to accord “fair and equitable treatment and full and constant protection and security” under Article 10(1) of the BIT would enable the affected companies to bring a claim for denial of justice. The question, however, is how such a claim could be formulated.

1. Meeting the high threshold for denial of justice

Despite more than a century of jurisprudential developments, denial of justice remains an “elusive” concept. In theory, cases of denial of justice are essentially instances of states failing to live up to their fundamental duty to provide judicial protection to investors’ rights. But this duty has proven anything but simple to operationalize. The obligation to maintain an adequate system of justice is namely not one imposing defined, substantive outcomes. In its essence, it foremost entails that, whenever a foreigner seeks to vindicate substantive rights through a domestic judicial process, that foreigner be accorded procedural justice, as determined by reference to the standards provided by international law.³⁸ Today, these standards are deemed to include elements such as the right to an independent and impartial court established by law, the right to have the case heard and determined within a reasonable time, the right to a reasonable opportunity to present the case, the right to equality of arms, and the right to a reasoned decision.³⁹ Not unexpectedly then, instances of denial of justice have most commonly been associated with (1) denial or obstruction of access to courts; (2) unwarranted judicial delays; (3) serious deficiencies in the conduct of judicial proceedings; and (4) potentially judgments that are manifestly unjust or otherwise improper.⁴⁰

In fashioning their claim as one of denial of justice, the Japanese companies would have a high standard to meet. To be considered discreditable from the standpoint of international law, the improprieties experienced in judicial proceedings need to be grave or serious ones, with the claimant being effectively required to prove that the national system as a whole had failed to satisfy minimum standards.⁴¹ The burden of proof will thereby not be lightly discharged. Tribunals have namely accepted that a national legal system will benefit from the general evidentiary presumption that its courts have acted properly, and that the judiciary shall be permitted a margin of appreciation before the threshold of a denial of justice will be met.⁴² An additional burden lies in the judicial finality rule, which entails that denial of justice can only arise once the judicial system as a whole has been tested and the existing judicial remedies have failed to correct the deficiencies in the lower courts’ judgments, or that the remedies are such that they do not afford the foreign investor any reasonable prospect of correcting those deficiencies in a timely, fair, and effective manner.⁴³

Whether or not Korea could be said to have failed to make available to the Japanese companies a fair and impartial system of justice that afforded adequate judicial protection to their rights would have to be appreciated in the light of the circumstances of each individual lawsuit. In general, however, there appears to be no issue of Japanese companies being denied access to Korean courts. Nor is there anything yet to suggest that the treatment of the forced labour claims has been subject to delays that, by reference to existing case-law, could be said to be unreasonable.⁴⁴ In the light of publicly available information, there is furthermore little indication that in the litigation of those cases the Japanese companies had not been accorded due process, or had otherwise experienced procedural treatment falling below the standards of judicial propriety. With respect to the forced labour litigation, a potential claim would appear to turn primarily on the question whether the individual judicial decision ordering the payment of compensation in itself amounted to a denial of justice. This is a question to which we turn next.

2. Responsibility for the content of judicial decisions?

One of the most vexed issues in the discussions on denial of justice has traditionally concerned the circumstances in which a state could be held responsible for the content of judicial decisions.⁴⁵ What has been contested is not so much whether a denial of justice could arise as a result of defects in the substance of judgments as such.⁴⁶ Rather, differing views have existed about the kind of defects that would render a judgment improper from the perspective of international law. This traditionally proved to be a delicate task, in part because of the general admonition that international adjudicatory bodies are not there to function as courts of appeal from domestic judicial systems. In the classical writings on the topic, the view was nonetheless maintained that a judgment was to be considered defective from the standpoint of international law when it was “manifestly unjust”⁴⁷—a formula fraught with imprecision that was difficult to apply in practice.

In the context of present-day investment arbitration, in contrast, the view has gained support that the substance of a judgment can only be of relevance to the extent that it provides evidence of lack of due process or other procedural improprieties, and that an international tribunal never really engages in a review on substance.⁴⁸ The benchmark most commonly used has been that of “judicial propriety”.⁴⁹ According to the Mondev Tribunal, “[t]he test is not whether a particular result is surprising, but whether the shock or surprise occasioned to an impartial tribunal leads, on reflection, to justified concerns as to the judicial propriety of the outcome”—an issue which ultimately boils down to the question whether the judgment can be considered “clearly improper and discreditable”.⁵⁰ But as adjudicators applying the imprecise formula of “manifest injustice” in the past, investment tribunals of today disagree as to when a judgment would meet the required level of impropriety. In the view of some tribunals, denial of justice could be pleaded in the event of a “clear and malicious misapplication of the law”.⁵¹ In the view of others, the judgment would have to be “egregiously wrong”, of the kind that no “competent judge could reasonably have made”.⁵² Then again others considered the test to turn on the question of “clear and manifest illegality”.⁵³ Whilst the assessment frequently appeared to depend on the adjudicators’ own perceptions of what is proper, the threshold for the requisite “impropriety” was generally deemed to be a high one. In order to engage the responsibility of the state, much more has been considered necessary than a mere error in the interpretation or application of the law.

Most of the discussions on “substantive” denial of justice traditionally turned on the circumstances under which the application of domestic law itself was such as to give rise to the responsibility of the state under international law. But in the classic writings on the topic, it was at the same time accepted that denial of justice could also be predicated upon domestic courts’ misapplication of international law.⁵⁴ The cases typically invoked were those of judicial decisions rendered in violation of provisions of an extradition treaty, of judgments contrary to special rights secured to foreigners under certain treaties of commerce, or of decisions rendered in the exercise of a competence not recognized under international law.⁵⁵ But as the idea was gradually recognized that domestic courts can also violate international law in other ways than through denying justice to foreigners,⁵⁶ this prompted discussions as to the precise line dividing instances of denial of justice from “other” international wrongs occasioned by the judiciary. On the one hand, some have considered the general duty of judicial protection to concern solely how courts apply domestic law with respect to foreigners, treating those courts’ misapplications of international law as discrete violations of customary or conventional obligations—and not as denials of justice.⁵⁷ On the other hand, others considered the non-observance by courts of concrete obligations under treaties or customary international law to be capable of qualifying as denials of justice whenever the international norm in question was one aiming at the legal protection of an individual right. For every time one such norm failed to obtain recognition in the context of a domestic adjudicative procedure, one could arguably also speak of a non-fulfilment of the duty of judicial protection and thus of denial of justice.⁵⁸

The distinction is of course relevant if the Japanese companies were to attempt to challenge the impugned judgments on the ground of denial of justice. At the heart of the judgments arising out of the forced labour litigation is namely not so much the question of misapplication of Korean law, but that of the purportedly wrong interpretation and application of the 1965 Agreement. Under a broad understanding of the duty of judicial protection, a failure on the part of the Korean courts to give effect to a substantive right granted to Japanese companies under the 1965 Agreement could possibly then serve as a predicate of a denial of justice claim, thereby allowing the question of conformity of Korean courts’ judgments with the 1965 Agreement to be brought within the jurisdictional scope of the Korea-Japan BIT. Such a move would not be entirely unprecedented. In the Al Warraq case (2014), the Tribunal similarly drew on substantive rights granted to the investor under another treaty—namely, the right to a fair trial under Article 14 of the International Covenant on Civil and Political Rights [ICCPR]⁵⁹—with a view to determining whether the Respondent had violated the FET standard through its courts failing to accord the investor justice in domestic criminal proceedings.⁶⁰

Predicating the claim on the violation of the 1965 Agreement would potentially also lower the otherwise very high threshold for a finding of denial of justice. In accordance with classical writings on the topic, claims predicated on the misapplication of international rules did not require proof of maliciousness on the part of the courts: mere error in the interpretation or application of the international rule was deemed sufficient to give rise to responsibility, even where the judgment in question had been rendered in perfect good faith, by an honest and competent court.⁶¹ Thus also the Arbitral Tribunal in the Martini case (1930)—in holding Venezuela responsible for manifest injustice on account of its Court of Cassation's Judgment that failed to observe Venezuela's international obligations under an earlier arbitral award—did not pay heed to the argument that the Court's error was merely an exercise of bad judgment, and as such insufficient to establish responsibility.⁶² In the view of the Arbitral Tribunal, the psychological motives of the judges were of “no importance” with regard to Venezuela's responsibility for an obvious violation of its international obligations.⁶³

3. Does Article 2 create rights or entitlements for private parties?

The success of any such claim of denial of justice would however be conditional upon whether the 1965 Agreement actually guarantees Japanese companies specific material rights—or at the least some sort of legal entitlement that would confer upon them standing to invoke the violation of the Agreement. Today, it is generally established that private parties may be the recipient of direct rights under international treaties where the States Parties so intended.⁶⁴ It is furthermore accepted that, even in the absence of such direct rights, private parties may be entitled to enforce violations of certain state obligations to the extent that such obligations create benefits in their favour.⁶⁵ Arguably, it is precisely in their position of third-party beneficiaries of arrangements under investment treaties that investors are considered to enforce violations of treaty standards through the mechanism of investor-state arbitration.⁶⁶

Whether private parties have been granted direct rights or conferred-upon benefits of course depends on the concrete language of the treaty. This is ultimately a matter of treaty interpretation. In following the ICJ, any rights or interests of such a kind would, at any rate, have to be conferred upon in an unequivocal manner, either directly or by a clearly necessary implication.⁶⁷ In the case of Article 2 of the 1965 Agreement, one can note that the provision is not formulated in the language of rights or obligations towards private parties. The Contracting Parties merely “confirm” that the problem concerning property, rights, and interests and concerning claims “is settled completely and finally”. However, insofar as Article 2 makes explicit reference to Contracting Parties’ “nationals (including juridical persons)”, it seems nonetheless possible to construe the provision as at the very least creating benefits for private parties. What is then debatable is rather the nature and scope of the benefits thus conferred.

Though avoiding explicitly addressing the question of Japanese companies’ standing to invoke Article 2 of the 1965 Agreement, the majority judgment in the New Nippon Steel case (2018) implicitly accepted the premise that those stipulations were capable of conferring certain rights or benefits upon Japanese companies. In following the majority's reasoning, it was just that the legal import of those rights or benefits was not such as to extend to the type of claims that were at issue in the proceedings.⁶⁸ Admittedly, the language of Article 2 appears capable of accommodating other interpretations, too: on the face of it, the reference to the “problem concerning property, rights and interests of the two Contracting Parties and their nationals (including juridical persons) and concerning claims between the Contracting Parties and their nationals” is broad enough to apply to the type of claims that were at issue in the forced labour cases.

It is a different question whether the benefits derived from Article 2 also operate as between the nationals and entities of the two Contracting Parties; that is, whether the waiver also has horizontal effects. The reference to the “problem concerning property, rights, and interests of [emphasis added] the two Contracting Parties and their nationals” does appear broad enough to extend to nationals’ right to compensation. The reference to “claims between [emphasis added] the Contracting Parties and [emphasis added] their nationals” appears, on the contrary, more limitative, as it could be construed as excluding claims between nationals inter se. But the language is sufficiently ambiguous that one cannot entirely rule out the possibility that claims brought by nationals of one state against the companies of another state fall equally within the purported waiver under Article 2 of the 1965 Agreement.

All in all, the language of Article 2 therefore appears capable of being construed as conferring upon the Japanese companies the right not to be exposed to claims for damages of the kind that have been at issue in the forced labour case, and thus providing the predicate for a claim of denial of justice grounded in the courts’ failure to fulfil their duty of judicial protection with respect to that right.

B. Breach of FET Through a Judicial Frustration of Legitimate Expectations

Though perhaps a less obvious route, another way for the affected companies to challenge the adverse judgments would be to present them as a breach of legitimate expectations—such as those that may arise on the basis of promises or assurances that a host state makes with a view to attracting investments, but also generally in relation to the state of the law and the totality of the business environment existing at the time of the investment. The protection of such expectations, alongside the maintenance of regulatory stability, belongs to the core elements of the FET standard.⁶⁹ For an external observer it is of course impossible to determine whether any of the Japanese companies affected had at any time received specific assurances from the Korean government as to them being shielded against forced labour claims in the Korean courts. This can only be determined by reference to the concrete circumstances of each case. What can nonetheless be considered, however, is whether the companies affected may have legitimately expected that the Korean courts would have interpreted and applied the 1965 Agreement in a certain way.

Investment tribunals have readily accepted that, as a matter of principle, an investor's legitimate expectations can be frustrated on account of the conduct of the courts themselves.⁷⁰ Indeed, on at least one occasion, a state has been held responsible for breaching the investor's legitimate expectations through its courts’ annulling an investor's contract—albeit that much of that finding rested heavily on the conduct of other state organs that had previously created the expectation that the contract was legally valid.⁷¹ In the absence of inconsistent representations by other state organs, challenging the conduct of courts under this heading of the FET standard could perhaps be attempted in two ways. As in the case of unexpected changes to the regulatory environment, which in some cases were deemed a sufficient reason for finding a violation of the standard,⁷² the injured investor could argue that the courts’ adjudication departed from established jurisprudence. However, not every such departure would necessarily amount to a breach of legitimate expectations. As the Tribunal in Eli Lilly v. Canada (2017) aptly observed, an investor “should have, and could have, anticipated that the law would change over time as a function of judicial decision-making”.⁷³ In the context of the Korean forced labour litigation, of course, the issue is not whether the courts’ jurisprudence has been subject to change. The most recent decisions of the Korean Supreme Court do not appear to have departed from previously established case-law, as the scope of Article 2 of the 1965 Agreement had not been subject to adjudication earlier.

Another way would perhaps be to challenge the outcome of domestic adjudication for being contrary to a specific, legitimately expected outcome. This touches on the more fundamental question whether the investors can legitimately expect that domestic courts shall necessarily act in accordance with obligations incumbent upon the host state under international law.⁷⁴ The issue has thus far presented itself in the White Industries v. India (2011) case. There, the Tribunal was called to consider whether the investor could legitimately have expected Indian courts to apply the 1958 New York Convention⁷⁵ properly and in accordance with international standards. But the Tribunal dismissed such a possibility in circumstances where Indian courts were known for regularly entertaining set-aside applications in respect of foreign awards, despite such practice being contrary to India's obligations under the said Convention.⁷⁶ Similar considerations would seem to apply in the circumstances of the Korean forced labour litigation. While the 2018 decisions provide the Supreme Court's definite ruling on the scope of Article 2 of the 1965 Agreement, the interim decisions rendered in 2012 had already demonstrated that the Supreme Court would not consider the said provisions to constitute a bar to domestic forced labour claims.⁷⁷ In circumstances where the Supreme Court had already developed its views on the scope of Article 2, the Japanese companies could not legitimately expect that the Court would interpret the provision differently.

There are furthermore principled reasons why it may be inappropriate to legitimately expect judicial organs to always abide by international law. For one, in many legal systems, courts may not have the ability to reach decisions that are in conformity with, rather than contrary to, international law. In speaking law in the name of the state, domestic courts owe their allegiance to the domestic constitutional order, which may or may not grant effect to international law norms. But even where courts are in a position to directly interpret and apply international law, any expectations of a particular judicial outcome can only arise where the applicable principles and rules of international law are sufficiently clear. It is one thing to expect domestic courts to interpret and apply legal rules in a particular way where there is consistent and well-established jurisprudence, preferably developed by international courts or tribunals. It is quite another to have such expectations in relation to rules that have not yet been subject to prior adjudication. In circumstances where the Contracting Parties themselves have held opposing views as to the meaning of Article 2 of the 1965 Agreement, it would be rather incongruous for Japanese companies to have developed legitimate expectations as to any particular way in which the provision should be interpreted and applied by Korean courts.

C. Non-compliance with the 1965 Agreement as a Judicial Expropriation?

Another treaty obligation susceptible to being invoked as a ground for challenging the impugned judgments is the prohibition of uncompensated expropriations under Article 10(2) of the Korea-Japan BIT. As most expropriation clauses in investment treaties, the latter invariably focuses on the act of taking itself, by prohibiting the parties to “expropriate or nationalise investments in its territory of investors of the other Contracting Party or take any measure tantamount to expropriation or nationalisation”. The provision thus seems perfectly open to be engaged through judicial measures. Indeed, in the practice of investment tribunals, there is now a growing jurisprudence involving expropriation claims predicated upon judicial conduct.⁷⁸

Whether judicial measures affecting Japanese companies can be presented as instances of direct expropriation, or as measures tantamount to expropriation, will depend on the circumstances of each individual case. Most of the proprietary interests susceptible to being affected by judicial decisions—such as rights to property in rem, shareholding rights in companies, contractual rights, rights granted under public law, or intellectual property rights—are each individually capable of constituting a protected asset under Article 1(2) of the Korea-Japan BIT. When the interest affected itself forms a protected investment, the claim could be brought as a direct judicial expropriation. A judicial measure such as a court-ordered liquidation of property seized from Japanese companies in satisfaction of the forced labour judgments will namely affect the very title to the proprietary interest. Where the legal interests affected instead will not separately qualify as a protected investment, an impugned judicial measure may still be challenged as an instance of indirect expropriation where the effects of the measures are such that they permanently deprive the relevant company of the use, benefit, management, or enjoyment of a substantial part of its investment. But this, as is known, imposes a demanding condition as a claimant company needs to be “radically deprived of the economical use and enjoyment of its investments, as if the rights related thereto … had ceased to exist”.⁷⁹

For a judicial expropriation claim to succeed, one will be required to demonstrate that the impugned judicial measure was unjustified. The mere fact that a judgment affects a legal right, which constitutes a protected asset, does not automatically translate that judgment into an act of expropriation. As aptly noted by the Tribunal in Garanti Koza v. Turkmenistan (2016), “[a] seizure of property by a court as the result of normal domestic legal process does not amount to an expropriation under international law unless there was an element of serious and fundamental impropriety about the legal process”.⁸⁰ Investment tribunals have by and large accepted that, in order for a judicial measure to amount to an unwarranted deprivation of the investor's proprietary rights, that measure itself had to be reproachable from the standpoint of international law.⁸¹ In order to determine whether a judicial decision went beyond the normal exercise of adjudicative functions, however, they often reverted to the very same standard of denial of justice. This is not unusual given that the latter still provides the most fundamental normative benchmark for determining whether the domestic judicial process presents deficiencies that are unacceptable from the standpoint of international law. But the approach does raise the question as to how a judicial expropriation claim then actually differs from a claim of denial of justice.

Then again, tribunals considered denial of justice to not be the only benchmark for assessing the propriety of judicial conduct in the context of an expropriation claim.⁸² In some cases, the failure on the part of the courts to respect obligations incumbent upon states under specific treaties was deemed sufficient a reason to treat a judicial measure as expropriatory in nature. Most notably, the Tribunal in Saipem v. Bangladesh (2009) found the Bangladeshi courts to have expropriated the Claimant's contractual rights under a local commercial award on the ground that the courts—by revoking the authority of the local arbitral tribunal and by subsequently annulling the award itself—failed to give effect to the Claimant's right under Article 2 of the 1958 New York Convention to have its arbitration agreement recognized.⁸³ The approach is not unsound since, in failing to give effect to a particular international obligation, courts may thereby also “deprive” investors of any substantive rights or benefits accruing to them as a result of such an obligation.

In following the Saipem precedent, the affected Japanese companies could attempt to mount a similar expropriation claim, by arguing that the impugned judgments wrongfully deprived them of their investments in failing to respect the rights or benefits to which they were entitled under Article 2 of the 1965 Agreement. Were one to accept that the latter confers upon the Japanese companies the right not to be exposed to claims for damages of the kind that have been at issue in the forced labour case, such an expropriation claim may then have some chance of success—provided, however, that the concrete judicial measures, as a matter of fact, also permanently deprived the relevant companies of a substantial part of their investment.⁸⁴

D. Violation of the 1965 Agreement as a Breach of the Umbrella Clause?

This brings us to consider two further treaty obligations that could provide a basis for challenging the impugned judgments: the umbrella clause and the effective means clause. Compared to the other grounds just discussed, the use of such clauses is possibly the most hypothetical one, since neither of them is actually available under the Korea-Japan BIT. Yet, as one can find them in other investment treaties entered into by Korea,⁸⁵ it is worth briefly considering their potential effects, for such clauses could be directly applicable to some investment structures, or else be capable of being invoked by relying on the Most Favoured Nation [MFN] treatment provision in Article 2(2) of the Korea-Japan BIT.⁸⁶

As between the two, the umbrella clause appears more likely to provide some advantage in vindicating any putative rights that the Japanese companies may have under the 1965 Agreement. Such clauses are namely designed to bring specific commitments that a host state has assumed vis-à-vis a foreign investor or its investment under the protective scope of the investment treaty, by turning their observance into a separate obligation under international law; and in practice there is nothing to suggest that such clauses could not be violated through the intermediary of the courts.⁸⁷

The type of commitments protected depends on the language of the concrete treaty provision. In the case of umbrella clauses available under other Korean BITs, this can vary from more narrowly formulated “any contractual obligation … entered into towards an investor”⁸⁸ or “any commitments … entered into with the investors”,⁸⁹ to the more open-ended “any obligation … entered into with regard to investments”⁹⁰ or “any other obligation … assumed with regard to investments”.⁹¹ In practice, it has been generally accepted that protection under umbrella clauses is not restricted to contractual commitments.⁹² Where their language is not qualified, such protection may extend to obligations unilaterally assumed by the host state, such as through law or regulation,⁹³ provided that such obligations are sufficiently specific and quantifiable.⁹⁴

Whether this protection may also extend to obligations assumed under an international treaty is less clear from current arbitral practice. The Tribunal in Noble Ventures v. Romania expressed hesitation about such a possibility: not only because states usually do not conclude special international agreements with reference to specific investments, but also because such agreements, if concluded, would already be subject to the general principle of pacta sunt servanda, arguably making an umbrella clause redundant.⁹⁵ As a matter of general proposition, however, the International Centre for Settlement of Investment Disputes [ICSID] Annulment Committee in CMS v. Argentina accepted the possibility that a treaty provision demanding respect of “any obligations [the host state] may have entered into with regard to investments” was capable of extending to consensual obligations arising “possibly under international law”.⁹⁶ Whilst examples of successful umbrella clause claims predicated on breaches of international law are presently lacking, at least as a matter of principle, there is no reason why the notion of “obligation” would necessarily have to be limited to obligations under contracts or host state law.

Hence, whether or not the Japanese companies could predicate an umbrella clause claim on alleged judicial violations of the 1965 Agreement would ultimately depend on the concrete language of the umbrella clause invoked, and on how one construes Korea's obligations under Article 2 of the 1965 Agreement. The Agreement being a treaty between Korea and Japan, one would have difficulty treating them as commitments entered into “with investors” (within the meaning of the narrower umbrella clauses available under Korean BITs), as the Japanese companies are not privy to the 1965 Agreement. On the other hand, one could arguably construe them as obligations “entered into” with Japan or else broadly “assumed” by Korea (in the sense of the language used in the broader umbrella clauses). But the question remains whether such obligations have been entered into or assumed “with regard to investments” (as required under those same clauses).⁹⁷ Since the 1965 Agreement predates the Korea-Japan BIT, it is debatable whether the stipulations made in Article 2 thereof were really commitments agreed to by Korea in order to encourage specific investments from Japanese companies.⁹⁸ Given that the potential beneficiaries of Korea's commitments were not solely the Japanese companies now affected, but also Japan and private nationals, it is even arguable whether the commitments concerned “investments” at all. Nonetheless, were one to take the requirement on its face—as one extending to just any obligation applicable to an investment covered by the Korea-Japan BIT—it is possible for umbrella clauses such as those available under Article 3(4) of the Korea-Netherlands BIT (2003), or Article 10(2) of the Korea-Portugal BIT (1995) to provide an additional means for challenging the impugned judicial measures.

E. Violation of the 1965 Agreement as a Breach of the Effective Means Clause?

Less clear in this context is the potential usefulness of treaty obligations demanding the maintenance of “effective means of asserting claims and enforcing rights”, such as the one found in Article 2(8) of the Korea-Kuwait BIT (2004). It is true that, in Chevron v. Ecuador (2010), this type of obligation has been interpreted as going beyond the general guarantee against denial of justice, by purportedly setting out an “an independent treaty standard” that was capable of engaging the responsibility of the state for “a failure of domestic courts to enforce rights ‘effectively’”.⁹⁹ It is also true that, whilst considered to be directed “at many of the same potential wrongs as denial of justice”, the obligation was found to impose a “potentially less-demanding test” as compared to the latter.¹⁰⁰ Yet, it is not immediately evident whether relying on such an obligation would prove of added value in the context of the forced labour litigation.

The standard of “effectiveness” prescribed by such obligations was taken to apply to “a variety of state conduct that has an effect on the ability of an investor to assert claims or enforce rights”.¹⁰¹ Yet, given that the effectiveness of a judicial system significantly depends on its expediency, the obligation has thus far mostly been engaged in the context of judicial delays—but these are not at issue in the present case. Of course, the effectiveness of a particular means may also depend on the substantive quality of adjudication. But this does not mean that the obligation is directly concerned with substantive outcomes produced by the respective means. The duty to provide effective means is one demanding the establishment of “a proper system of laws and institutions” that enables investors to assert claims or enforce rights.¹⁰² It is not an obligation guaranteeing such claims or rights—not in general,¹⁰³ let alone in the very circumstances where the validity or existence of the latter is contested. Hence, the fact that the Japanese companies may not have succeeded in enforcing their putative rights under the 1965 Agreement—because Korean courts found such rights not to exist under the Agreement—would not in itself mean that Korea had failed to provide the companies effective means of asserting claims and enforcing rights.

III. PURSUING THE DIRECT ROUTE: CHALLENGING ADVERSE JUDGMENTS AS VIOLATIONS OF THE 1965 AGREEMENT?

To complete the analysis, it is necessary to consider whether the companies affected could also assert violations of the 1965 Agreement as an independent cause of action. As already mentioned, not all investment treaties necessarily restrain the scope of actionable claims to those arising under the treaty itself. In some investment treaties, the host state's consent to arbitrate disputes with eligible investors is expressed broadly, extending to any or all disputes concerning, relating to, arising out of, or being in connection with an investment,¹⁰⁴ or simply to any dispute between an investor and a host state.¹⁰⁵ This prompts the question whether such broadly expressed consent entitles an arbitral tribunal established pursuant to such a treaty to take cognisance of claims directly founded upon the 1965 Agreement.¹⁰⁶

In arbitral practice, the view has generally been taken that such broadly formulated dispute settlement clauses provide a treaty tribunal with authority to hear claims grounded in other sources of obligations than those based on the treaty from which they derive their jurisdiction. Such a stance was in most cases taken with respect to claims or counter-claims grounded in contracts entered into between the investor and the host state.¹⁰⁷ In some cases, however, tribunals treated broadly formulated dispute settlement clauses as equally capable of accommodating independent causes of action rooted in other sources of international law, such as claims of denial of justice or expropriation claims grounded in customary international law,¹⁰⁸ or counter-claims based on human rights obligations grounded in a treaty or general international law.¹⁰⁹

But such an extension of investment tribunals’ authority over claims extraneous to the underlying investment treaty is arguably not without limits. As pointed out by a commentator, the admissibility of a claim grounded in another treaty will essentially depend on two factors: (1) any mandatory choice-of-forum provisions potentially demanding the settlement of disputes exclusively through the procedures provided for by that treaty, and (2) the nature of the obligations breached, which has to be such that it actually confers upon the investor a legal entitlement.¹¹⁰ It is not without reason, for example, that the Tribunal in Achmea decided that it did not have competence to rule on alleged breaches of EU law, despite the applicable investment treaty conferring it jurisdiction over “all disputes … concerning an investment”.¹¹¹ Conformity with EU law is a matter on which the European Court of Justice will have the final say.¹¹²

Whether the Japanese companies affected would hence be entitled to bring direct claims for violations of the 1965 Agreement before an investment treaty tribunal with jurisdiction over “any disputes” concerning an investment would then seemingly depend on whether the dispute settlement provisions under Article 3 of the 1965 Agreement are exclusive in nature, as well as on whether the stipulations in Article 2 of the 1965 Agreement are of such a legal nature that they create legal entitlements to the benefit of Japanese companies. As already noted above, the language of Article 2 is equally capable of accommodating the latter reading, whereas Article 3, though providing in mandatory terms that disputes “shall be referred for decision to an arbitration board composed of three arbitrators”, does not otherwise suggest that such a stipulation would operate to the exclusion of other dispute settlement mechanisms. On the face of it, the affected companies could thus also be in a position to pursue the direct route—provided, of course, they are able to invoke the protection of an investment treaty with a broadly formulated dispute settlement clause.

IV. REMEDIES AGAINST WRONGFUL JUDGMENTS

The question that finally needs to be considered concerns the remedies that an investment tribunal established pursuant to an applicable investment treaty could eventually provide were it to find that the Korean Supreme Court's judgments are not in conformity with the treaty's investment protection disciplines or otherwise in violation of the 1965 Agreement. The essential principle in that respect is, of course, the one enunciated in the Factory at Chorzów case, that a “breach of an engagement involves an obligation to make reparation in an adequate form” and that such reparation “must, as far as possible, wipe out all the consequences of the illegal act and re-establish the situation which would, in all probability, have existed if that act had not been committed”.¹¹³ The adequate form of reparation will depend on the concrete circumstances of each case and the precise nature and scope of the injury. That also means that the type of obligation violated will be determinative as to whether the impugned judgment would have to be vacated. The ICJ, for example, has drawn a distinction between situations where a judgment was defective as a result of the courts’ failure to respect obligations of a procedural nature, and those where a judgment was contrary to a substantive rule of international law. Only in the latter case has the Court considered the most appropriate form of reparation to consist of the impugned measure being vacated, so that it ceases to have effect.¹¹⁴

Yet, one would be hard pressed to find in the practice of international courts and tribunals decisions directly annulling or setting aside domestic judgments that had been found to be wrongful.¹¹⁵ International adjudicators have by and large avoided directly proclaiming the impugned domestic judicial decisions to be without effect. The ICJ, in particular, has been very careful in imposing measures that would be intended to directly interfere in states’ domestic legal orders, preferring instead to order the respondent states to take steps, by means of their own choosing, to ensure that the decisions or orders emanating from their own courts ceased to have effect, both in the domestic legal order, and transnationally.¹¹⁶ International investment tribunals adopted a similarly deferential approach. The Tribunal in the Chevron/Texaco v. Ecuador case (2018) considered that it lacked “the power to annul” the Ecuadorian Judgment it had found to amount to a denial of justice. Whilst noting that the Judgment existed as “a concrete fact under Ecuadorian law” and that it also had “a legal effect and resulting consequences under international law”, the Tribunal explained that “the remedy of annulment, as such, lies with the Respondent's internal law”, but that as “an international tribunal” it had “the power to order the Respondent to take steps to secure that result”.¹¹⁷ The Tribunal in ATA v. Jordan (2010) took a somewhat different approach. Whilst similarly not interfering with the original domestic judgment that had been found to amount to a violation of the applicable investment treaty, the Tribunal made an order that had the effect of reinstating the contractual right to arbitration that had been wrongfully extinguished by the impugned judgment.¹¹⁸

In terms of remedies, the advantage provided to the Japanese companies by investment treaty arbitration are therefore obvious. Not only could they count on compensation in the event that the damage suffered turns out to be the result of an internationally wrongful act occasioned by the improper conduct of Korean courts. Investment tribunals are also capable of ordering the Korean state to take all necessary measures to have the judgments adversely affecting their investments cease to have legal effects, which would be equivalent to having the impugned judgments quashed.

V. CONCLUSIONS

The availability of investment treaties has opened up an additional avenue for private actors to seek redress against domestic courts’ judgments considered adverse to their interests; an avenue that has proven to provide concrete and effective remedies. The purpose of investment arbitration may not have been to provide a supranational instance of appeal against decisions of host states’ courts. But it is certainly its proper task to determine whether the conduct of host state organs, including judicial ones, conform with the standards of treatment to which states have subscribed in their investment treaties, and by necessary implication, to correct wrongs suffered at the hands of those organs acting in contravention of those standards. As the present contribution has demonstrated, some of those standards of treatment may be capable of enabling Japanese companies to offset the adverse consequences of the lawsuits that were brought against them in Korean courts by the victims of forced labour. Whether or not such a development is desirable is not the intention of this contribution to answer. What is certain, however, is that Korea would be better off if the issue of forced labour is resolved with Japan directly; hopefully then in a way that would also benefit the actual victims of forced labour. Recent reports on the two governments discussing a plan to accommodate compliance with the 1965 Agreement and respect for the Korean courts’ rulings is a welcome development in that respect.¹¹⁹