1 Introduction

The topic of corruption is gaining increasing importance in the field of investor–state dispute settlement (ISDS), which has been facing intense scrutiny and criticism with regard to its role in international crime and corruption, whether perceived or otherwise.1 While allegations of corruption throw up difficult factual and legal issues, the emergent trend in ISDS is that corruption allegations are almost never determinative of the outcome, with very few cases thus far having made express findings of corruption that had an impact on the ultimate outcome of the case.2 However, corruption allegations are being alleged with increasing frequency and often with preclusive effect,3 with around 30 awards involving allegations of corruption being rendered since 2010,4 and some recent cases have even featured high level corruption scandals at the heart of the dispute.5 This makes it crucial for investors and host states alike to be familiar with the contours of corruption issues in ISDS. This chapter therefore seeks to provide a concise snapshot of the potential issues that both investors and host states may face when either raising or being confronted with an allegation of corruption.

2 The Meaning of Corruption

Before getting to the discussion proper, it is necessary to set out the meaning of ‘corruption’ and its most common instance, ‘bribery’. Corruption is derived from the Latin word corrumpere, meaning ‘to break’, and encompasses all situations where ‘agents and public officers break the confidence entrusted to them’.6 It is defined in the Oxford English Dictionary as the ‘perversion or destruction of integrity in the discharge of public duties by bribery or favour; the use or existence of corrupt practices, esp. in a state, public corporation, etc.’. ‘Bribe’ is defined as ‘to influence corruptly, by a reward or consideration, the action of (a person) to pervert the judgment or corrupt the conduct by a gift’.7 These definitions have been described as correctly emphasising ‘the essence of corruption in its legal sense’.8

International consensus on a broad definition of both public and private sector corruption can be found in Articles 15 (bribery of national public officials), 16 (bribery of foreign public officials and officials of public international organisations) and 21 (bribery of the private sector) of the UN Convention Against Corruption (UNCAC, open for signature in 2003). These definitions are materially similar to the corresponding provisions of major international and national anti-corruption regimes, such as the Council of Europe Criminal Law Convention on Corruption, the OECD Convention on Combating Bribery of Foreign Public Officials in International Business Transactions9 (OECD Convention, 1999) and the United States’ Foreign Corrupt Practices Act 1977 (FCPA).10 In short, in the context of public sector corruption (which is the relevant form of corruption in investor–state arbitration), there is consensus that: (i) corruption by a bribe payer is the act of intentionally promising, offering or giving to a public official, directly or indirectly, of an undue advantage for the official himself or herself or another person or entity, in order that the official act or refrain from acting in the exercise of his or her official duties;11 while (ii) corruption by a bribe recipient is the act of intentional solicitation or acceptance by a public official, directly or indirectly, of an undue advantage for the official himself or herself or another person or entity, in order that the official act or refrain from acting in the exercise of his or her official duties.12

Other forms of corruption include ‘trading in influence’, which involves a person having ‘real or supposed influence’ over public bodies or officials, and trading the ‘abuse’ of such influence (as opposed to the payment of bribes) in return for an ‘undue advantage’ from a person seeking this influence.13 Another form of corruption is extortion, which involves a situation in which a private party is compelled to pay an official to be treated fairly (as opposed to ‘better than fair’ treatment), because the official’s capacity to withhold a service or benefit otherwise required by law exceeds the private party’s capacity to sustain the loss of that service or benefit.14 The most common form of extortion is that of facilitation payments (aka ‘speed’ or ‘grease’ payments), which should be noted as not facing universal condemnation, in that such payments are condemned in some legal regimes,15 but not in others.16

3 Corruption in the Investor–State Relationship

Allegations of corruption in the investor–state relationship usually throw up difficult factual and legal issues that the tribunal must resolve. In general, where corruption allegations are afoot, three broad categories of issues will arise:

  1. (a)

    Evidentiary Issues: whether a tribunal may conduct its own investigations into prima facie suggestions of corruption, and what the applicable burden and standard of proof are.

  2. (b)

    Attribution of Responsibility: whether the corrupt act in question may be attributable to the party that allegedly committed it.

  3. (c)

    Legal Consequences: what legal consequences will flow from a finding of corruption, where corruption is raised either as a defence by the host state or as a ground for a claim by the investor, or where the corruption at hand is jointly attributable to both parties.

3.1 Evidentiary Issues

A party who seeks to prove corruption generally faces an uphill task, as corruption is ‘notoriously difficult to prove’.17 The most common form of corruption in investment treaty disputes tends to be bribery, which is perpetrated by parties who have gone out of their way to avoid identification or detection.18 There is typically little or no direct physical or documentary evidence of corruption,19 and any independent evidence necessary to corroborate a party’s allegations of corruption will have to come from the officials or politicians that have been bribed, which is highly unlikely in the face of criminal liability in their home countries.20 Further, the complainant must not only establish the alleged improper conduct, but also a causal link to the relevant investment in the sense that such improper conduct had contributed to obtaining (or not obtaining) a right or benefit related to the investment.21

These difficulties are exacerbated by the nature of the arbitral process, as arbitral tribunals do not have the same subpoena and enforcement powers of a national court to compel the production of evidence.22 This is especially acute where the corruption alleged involves a third-party intermediary who is not a party to the arbitration. The typical complainant is therefore compelled to rely mostly on the strength of its witnesses’ oral testimony, which may have little to recommend itself over the wrongdoers’ evidence, or on circumstantial evidence. The inordinate difficulties faced by complainants in proving corruption naturally raise questions about: (a) which parties should be charged with the burden of proving corruption; and (b) what is the proper evidentiary standard that must be applied in ascertaining whether there was any corruption.

Even where neither party advances allegations of corruption, there are also evidentiary issues that raise broader questions on the role of the arbitral tribunal in the fight against transnational corruption. One issue in particular is whether there ought to be some form of proactivity on the part of the arbitral tribunal where the evidence discloses a prima facie suggestion of corruption.

3.1.1 Investigations by the Tribunal

Parties in international arbitration are generally nervous of raising the issue of corruption, or leading evidence that the arbitral tribunal may reasonably construe as ‘red flags’ of corruption, due to the possible mutuality of wrongdoing in corruption. It may then be up to the arbitral tribunal to independently raise the issue of corruption when the signs are too obvious to ignore.

Arbitrators are incentivised, and may indeed be obliged, to initiate sua sponte investigations into prima facie suggestions of corruption, because turning a blind eye to corruption may lead to any eventual award being unenforceable as well as the eroding of the integrity of the institution of international arbitration.23 National courts have come to a similar conclusion by holding that an arbitral tribunal’s duty to investigate is a proactive one and arises not only where there are allegations of corruption in the parties’ dispute, but also where the evidence in the case indicates possible corruption.24 As corrupt dealings by one or both parties would often have a dispositive effect on the enforceability of the claims submitted to the arbitral tribunal (e.g., where the concession was obtained through bribery), an investigation into prima facie suggestions of corruption would generally be relevant to the resolution of the parties’ dispute and would therefore not violate the principle of ultra petita.25 However, this does not mean that an arbitral tribunal is entitled to launch into a completely independent investigation of its own;26 rather, when faced with a stench of corruption that is too pungent to ignore, the arbitral tribunal should make appropriate inquiries, and allow (or even direct) parties to define the parameters of such an inquiry and then present the case for and against a specific finding of corruption.27

The practice of arbitral tribunals has thus far been inconsistent. Some tribunals have preferred a more hands-off approach, preferring instead to allow the adversarial process to take its course absent some compelling indication of corruption.28 Others have taken a more inquisitorial stance, ranging from addressing corruption allegations that the parties have decided to not rely upon in their own pleadings,29 to issuing procedural orders propio motu for the purpose of obtaining additional information in relation to the suspected corruption at hand.30

It is however important to note that there are inherent limits to an arbitral tribunal’s power to initiate investigations, such as the inability to directly compel the giving of evidence.31 Arbitrators may prefer to instead find other, more indirect ways of dealing with the corruption factor. For example, where a particular argument rests on an assumption that corrupt activities were performed, the arbitrators may highlight that to the parties and suggest to them to amend their pleadings accordingly. This would then bring the issues of corruption squarely before the arbitral tribunal, obviating the need for a sua sponte investigation.32

3.1.2 Burden of Proof

It is a prevailing principle in the adjudication of international disputes that each party bears the burden of proving the facts on which it relies (actori incumbit probatio).33 Therefore, when corruption is pleaded as a defence in ISDS, the burden is on the host state to prove the alleged corruption. Correspondingly, when corrupt solicitation and/or extortion is pleaded by investors, the burden is on the investors to prove the alleged corrupt activities.34

The more controversial issue is whether it would be appropriate, in certain instances, to reverse the burden of proof (i.e., requiring a party to disprove its involvement in corrupt activities) upon a prima facie showing of corruption (e.g., through ‘red flag’ evidence).35 The main justifications for such a reversal of the burden of proof are: (i) the high difficulty in proving corruption in international arbitration due to its inherently clandestine and complex nature36 vis-à-vis the relative ease by which a truly innocent party can produce countervailing evidence;37 and (ii) the arbitral tribunal’s lack of the same subpoena and enforcement powers of a national court to compel the production of evidence.38

Such direct burden-shifting has some support in international commercial arbitration, albeit with much caution and the need for special circumstances.39 In ISDS, tribunals have recognised that burden shifting may be warranted in certain circumstances.40 In the corruption context, some tribunals have requested that the investor (and not the host state alleging corruption) prove that it obtained the investment legally,41 while others have applied the actori incumbit probatio principle flexibly42 or determined on the basis of the evidence before it whether corruption had been established with reasonable certainty without resorting to rules of burden of proof.43

However, such reversal of the burden of proof has also been met with considerable criticism on the basis of its incompatibility with principles of natural justice and due process, and the risk of a slippery slope leading to a similar reversal for other issues for which proof is difficult to obtain.44 The rule that a party must prove the facts on which it wishes to rely may be too important to be derogated from, as it is intrinsically tied to the integrity of the fact-finding process in international arbitration.45 The answer to the evidentiary problems in proving corruption may lie more in the realm of the quality of evidence required to prove corruption instead, to which we shall now turn.

3.1.3 Standard of Proof

The standard of proof is concerned with the threshold of evidence necessary to establish a certain fact, contention or proposition, and is assessed not just on the party who bears the burden of proof, but rather on the overall accumulated evidence put forward by one or both parties.46 The standard of proof in international arbitration is often assumed47 to be on a balance of probabilities (i.e., more likely than not) or its civil law counterpart of ‘intime conviction’ (‘inner conviction’).48

In cases dealing with corruption issues, however, there is a prevailing arbitral practice of subjecting complainants to a higher standard of proof that appears to approximate the ‘beyond reasonable doubt’ standard in criminal law.49 This practice appears to mirror the standard of proof utilised in national law where serious allegations of wrongdoing are involved in civil proceedings, such as bribery or fraud,50 though this has been doubted in recent national jurisprudence.51 In ISDS, the tension between the two approaches is brought to the fore by the fact that there has been no consistent approach with regard to the standard of proof for serious allegations of wrongdoing,52 with some tribunals even taking the view that such evidentiary issues are open questions under the lex causae of international law, thereby leaving tribunals with ‘relative freedom in determining the standard necessary to sustain a determination of corruption’.53 The application of too strict a standard of proof may result in certain risks, such as the award being subsequently set aside by a court or annulment tribunal applying a different standard.54

The better approach is to simply have one single standard of proof in ISDS, that of a balance of probabilities, considering that the arbitral tribunal is dealing with the consequences of corruption on a matter of civil liability, not criminal liability, and that a criminal standard of proof would in most cases be impossible to satisfy, which would then be exploited by the corrupt person to avoid liability.55 The latter point is especially relevant in disputes relating to ‘intermediary’ or ‘agency’ agreements (e.g., payment of ‘facilitation fees’) due to the fact that procurement of the necessary evidence will have to come from the officials or politicians whom the intermediary has bribed, which is unlikely to occur in the face of potential prosecution.56

The balance of probabilities standard, however, should be understood and applied in a nuanced fashion in conjunction with the quality of the evidence required to cross the threshold. This means taking into account the particular circumstances of each case, including the seriousness of the allegations of corruption and their legal consequences if proven,57 the inherent likelihood or unlikelihood of corruption in the specific circumstances of the case, and the intrinsic difficulty of proving corruption. Simply put, the more inherently unlikely a certain alleged fact is, such as seeing a lion as opposed to a dog in a park, or high-level corruption across multiple government agencies, the more cogent the evidence would need to be in order to satisfy the arbitral tribunal that the fact is indeed made out on a balance of probabilities.58 However, this will have to be balanced against the ‘intrinsically difficult nature’ of demonstrating the clandestine activities of the corrupt.59 Such a flexible understanding of the balance of probabilities approach will enable arbitral tribunals to better match the evidentiary process with the ingenuity of those that conceal corruption. It has been observed that, in practice, what matters more than the standard of proof in relation to allegations of corruption is the tribunal’s approach to the assessment of the evidence.60

The flexibility of the balance of probabilities standard, combined with the wide discretion given to arbitral tribunals to determine the admissibility, relevance, materiality and weight of the evidence adduced,61 therefore allow an arbitral tribunal to consider indirect or circumstantial evidence, as well as draw adverse inferences, in determining whether the allegations of corruption have been proven to its satisfaction. ISDS tribunals are not unfamiliar with the use of circumstantial evidence.62 Examples include excessively high consultation fees paid to an intermediary along with little to no proof of any consultation services being provided in return, or remuneration assessment being based on the value of the contract awarded to the principal as opposed to the quantity or quality of services rendered.63

The presence of ‘red flags’ (i.e., potential indicia of corruption), such as those set out in the US Department of Justice’s A Resource Guide to the US Foreign Corrupt Practices Act (2012), the Woolf Committee’s Report on BAE Systems (2008), TRACE International’s Due Diligence Guidebook (2010) and the OECD Convention on Combating Bribery of Foreign Public Officials in International Business Transactions (2011), may also prove useful.64 Where necessary, expert testimony assessing the various indicia of corruption may be adduced to assist the tribunal in better determining their evidential value.65

In exercising its broad discretion in evaluating evidence, a tribunal may also draw adverse inferences from: (i) an impugned party’s failure without sufficient justification to provide evidence requested by the tribunal;66 and (ii) a party’s failure or inability to adduce counter evidence where prima facie evidence of its involvement in corruption has been produced.67 However, as silence can often be motivated by innocent reasons, adverse inferences should be drawn only if it is the natural inference from the facts, and only the most cogent or compelling inferences ought to be given dispositive effect.

3.1.4 Indirect Evidence of Corruption from Other Proceedings

Corruption allegations are generally not made in a vacuum. There are rare cases where a party freely admits the facts that establish corruption during the arbitral proceedings.68 Oftentimes, however, evidence of the corrupt act in question may be indirect and derived from other legal or similar proceedings that are already afoot, such as national court decisions,69 evidence presented to national authorities (including national prosecutors and courts),70 or even a national anti-corruption commission’s report.71 Questions then arise as to the admissibility of such indirect evidence to prove the corruption allegations and what effect it should have.

The general practice amongst international arbitral tribunals is that the tribunals would, in the absence of special circumstances, refrain from excluding evidence on technical grounds of inadmissibility, and evaluate the relevance, credibility and weight of the evidence instead.72 This opens the door for tribunals to consider indirect evidence of corruption derived from other legal proceedings, including national ones. However, in ISDS, tribunals must ultimately make independent assessments of key facts, irrespective of the findings of national authorities. Holding otherwise would imply giving host states, who have consented to reciprocal protection of investments in accordance with law, the power to withdraw their consent unilaterally.73

International arbitral tribunals have in practice considered indirect evidence of corruption from other proceedings, but with varying degrees of importance. On one end of the spectrum are national anti-corruption commission reports and evidence from pending criminal investigations and legal proceedings, which have generally been held to be inconclusive as proof of any alleged corruption.74 On the other end of the spectrum is evidence adduced in completed legal proceedings (e.g., witness evidence) and national court judgments (e.g., criminal convictions). On the former, witness statements and oral testimony from other legal proceedings would generally amount to hearsay evidence,75 which, while generally admissible in arbitral proceedings, may be ascribed little weight due to the inability to test it by cross-examination. On the latter, the traditional common law rule is that a judgment in personam delivered in civil or criminal proceedings is generally inadmissible against a stranger (or against a party to those proceedings for a stranger) as evidence of the facts found or legal conclusions drawn in that judgment;76 but international tribunals nonetheless appear to be willing to consider national court judgments in determining the existence of the alleged corruption.77 In any event, such national court judgments are unlikely to have preclusive effect in ISDS proceedings, as the doctrine of res judicata generally requires commonality of parties and subject matter in the earlier and current proceedings, and ISDS proceedings are ordinarily concerned with issues encompassed by the investment treaty (or other arbitration agreement), and not those that previously arose in the national court litigation.78

It is worth noting that the absence of criminal investigation or prosecution may equally be taken into account by the arbitral tribunal. For example, the host state’s failure to bring any prosecution or investigation against any of the individuals allegedly involved in the corrupt act, along with an inability to provide a convincing explanation of its efforts in relation thereto, may negatively impact the credibility of the host state’s allegations on corruption.79

3.2 Attribution of Responsibility

Where the corruption at hand is alleged to have been committed by the investor at hand, the attribution of responsibility to the investor is generally not an issue and is often a mere question of fact. More problematic is when the corruption at hand is alleged to have been committed by the host state, at least in part, through an individual or entity that it is attributable to, since host states can only act through individuals and entities. This section will therefore deal with the issue of attribution of responsibility under international law rules.

So far, very few investment cases have engaged in a discussion on state responsibility for corruption in any significant way. In the context of corruption solely on the part of the state, it has been recognised that: (i) the corrupt solicitation of a bribe (i.e., extortion) by a state agency would amount to a violation of the obligation of fair and equitable treatment owed to the investor pursuant to the BIT, as well as a violation of intentional public policy; and (ii) where the host state exercises its discretion on the basis of corruption, that would result in a ‘fundamental breach of transparency and legitimate expectations’.80 A rare example can be found in Chevron v. Ecuador, where the corrupt issuance of a judgment by the host state’s judiciary against the investor was found to be ‘cloaked with governmental authority’ and therefore attributable to the host state.81

In the context of the invoking of corruption as a defence by the host state, it has also been recognised that, even if an investor may have violated important national laws when making its investment, the host state may, at least in principle, be estopped from invoking that illegality had it been aware of it but nonetheless proceeded in tacit approval, or at least acquiesced, to it.82 There is also some suggestion that that the corruption of public officials or their intermediaries can be attributed to host states for purposes of allocating the costs of arbitration.83

3.2.1 Attribution for State-Only Corruption

The primary legal regime for state responsibility in international law is the United Nations International Law Commission’s Draft Articles on Responsibility of States for Internationally Wrongful Acts (ILC Articles),84 which were adopted without a vote by the United Nations General Assembly in August 2001. The ILC Articles set out the principles that are to apply to all manner of internationally wrongful acts, without setting down any specific rules in substantive areas of international law. The ILC Articles have been treated by international courts and tribunals as a functional equivalent of the customary international law on state responsibility.85 While the ILC Articles were drafted with inter-state relations as the basic archetype, they arguably apply in the context of investment arbitration as well.86

Under the ILC Articles, a state is ‘internationally responsible’ for every one of its ‘internationally wrongful’ acts.87 There is an internationally wrongful act of a state when conduct consisting of an action or omission: (i) is attributable to the state under international law; and (ii) constitutes a breach of an international obligation of the state.88 Whether an act of a state is ‘internationally wrongful’ is an issue governed by international law, and is unaffected by how national laws may characterise the act.89 In the context of ISDS, the internationally wrongful act would be a breach of the host state’s obligations under the relevant investment treaty.

As a state can only act through its agents, one of the most important questions in invoking state responsibility is whose conduct is attributable to the state. With regard to state organs, the conduct of any state organ is to be considered an act of that state under international law.90 With regard to non-state organs, only certain conduct is attributable to the state.91 For example, for persons or entities exercising elements of governmental authority, their conduct is to be considered an act of the state under international law provided that the person or entity is acting in that capacity in the particular instance,92 while for a person or a group of persons, his or their conduct is to be considered an act of state under international only if he or they are in fact acting on the instructions of, or under the direction or control of, that state in carrying out the conduct.93

Generally, the fact that the conduct of the state organ or person or entity empowered to exercise elements of governmental authority in question exceeds the authority or contravenes instructions does not affect its attribution to the state under the ILC Articles, so long as that person or entity was acting ‘in that capacity’.94 The commentary to the ILC Articles stated that one form of ultra vires conduct covered by Article 7 would be for a state official to ‘accept a bribe to perform some act or conclude some transaction’.95 Therefore, in situations where public officers have solicited or extorted bribes while acting with apparent authority, such corrupt acts should, if proven, engage the responsibility of the host state.96 However, even where the corrupt conduct is attributable to the host state, there are several doctrines in international law that may operate to preclude the investor from invoking the host state’s responsibility, such as the doctrines of consent, waiver, acquiescence and estoppel.

Where the investor has participated in the corruption (e.g., by offering and paying a bribe to the public official involved), in full knowledge that the public official is acting illegally and with private enrichment in mind, the attribution rules under the ILC Articles would arguably not apply.97 There are various arguments for this result under the ILC Articles, such as the investor’s participation amounting to its consent to the host state’s otherwise internationally wrongful conduct,98 or that the investor would have known that the public official was not acting ‘in that capacity’ and could not therefore have been engaging in an act of state vis-à-vis the investor.99 In such scenarios, the investor may need to look to other applicable rules of attribution in order to preclude the host state from invoking the corruption defence (discussed further below).

If attribution of the allegedly corrupt act to the host state is established, the other half of the equation is in establishing that the relevant act amounts to a breach of an international obligation of the state, which would include an obligation under the relevant BIT or other treaty. These obligations would include the obligations of fair and equitable treatment, providing full protection and security, refraining from arbitrary/discriminatory action, and refraining from uncompensated takings without public purpose (elaborated below).

3.2.2 Attribution for State-Complicit Corruption

More difficult is the question of attribution of state responsibility for corruption committed by both the host state and the investor. This will generally arise only where the state attempts to run a defence against the jurisdiction of the tribunal and/or the merits of the claim (elaborated further below), and the investor seeks to argue for the preclusion of the host state’s ability to run such a defence on account of the state’s complicity in the corruption in question.

The first difficulty is that investment tribunals often adopt a ‘zero-tolerance’ policy towards corruption-tainted investments. Thus, where the investment has been found to be tainted by corruption as a fact, that results in a complete defence against the investor’s claim, even where a public official of the host state had been complicit in the corruption at hand, such as by soliciting or extorting the bribe in question.100 This has been said to result in an ‘attribution asymmetry’, where a public official’s actions in soliciting or extorting bribes from foreign investors is not attributable to the host state once that solicitation meets acceptance by the investor (e.g., by paying the bribe).101 This may result in a perverse incentive for host states to strategically engage in or tolerate corruption in order to deploy it as a shield in the event of future ISDS proceedings.

The second difficulty is that the attribution rules under the ILC Articles would arguably not apply in determining the question of whether the host state ought to be precluded from asserting the corruption defence. This is because the investor here is not raising the corrupt conduct in question to ground its claim against an ‘internationally wrongful act’, but rather to argue against the raising of a positive defence by the host state on the account of the host state’s own unilateral conduct.102 It has been argued that the ambit of the rules of attribution under the ILC Articles are limited to establishing that there is an act of state for the purposes of international responsibility.103 On this view, they do not extend to other purposes for which it may be necessary to define the state or its government, such as other international law processes by which particular organs are authorised to enter into commitments on behalf of the state, which depend not on the rules of state responsibility, but rather the international law rules relating to the expression of the will of the state. Even if the ILC Articles do apply, attribution under them would arguably not be possible once the investor has participated in the corruption in question (as discussed above). It may therefore be necessary to look to other rules of attribution that would support an argument for the preclusion of the raising of the corruption defence.

The main international law doctrines that the investor would be seeking to invoke to argue for the preclusion of the corruption defence would be: (i) recognition; (ii) acquiescence; and (iii) estoppel (elaborated further below), all of which concern the unilateral conduct of the state. In the event that the ILC Articles are inapplicable, the relevant attribution rules may instead be found in the ILC’s ‘Guiding Principles applicable to unilateral declarations of States capable of creating legal obligations’ (ILC Guiding Principles), which were adopted by the ILC following the ILC’s and Special Rapporteur Victor Rodriguez Cedeño’s detailed study of the subject of unilateral state conduct.

Under Principle 1 of the ILC Guiding Principles, it is stated that ‘Declarations publicly made and manifesting the will to be bound may have the effect of creating legal obligations’ and that when the conditions for this are met, ‘the binding character of such declarations is based on good faith; States concerned may then take them into consideration and rely on them; such States are entitled to require that such obligations be respected’, though both the Preamble of the ILC Guiding Principles and Cedeño clarified that the ILC Guiding Principles are applicable not just to positive forms of unilateral state conduct, such as recognition, but also to negative forms of unilateral state conduct, which arguably include inaction in circumstances which may amount to acquiescence and/or estoppel.104 Principle 6 provides that unilateral state conduct may be addressed not just to other states or the international community as a whole, but also ‘to other entities’, which should, in the investment treaty context, include investors who have been conferred the right to bring international law claims under the investment treaties.

Regarding attribution, the ILC Guiding Principles are narrower in the scope of attributable conduct than the ILC Articles. Principle 4 of the ILC Guiding Principles states that unilateral declarations ‘bind the State internationally only if it is made by an authority vested with the power to do so’, with heads of state, heads of government and ministers for foreign affairs being competent to formulate such declarations ‘by virtue of their functions’, and other persons representing the state in specified areas being also authorised to bind the state through their declarations ‘in areas falling within their competence’. Arguably, however, even statements made by officials that are ultra vires or illegal under municipal law can be attributed to and bind the state as unilateral declarations of intent, so long as they are made in furtherance of the duties assigned to the official by the state,105 and even if the addressee of the unilateral declaration knows that the official giving the declaration is in fact not competent under his municipal law to make such declaration or is otherwise acting illegally in doing so. While a state can invoke the violation of its domestic law as invalidating its unilateral declaration, similar to how a state’s consent to a treaty may be invalidated under Article 46 of the 1969 Vienna Convention on the Law of Treaties, it is clear that such violation must be ‘objectively evident’ and be concerned with an ‘internal law of fundamental importance’, which relates only to the procedural provisions of internal constitutional law.106 This will not be the case in every instance of corruption committed between the investor and the host state, as the corrupt act will usually concern a violation of substantive anti-corruption policy and rule of law considerations.

Under the ILC Guiding Principles, attributable conduct would include the solicitation and extortion of bribes by high-ranking government officials in return for the procurement of the investment, as in the case of World Duty Free v. Kenya where the Kenyan president (and other high ranking officials) had solicited and received bribes from the investor in return for the investment contract.107 It may also include other types of conduct, such as a deliberate choice to not prosecute or otherwise punish the corrupt government officials in question, though it will be generally difficult to establish such conduct to be a clear statement of acquiescence in corruption (elaborated below).

3.3 Legal Consequences

The legal consequences of corruption allegations in ISDS are affected by a multitude of factors, most especially who pleaded the corruption and when the corruption allegedly occurred. In essence, corruption is usually used a shield by the host state (to defend against an investor’s claim), while investors generally use it as a sword (to ground a claim for a breach of the investment treaty or agreement). More complicated is the situation where the host state attempts to use corruption as a shield and the investor seeks to preclude the host state from doing so on the account of the host state’s complicity in the corruption at hand. This section provides a brief sketch of the general legal consequences of a finding of corruption, as well as the emergent trends of modern ISDS practice, thereby setting the stage for more detailed case studies in the ensuing country reports in this volume.

As a preliminary note, it should be observed that issues in ISDS are generally governed by public international law (unless the arbitration stems from an investment contract).108 Thus, unlike commercial arbitration, there is generally no need to delve into a choice of law analysis to determine the applicable law for the legal consequences flowing from corrupt conduct, though national law may sometimes come into play where the protection over the investment is subject to compliance with the host state’s laws (elaborated below).

3.3.1 Investor Corruption

Corruption is raised mostly by host states as a complete defence to investors’ claims, usually in one of three ways: (i) denial of jurisdiction; (ii) denial of admissibility; or (iii) invalidation of the investment agreement. Other consequences may come in the form of: (i) providing the host state with a defence on the merits of the case; (ii) being a relevant factor in the assessment of damages and apportionment of costs; and (iii) various procedural and evidential consequences.

Denial of Jurisdiction

The jurisdiction of an investment treaty tribunal generally depends on whether the claimant-investor satisfies four necessary jurisdictional requirements that establish the existence of adjudicative power: (i) ratione voluntaris (whether there was unqualified consent to arbitrate the claim in question); (ii) ratione personae (whether the claimant is a covered investor under the treaty); (iii) ratione materiae (whether the subject matter of the claim is within the scope of the treaty, i.e., whether there is a covered investment); and (iv) ratione temporis (whether the treaty was in force when the dispute arose).109

Where corruption is raised by host states as a jurisdictional objection, it is most commonly for the purpose of contesting the tribunal’s jurisdiction ratione materiae on the basis that the investment was corruptly made and therefore not covered by the bilateral investment treaty (BIT) or other investment treaty. BITs commonly contain provisions providing that a protected investment is one that is made ‘in accordance with the law’ (i.e., a ‘legality clause’), which excludes from the treaty’s protection coverage any and all investments that violate the host state’s laws.110 Some recent BITs even have ‘legality plus’ clauses, which accompany the legality clause by explicitly excluding investments obtained through corruption from the treaty’s scope of protection.111 As this is a jurisdictional requirement, claims concerning an investment that was made in violation of a legality clause face dismissal by arbitral tribunals for want of jurisdiction ratione materiae,112 though this generally does not extend to trivial breaches of local law.113 As the laws of most host states criminalise corruption, where an investment is procured by corruption—in particular bribery—such an investment will most likely violate the host state’s law and be denied protection under a BIT with a legality clause.114 Performance corruption,115 on the other hand, is unlikely to lead to jurisdictional issues and may instead be addressed during the admissibility, merits and/or damages phases of the arbitration.116 In cases where it is unclear whether the corruption at hand was foundational or performance-related in nature, the tribunal may, as a case management decision, opt to hear the corruption allegations in one set of proceedings (i.e., without bifurcating the proceedings to hear jurisdictional issues separately), so as to allow a full investigation into the facts of the matter.117

The most prominent example of denial of jurisdiction in the corruption context is Metal-Tech v. Uzbekistan.118 The tribunal considered that Uzbekistan’s consent to arbitration was limited under Article 8(1) of the 1994 Israel–Uzbekistan BIT to only those disputes concerning ‘lawfully implemented investments’, defined under Article 1(1) of the BIT as ‘any kind of assets, implemented in accordance with the laws and regulations of the Contracting Party in whose territory the investment is made’. Due to the tribunal’s finding that the investor had engaged in ‘sham’ consultancy agreements that were designed to facilitate bribery, in contravention of the anti-corruption laws of Uzbekistan, the tribunal found that the investor was not compliant with Article 1(1) of the BIT and its investment did not fall within Article 8(1) of the BIT. This rendered the tribunal without subject matter jurisdiction over the dispute.119

Where there is no legality clause in the investment treaty, the position is far less clear. Some tribunals have taken the position that all investment treaties contain an implicit legality clause.120 Yet this has been criticised on several fronts, including the lack of a clear basis for such an implication.121 Other tribunals have taken the view that where there is no legality clause, corruption allegations are best addressed as questions of admissibility (discussed further below).122 Yet other tribunals may deal with an apparently jurisdictional issue together with the merits, so that a broader investigation into the history of the investment may be conducted for the purpose of obtaining positive evidence of the alleged corruption.123 In such scenarios, corruption allegations will no longer serve as a jurisdictional ‘trump card’ in the hands of the host state.

The host state may also raise corruption to contest the tribunal’s jurisdiction ratione personae, by seeking to establish that the investor does not have the requisite nationality to qualify as a foreign investor. In Siag v. Egypt,124 a case under the 1989 Italy–Egypt BIT, the claimants alleged that Egypt had expropriated their investment as principal shareholders of a hotel management company that had purchased from the government a large parcel of ocean-front land on the Gulf of Aqaba on the Red Sea in order to develop a tourist resort. Egypt objected to the tribunal’s jurisdiction partly on the basis that one of the claimants (previously holding Egyptian nationality) had corruptly acquired his purported Lebanese nationality via bribery and should therefore be treated as never having lost his Egyptian nationality, which would have rendered the tribunal without jurisdiction rationae personae over the dispute. The tribunal eventually dismissed the jurisdictional objection on the ground of insufficient evidence.

A host state may also raise corruption to contest the tribunal’s jurisdiction ratione temporis, by seeking to establish that the dispute had arisen before the relevant treaty had entered into force, for instance by impugning the integrity of a local judgment that purported to end an earlier dispute at a time prior to the treaty having come into force. In Lucchetti v. Peru, a case under the 2001 Peru–Chile BIT, the Chilean investor (the majority shareholder of a Peruvian company) constructed a plant for the manufacture and sale of pasta on property that it owned in the municipal district of Chorrillos, Lima. In 1997, the Council of the Municipality of Lima and related parties issued a series of decrees that declared the investor’s construction licence as null and void. In response, the investor instituted Peruvian legal proceedings against the entities that issued the decrees, which resulted in a series of Peruvian court judgments that allowed the investor to proceed with construction and operation of the pasta plant. However, in August 2001, the Council of the Municipality of Lima again revoked the investor’s operating licence through a decree that expressly noted that the judicial decisions rendered in favour of the investor were fraudulently and corruptly obtained, and the investor’s plant was consequently forced to close. Subsequently, the Peru–Chile BIT came into force on 3 August 2001. The investor filed for ICSID arbitration under the Peru–Chile BIT, but Peru contested the tribunal’s jurisdiction ratione temporis by arguing that its dispute with the investor had already arisen at the time the BIT came into force. The investor sought to counter this by pointing to the Peruvian judgments made in its favour, which were final and thus res judicata, thereby effectively terminating the initial dispute between the investor and the Municipality of Lima. However, Peru responded by arguing inter alia that the Peruvian judgments were obtained through corrupt conduct and therefore could not be deemed to have ended the dispute between the parties that began in 1997. The tribunal eventually held that it did not have jurisdiction on other grounds without dealing with the corruption allegations, but opined that if the corruption had been proved, it would have provided an established fact that would have substantiated dismissal of the case on ratione temporis grounds.125

Where corruption is proven, it is notoriously difficult for the investor to bring a successful counter-defence to prevent the dismissal of jurisdiction, especially in the context of a legality clause in the investment treaty in question. Most investors will seek to argue that the host state is precluded from relying on the corruption defence on the grounds of estoppel or acquiescence due to its own complicity. Others may seek to argue that the issue should be treated as a question of admissibility rather than jurisdiction, which would allow the tribunal to engage in a balancing act between the investor’s wrongdoing against the state’s misconduct.

Denial of Admissibility

The admissibility of a claim is a question on whether the tribunal should rule that the claim ought not to be heard by the tribunal (or at least not yet),126 and arises only after the tribunal has established the existence of its jurisdiction.127 Under national law, the equitable maxims ex turpi causa non oritur actio and nemo auditur turpitudinem suam allegans (an unlawful or morally reprehensible act cannot serve as the basis of an action in law), which are expressions of the ‘clean hands doctrine’,128 procedurally bar a claimant’s claims due to its illegal or improper conduct in relation to those claims. While there is no express reference to the admissibility or preclusion of claims under the ICSID Convention, the ICSID Arbitration Rules and the UNCITRAL Arbitration Rules, the clean hands doctrine is recognised by some tribunals as a rule of international law (though its precise contours are unclear).129 It also has been argued that ISDS tribunals, as creatures of public international law, ‘should be viewed as having inherent or incidental jurisdiction to find that claims are inadmissible for abuses of process or other serious forms of misconduct’.130

Where corruption allegations are treated as questions of admissibility, the jurisdictional trump card no longer exists. Both host states and investors are then free to raise general principles of law in their defence, which can lead to a ‘battle of principles’.131 The host state, on one hand, is likely to raise principles such as good faith,132 the clean hands doctrine133 and international public policy.134 The investor, on the other hand, is likely to respond with principles such as recognition, estoppel and acquiescence (discussed further below). Where such principles are pitted against each other, the tribunal will need to weigh the wrongdoings of all parties and perform a balancing exercise in deciding whether the investor’s claims are admissible. Even where the claims are admissible, the investor may still have to grapple with the consequences of its own participation in the corruption in the merits and/or damages phases of the arbitration (discussed further below).

Invalidation of Investment Agreement

In cases where the arbitration is based on a contract rather than an investment treaty, host states may also raise corruption as a defence by seeking the invalidation of the investment agreement on the ground of illegality. In World Duty Free v. Kenya, a contract-based ICSID arbitration, the tribunal held that the host state was legally entitled to avoid the entire investment contract on the basis that the upholding of claims based on contracts obtained through corruption was contrary to international or transnational public policy as well as the applicable laws of the contract (English and Kenyan law). The tribunal rejected the investor’s argument that the tribunal should undertake a discretionary balancing exercise in which the investor’s misconduct was weighed against that of Kenya’s, noting that while the tribunal was ‘sympathetic’ to the investor’s argument, the House of Lords in Tinsley v. Milligan had overruled such a discretionary test.135 In Niko v. B&P, another contract-based ICSID arbitration in which corruption allegations impugning the procurement of the contract were also made, the tribunal clarified that, as a general principle of public international law, contracts obtained by corruption are voidable at the option of the host state, but contracts of corruption are void ab initio.136

However, there might remain some room for some form of recourse for the investor even if the investment agreement had been tainted by corruption, as a non-contractual claim in unjust enrichment may nonetheless be available in situations of host-state complicit corruption (discussed further below). The World Duty Free v. Kenya tribunal had notably left open the possibility of some form of restitution even following the avoidance of the contract, though not through the return of the bribe to the investor.137 It has been argued that in the light of the decision of Patel v. Mirza,138 the World Duty Free v. Kenya tribunal’s reasoning might have taken a different direction if it were decided today, though the conclusion might not be different.139

Failure of Investor’s Claims on the Merits

Assuming that the investor’s claim is not defeated either on grounds of lack of jurisdiction or inadmissibility, the host state may argue that the claim should fail on the merits, because the regulatory action interfering with the investment that the investor is complaining of as violating investment protection standards can be justified as a response to illegal conduct by the investor. Investment arbitration tribunals have held that a host state may be justified in revoking the investor’s investment if it was done as a response to the investor’s illegal conduct, thereby absolving the host state from any liability stemming from an alleged violation of investment protection standards.140 In the corruption context, a host state that has been the victim of investor corruption can similarly argue that it is not liable for a breach of investment protection standards because its actions in revoking the investment were justified as a response to the investor’s corrupt conduct.

Other Consequences

Corruption on the part of the investor, like other instances of investor misconduct, may potentially have other legal consequences, such as: (i) providing the state with a defence on the merits of the case, including one based on a lack of due diligence, negligence or wilful blindness on the part of the investor with regard to signs of crime or misconduct;141 (ii) serving as a relevant factor in the assessment of damages, for instance finding that the investor was partly responsible for the damages in question due to its misconduct, thereby lowering the amount of damages to be awarded;142 and (iii) serving as a factor in the apportionment of costs.143

In terms of procedural issues, where corruption allegations have been made by the host state, the tribunal may choose to bifurcate proceedings in order to first resolve the corruption allegations, which may have a dispositive effect.144 In terms of evidentiary issues, corruption allegations may result in the tribunal considering indirect or circumstantial evidence or the drawing of adverse inferences (see above).

3.3.2 Host State Corruption

Where the investor raises corruption, it is often in the form of an allegation of attempted extortion or solicitation of a bribe by the host state’s public officials, in violation of an investment treaty.145 This is so because the investor is implicated in the corruption once it pays the bribe, which would lead to the above-mentioned legal consequences for the investor. Further, the implication of the investor would arguably bar it from invoking the attribution rules of the law on state responsibility (see above). The ways in which the investment treaty may be violated by the host state’s corrupt act are discussed below.

Obligation of Fair and Equitable Treatment

Where the host state attempts to solicit or extort a bribe from an investor, or engage in other types of corrupt conduct, that may result in a breach the host state’s obligation of fair and equitable treatment. Where the BIT contains a ‘fair and equitable treatment’ clause,146 which has the primary aim of promoting a stable and predictable investment environment in a host state,147 the host state may, depending on the specific wording of the clause, have certain obligations such as: (i) transparency and treatment in accordance with the investor’s legitimate expectations; (ii) compliance with contractual obligations; (iii) procedural propriety and due process; (iv) good faith; and (v) freedom from coercion and harassment.148 Such obligations would naturally be breached where a person or entity attributable to the host state were to solicit, demand or extort a bribe from an investor, or were to threaten to impose disadvantageous treatment against the investor in the event a bribe is not paid.

In EDF v. Romania, the investor argued that an alleged demand for a USD2.5 million bribe by the Chief of Cabinet to the Prime Minister of Romania, coupled with the state’s subsequent refusal to extend an investment contract, amounted to a violation of the fair and equitable treatment clause in the relevant BIT. Although the tribunal held that there was insufficient evidence to prove such allegations, the tribunal stated that a request for a bribe by a state agency is a violation of the fair and equitable treatment obligation owed to the investor pursuant to the BIT, as well as a violation of international public policy, and that the exercise of a state’s discretion on the basis of corruption was a ‘fundamental breach of transparency and legitimate expectations’.149

In Rumeli v. Kazakhstan, the investor argued that systemic corruption in the Kazakh judiciary and the solicitation of a bribe by a Kazakh judge in return for preventing the seizure of the investor’s investment had resulted in violations of the 1992 Turkey–Kazakhstan BIT. While the tribunal dismissed the corruption allegation on the ground of lack of evidence, it seemed to accept, or at least did not reject, the notion that the fair and equitable treatment obligation could be breached by the host state’s judiciary rendering judgments against the investor due to corruption.150 Such an argument was accepted in Chevron v. Ecuador. The tribunal held that a USD9.5 billion judgment rendered by an Ecuadorian judge against the investor was procured through fraud, bribery and corruption (and which was left unremedied by the host state’s appellate, cassation and constitutional courts), and thereby constituted, among other things, a breach of the fair and equitable treatment clause under the US–Ecuador BIT.151

Obligation to Provide Full Protection and Security

Corrupt conduct on the part of the host state may also violate its obligation to provide full protection and security. Most BITs contain a clause stipulating that the host state is obliged to grant full protection and security in its territory for investors and their assets. This is commonly understood to impose an obligation upon the host state to actively protect the investment from adverse actions by the host state itself, by its authorities or by third parties,152 which is an obligation of due diligence,153 and has been recognised to extend to providing protection from physical violence against the assets and individuals connected with an investment as well as the protection of investors’ commercial and legal rights.154 It has been argued that a host state might breach the obligation to provide full protection and security due to a lack of appropriate due diligence that resulted in the investor being a victim of corruption.155

Obligation to Refrain from Arbitrary or Discriminatory Action

By taking or threatening to take adverse action against a person who refuses to pay a bribe, a host state may also violate its obligation under most BITs to refrain from ‘arbitrary’ or ‘discriminatory’ action. ‘Arbitrary’ actions have been explained by international tribunals to mean those that fly in the face of the rule of law, that is, where there is a ‘wilful disregard of due process of law, an act which shocks, or at least surprises, a sense of juridical propriety’.156 A ‘discriminatory’ act is one that results in the treatment of an investor that is different to that accorded to other investors in a similar or comparable situation.157

A discriminatory act often involves a breach of the host state’s national treatment obligations. Most BITs contain a national treatment clause in which the host state is obliged to accord to foreign investors (and/or covered investments) treatment that is no less favourable than it accords, in like circumstances, to its own investors (and/or to investments in the territory of its own investors), with respect to certain aspects of investments (e.g., establishment, expansion, management, disposition).158 Such an obligation may be breached by a host state where it had sought to induce certain business outcomes for the purpose of corruptly favouring specific domestic businessmen.159

Obligation to Refrain from Uncompensated Takings without Public Purpose

The extortion of a bribe from an investor by a host state may result in it violating its obligation to refrain from uncompensated takings without a public purpose. Most BITs contain an expropriation clause that provides that the host state cannot expropriate a covered investment either directly or indirectly except: (a) for a public purpose; (b) in a non-discriminatory manner; (c) on payment of prompt, adequate and effective compensation; and (d) in accordance with due process of law.160

3.3.3 Complicit Corruption by the Host State

Investors who are faced with host states asserting corruption committed by both sides as a defence against the investor’s claims will in turn seek to argue that the host state is precluded from relying on such a defence. This may be maintained on the basis of the host state’s own complicity in the corruption at hand as a legal consequence of the host state’s own unilateral conduct, most commonly under the doctrines of recognition, acquiescence and estoppel,161 which are considered to be part of the corpus of ‘general principles of law recognised by civilised nations’ under Article 38(1)(c) of the 1945 Statute of the International Court of Justice (ICJ Statute).162All of these three doctrines have the same effect of precluding a state from contradicting or objecting to a given factual or legal situation that it had earlier accepted as legitimate.163

Recognition

Recognition has been defined as a unilateral expression of will by a state ‘acknowledging the existence of a de facto or de jure situation or the legality of a legal claim, with the intention of producing specific legal effects, and in particular accepting its opposability as from that time or from the time indicated in the declaration itself’,164 so that the recognising state is obliged to act in a manner that is consistent with its affirmation in relation to the addressee of the act. For binding recognition to arise, there must be: (i) an act or declaration evincing recognition that is made publicly and manifests the will of the state to be bound; and (ii) knowledge or cognisance on the part of the addressee of the act and its content.165 Where the intended purpose of recognition is to regard as legitimate (or to permit) an illegal or otherwise invalid act, one of the recognition’s potential effects is to deprive the recognising state of the right to subsequently argue the invalidity and/or illegality of the recognised acts.166

In the investment treaty context, recognition would most likely be raised in a situation where high-ranking government officials of the host state had engaged in bribe solicitation and extortion in return for the right to invest in the host state. For example, in World Duty Free v. Kenya, the Kenyan president (and other Kenyan officials) were found to have solicited and received bribes from the investor as ‘payment for doing business with the Government of Kenya’167 during a time when it was widely acknowledged that the Kenyan government was highly corrupt.168 Where the host state could in the circumstances of the case be fairly described as a kleptocracy, conduct such as solicitation and extortion from high-ranking government officials could possibly be argued as representing the host state’s de facto will to recognise a corruptly procured investment as being valid and entitled to the protections under the relevant investment treaty, notwithstanding their illegality under the host state’s municipal laws.169

Acquiescence

Acquiescence is similar to recognition, except that it is derived from silence rather than affirmative action or active conduct. Acquiescence has been said to be equivalent to ‘tacit recognition manifested by unilateral conduct which the other party may interpret as consent’.170 The ‘silence’ here usually involves the inaction of a state, including unreasonable delay,171 that leads to the loss of a right or claim if, under the circumstances, the state could reasonably be expected to act in a certain manner.172 Naturally, such circumstances will arise only where the acquiescing party had knowledge of the facts against which it refrained from making a protest against the acts of corruption. On one view, the acquiescence argument is far more persuasive in the context of investment treaties without explicit legality clauses, due to the lack of a jurisdictional trump card in the hands of the host state (discussed above).173

In the context of corruption in ISDS, acquiescence would most likely be raised in a situation where the host state’s public officials were made aware of the corruption of its other public officials but took no effective steps to prosecute or otherwise punish those officials. That is arguably something that the host state could reasonably be expected to do in the light of international anti-corruption norms and national anti-corruption rules. Such inaction may be viewed ‘as disinterest at best and complicity with corrupt acts at worst’.174 Indeed, it has been recognised in the investment treaty context that a host state should be precluded from raising violations of its own laws as a jurisdictional defence when it knowingly overlooked them and endorsed an investment that was not in compliance with its laws,175 and the failure of the host state to prosecute or punish illegal acts by its own public officials has also played a role in other cases where the tribunal either refused or expressed reluctance in entertaining defences of corruption or other forms of illegal conduct.176 However, it may be difficult to assert that the host state’s decision to not prosecute was clearly a form of acquiescence, considering that state prosecutors possess considerable discretion, and that there may be valid reasons why a host state failed to prosecute, such as insufficient evidence to meet the criminal standard of proof and the availability of resources in mounting a high-profile, high-stakes public prosecution of governmental officials.177 Investors will therefore need to show something more, perhaps in the form of a government cover-up or widespread systemic corruption,178 in order to show political will behind the decision to not prosecute.

It should be noted that there can be no acquiescence if the host state had no knowledge of the corruption at hand, or where the host state had not yet reached the point where it would have been expected to prosecute or otherwise punish the public officials in question (e.g., where the corruption was discovered only recently). In World Duty Free v. Kenya—while the tribunal found the fact that no proceedings had been initiated by the host state to prosecute its former head of state for soliciting and accepting bribes or to recover the bribe that was paid, even after the host state was made aware of the bribe paid, was ‘highly disturbing’—the tribunal ultimately held that there could be no ‘affirmation or waiver’ because the host state had only known of the bribery during the arbitration itself, around 30 months after the claimant filed its request for arbitration.179 However, it should be noted that it is not the length of time that is relevant, but rather whether the host state had failed to act in circumstances where it would be expected to do so.180 In view of this, where corruption allegations concerning the activities of public officials arise, tribunals should inquire as to whether concrete steps will be or have been taken by the host state to prosecute or otherwise punish the public officials at hand, and if not, the reasons behind the decision to not do so.

Separately, it has been argued that the failure to investigate and prosecute not only forms the basis of acquiescence, but also a separate violation of substantive international anti-corruption law by the host state.181 On this view, the host state has a positive duty to prevent and redress corruption committed by its public officials under national law and international anti-corruption treaties (e.g., UNAC and OECD), and a failure to act will trigger state responsibility. Such a duty is made clear in some of the more recent BITs, which have included express provisions to this effect.182 The legal consequences that flow from such provisions, however, remain unclear.183

Estoppel

Estoppel is a principle that precludes one from asserting a particular state of things against another if one had previously, by words or conduct, unambiguously represented to the other the existence of a different state of things, and if the other had, on the faith of that representation, so altered his position that the establishment of the truth would injure him.184 There are two competing notions of estoppel under public international law, one restrictive and the other expansive, which have both been relied upon by ISDS tribunals.185 Under the restrictive approach, the essential elements of estoppel are: (i) a statement of fact or conduct that is clear and unambiguous; (ii) this statement or conduct must be voluntary, unconditional and authorised; and (iii) reliance in good faith upon the statement or conduct, either to the detriment of the party so relying on the statement or conduct or to the advantage of the party making the statement or conduct.186 Under the expansive approach, the third element of the test is discarded in favour of a more flexible approach based on the underlying principle of allegans contraria non audiendus est (‘a person adducing to the contrary is not to be heard’).187 Investment tribunals have recognised estoppel as a viable counter-defence against jurisdictional defences run by the host state.188

In the investment treaty context, estoppel would most likely be raised in a situation where high-ranking government officials of the host state had engaged in bribe solicitation and extortion in return for the right to invest in the host state, as in the situation in World Duty Free v. Kenya. While difficult, the corrupt conduct of high-ranking government officials that are attributable to the state could arguably amount to a clear, voluntary and unambiguous statement from the host state that the investment in question is legally valid. Should the investor subsequently rely on such a statement to its detriment by incurring the investment costs (and to the host state’s advantage by reaping the investment’s economic benefits), the host state ought to be estopped from raising defences based on its own corrupt conduct. One of the key questions in such scenarios is whether the conduct of the official in question is attributable to the host state,189 which has unfortunately been given scant attention by investment tribunals thus far.190

Estoppel may also be raised in a situation where the host state has condoned its own government officials’ corruption by failing to prosecute or otherwise punish those government officials or failing to even investigate the circumstances suggesting the government officials’ participation in the corruption at hand. This point was raised in the Wena Hotels v. Egypt case, where the tribunal raised suspicions that the host state had knowledge of the corruption at hand but had decided, for whatever reason, not to prosecute the government official in question, and was therefore ‘reluctant to immunize’ the host state from liability in the arbitration.191 In circumstances where the host state has made known its awareness of the corruption to the investor, but subsequently fails to prosecute or punish the government official in question, that may arguably amount to a clear statement that the investment is legally valid notwithstanding the corruption at hand that the investor had relied upon to its detriment. However, as mentioned above, it may be difficult to assert that there was indeed such a clear statement as the decision to prosecute or punish is not necessarily a political one.

Investors may, however, find it hard to establish detrimental reliance in most situations involving agency corruption, such as the payment of bribes by intermediaries (often known as ‘consultants’) without the investor’s knowledge to government officials for the procurement of the investment contract or licences to operate in the country. This is because the relevant facts that reveal the corruption by agents will usually be known only during the course of the arbitration, by which time the investor would have already raised all of its claims. This makes it difficult for the investor to establish that it had suffered any of its alleged damage as a result of it relying, to its detriment, on the host state’s participation in the corruption.

Costs, Damages and Restitution

Host states that have been found to be complicit in the corrupt act in question (e.g., by failing to investigate and prosecute) may find themselves subject to a contributory fault regime and/or saddled with negative costs orders from the tribunal.

Where there is an express legality clause in the investment treaty, the investor is unlikely to overcome the jurisdictional barrier posed by such a clause.192 However, it may be able to persuade the tribunal to hold the host state at least partially responsible for its complicity in the corruption by issuing a costs award that takes into account the host state’s involvement in the corruption in question. This occurred in both Metal-Tech v. Uzbekistan and World Duty Free v. Kenya, where the tribunals (to varying degrees) recognised the parties’ mutual involvement in the corruption and thereby ordered the parties to bear their own legal fees and to share in the arbitration costs.193

An innovative costs award can be found in Spentex v. Uzbekistan, in which the majority of the tribunal urged the respondent to make certain reforms in its anti-corruption policy and to make a monetary contribution to an international programme targeting corruption.194 The tribunal accomplished this by including in the costs order a choice of two options: (i) Uzbekistan donates USD8 million to one of the United Nations’ anti-corruption funds within 90 days in addition to covering its own legal fees and 50% of the costs of the proceedings; or (ii) Uzbekistan pays 75% of more than USD17 million of the claimant’s legal fees and 100% of the costs of the proceedings in addition to its own legal fees. While Uzbekistan has reportedly since made a contribution to the United Nation’s anti-corruption programme and has initiated joint projects aiming at combating corruption in the country, questions have been raised (in particular by the dissenting arbitrator, Professor Brigitte Stern) as to whether the Spentex tribunal, by issuing such a costs order, had ventured beyond the boundaries of its competences.195 Separately, it has also been observed that, since the BIT did not contain an explicit legality clause and the tribunal ultimately considered the jurisdiction–admissibility distinction to be irrelevant, the Spentex costs order opens the door for future tribunals to make similar costs awards regardless of the basis on which the claims had been refused to be heard.196

It is worth noting that negative costs orders may not only feature in cases where the corruption defence was successfully raised; costs orders have also been used by tribunals as warnings against host states spuriously raising the corruption defence. In Cortec v. Kenya, the tribunal disapproved of Kenya’s conduct in failing to support its corruption allegations with credible evidence; this resulted in the tribunal reducing the costs award by 50%.197 In Tethyan v. Pakistan, the tribunal disapproved of Pakistan’s ‘entirely meritless’ defences, including unproven allegations of corruption, and ordered Pakistan to bear the full costs of both the arbitration and the claimant’s legal fees and expenses.198

Where there is no express legality clause in the investment treaty, or where the corruption at hand amounts to performance corruption, the corruption allegations are likely be treated as an admissibility issue (as discussed above). In such circumstances, the arbitration may proceed to the merits stage of the proceedings, which gives rise to the possibility of awarding damages based on a contributory fault regime.199 In the context of performance corruption, it has been argued that the tribunal should apportion fault between the investor and the host state in consideration of three factors: (i) the nature of the corruption (e.g., a mere ‘grease payment’ would not weigh as heavily on an investor as opposed to a bribe for an illegal benefit); (ii) the prevalence in the host state of the type of corruption (e.g., the payment of bribes would not weigh as heavily against an investor where corruption is endemic to doing business in the host state); and (iii) the degree to which the host state was actively involved in the corruption (e.g., an investor that initiated a bribe is much more at fault than an investor that faced an extortion for a bribe from the host state).200

In contract-based disputes where the investment contract has been invalidated on the grounds of illegality (as discussed above), a non-contractual claim in restitution may, in some circumstances, still be available to account for the extent to which the host state was unjustly enriched at the expense of the investor.201 On one view, the investor should be entitled to ‘an allowance in money for the work done, corresponding to the value of the infrastructure project’.202 This approach finds support in the UNIDROIT Principles of International Commercial Contracts203 as well as recent English law.204

4 Conclusion

The legal issues that may be raised in ISDS proceedings with allegation corruptions are complex and may arise in a myriad of factual scenarios and varying degrees of moral turpitude. Evidential issues concerning burden and standard of proof frequently come to the fore due to the difficulties of obtaining direct evidence of corrupt dealings. This may even prompt a more proactive approach from the tribunal, though any such approach will be subject to limitations inherent in the arbitral process. However, it is not enough to simply establish proof of corruption, as the presence of a host state as a party and the involvement of corrupt public officials will necessarily require a tribunal to engage in questions of attribution of responsibility. Even where the issues of evidence and attribution of responsibility are overcome, there remains the difficulty of deciding what legal consequences should apply to a finding of corruption, which is especially thorny in situations where both parties had been complicit in the corruption. While the issue of corruption in ISDS proceedings is a fast-moving subject that continues to throw up vexing questions for parties and tribunals alike, it is hoped that this chapter has managed to capture a concise snapshot of the zeitgeist of current ISDS practice on corruption allegations.

Notes

  1. 1.

    See e.g., Hamby 2016; Ross 2016.

  2. 2.

    See Llamzon 2014, paras. 7.03–7.09.

  3. 3.

    Llamzon 2014, paras. 7.33–7.34.

  4. 4.

    Caprasse and Tecqmenne 2022, p. 520.

  5. 5.

    See e.g., Elliott Associates L.P. v. Republic of Korea, PCA Case No. 2018-51, UNCITRAL, Award (20 June 2023) (Elliott v. South Korea (Award)); outcome reported in Moody 2023.

  6. 6.

    Nicholls et al. 2011, para. 1.01.

  7. 7.

    Nicholls et al. 2011, para. 1.02.

  8. 8.

    Nicholls et al. 2011, para. 1.03.

  9. 9.

    As of September 2023, 45 countries have ratified or acceded to the convention, including Japan and South Korea.

  10. 10.

    On these two treaties, see further especially Chaps. 4 and 5 in this volume.

  11. 11.

    See Article 15(a), UNCAC.

  12. 12.

    See Article 15(b), UNCAC.

  13. 13.

    See Article 18, UNCAC.

  14. 14.

    Reisman 1979, p. 38.

  15. 15.

    See e.g., the UK Bribery Act 2010.

  16. 16.

    See e.g., the OECD Convention and the FCPA (and Chap. 4 in this volume).

  17. 17.

    EDF (Services) Limited v. Romania, ICSID Case No. ARB/05/13, Award (8 October 2009) (EDF v. Romania (Award)), para. 221.

  18. 18.

    Mills 2003, p. 295.

  19. 19.

    Scherer 2002.

  20. 20.

    Nicholls et al. 2011, para. 9.134.

  21. 21.

    See Tethyan Copper Company Pty Ltd v. Islamic Republic of Pakistan, ICSID Case No. ARB/12/1, Decision on Respondent’s Application to Dismiss the Claims (With Reasons) (19 November 2017) (Tethyan Copper v. Pakistan (Dismissal)), para. 6.

  22. 22.

    See e.g., Chevron Corp & Texaco Petroleum Co v. Republic of Ecuador, UNCITRAL, Second Partial Award on Track II (30 August 2018) (Chevron Ecuador (Partial Award)) (tribunal repeatedly sought to secure the attendance of the allegedly bribed individual at the hearing but ultimately failed to do so).

  23. 23.

    Cremades and Cairns 2003, pp. 65, 85. See also China Machine New Energy Corp v. Jaguar Energy Guatemala LLC [2018] SGHC 101 (China Machine v. Jaguar Energy), para. 224 (Singapore High Court accepted that an arbitral tribunal has ‘the duty and mandate to investigate matters raised which, if proven, would render the award unenforceable for being contrary to public policy’).

  24. 24.

    See e.g., Lao Holdings NV and Sanum Investments Ltd v. Government of the Lao People’s Republic [2021] SGHC(I) 10, para. 153 (‘arbitral tribunals have a pro-active role and cannot simply ignore evidence of corruption’); China Machine v. Jaguar Energy, para. 226; Kenya Airports Authority v. World Duty Free Company Limited t/a Kenya Duty Free Complex [2018] eKLR (High Court of Kenya), para. 35; see also Kendra and Coleman 2018 (and Chap. 12 in this volume).

  25. 25.

    See Hwang and Lim 2012, para. 17.

  26. 26.

    See Consultant v. State Agency and others, Final Award, ICC Case No 7047 (1994), in van den Berg 1996, pp. 79–98.

  27. 27.

    Hwang 2018, pp. 405–419.

  28. 28.

    See Ioan Micula and others v. Romania, ICSID Case No. ARB/05/20, Decision on Jurisdiction and Admissibility (24 September 2008), para. 95.

  29. 29.

    Infinito Gold Ltd v. Republic of Costa Rica, ICSID Case No. ARB/14/15, Award (3 June 2021), paras. 178–181.

  30. 30.

    See Metal-Tech Ltd v. Republic of Uzbekistan, ICSID Case No. ARB/10/3, Award (3 October 2013) (Metal-Tech v. Uzbekistan (Award)), paras. 86, 274; Niko Resources (Bangladesh) Ltd v. Bangladesh Petroleum Exploration & Production Company Limited (“BAPEX”) and Bangladesh Oil Gas and Mineral Corporation (“PETROBANGLA”), ICSID Case Nos. ARB/10/11 and ARB/10/18, Procedural Order 13 (26 May 2016).

  31. 31.

    See F-W Oil Interests, Inc v. The Republic of Trinidad and Tobago, ICSID Case No. ARB/01/14, Award (3 March 2006) (F-W Oil v. Trinidad and Tobago (Award)), para. 211; cf. Article 4(9), IBA Rules on the Taking of Evidence in International Arbitration (17 December 2020) (IBA Rules 2020).

  32. 32.

    Hwang 2018, pp. 412–419.

  33. 33.

    See e.g., Churchill Mining PLC and another v. Republic of Indonesia, ICSID Case No. ARB/12/14 and 12-40, Award (6 December 2016) (Churchill v. Indonesia (Award)), para. 238.

  34. 34.

    While there may a preliminary issue of the law applicable to the issue of the burden of proof, this is unlikely to be an issue in practice due to the broad consensus across national jurisdiction on the principle actori incumbit probatio and its likelihood of being a general principle of law pursuant to Article 38(c) of the ICJ Statute: see Caprasse and Tecqmenne 2022, pp. 531–533. See also Rule 36(2), ICSID Arbitration Rules 2022; Article 27(1), UNCITRAL Arbitration Rules (2021).

  35. 35.

    See e.g., Valle and Carvalho 2022, p. 845.

  36. 36.

    See e.g., EDF v. Romania (Award) (the investor alleged that it was the victim of senior Romanian officials’ demands for bribes made during private conversations but could only rely on the testimony of its employees who allegedly received the bribe requests, which were countered by denials by the host state’s witnesses (the same persons accused of soliciting bribes) and was deemed by the arbitral tribunal to be insufficient to prove corruption).

  37. 37.

    Lamm et al. 2010, p. 701.

  38. 38.

    See e.g., Mills 2003; Rose 2014.

  39. 39.

    See ICC Case No. 6497 (1994).

  40. 40.

    Asian Agricultural Products Ltd (AAPL) v. Republic of Sri Lanka, ICSID Case No. ARB/87/3, Award (27 June 1990) (AAPL v. Sri Lanka (Award)), para. 56; Marvin Roy Feldman Karpa v. United Mexican States, ICSID Case No. ARB(AF)/99/1, Award (16 December 2002), para. 177.

  41. 41.

    See International Thunderbird Gaming Corporation v. The United Mexican States, NAFTA Arbitration, Award (26 January 2006) (Thunderbird v. Mexico (Award)); Inceysa Vallisoletana, S.L. v. Republic of El Salvador, ICSID Case No. ARB/03/26, Award (2 August 2006) (Inceysya v. El Salvador (Award)) (in the context of a legality clause).

  42. 42.

    See Spentex Netherlands B.V. v. Republic of Uzbekistan, ICSID Case No. AR/13/26, Award (27 December 2016) (Spentex v. Uzbekistan (Award)), reported in Djanic 2017 (tribunal applied the principle as a starting point but also considered the manner in which the parties had cooperated during the fact-finding process).

  43. 43.

    See e.g., World Duty Free Company Limited v. The Republic of Kenya, ICSID Case No. ARB/00/7, Award (4 October 2006) (World Duty Free v. Kenya (Award)), para. 166; Metal-Tech v. Uzbekistan (Award), para. 243.

  44. 44.

    See Hwang and Lim 2012, para. 37; Partasides 2010.

  45. 45.

    See Himpurna California Energy Ltd (Bermuda) v. PT (Persero) Perusahaan Listruk Negara (Indonesia), Final Award (4 May 1999), paras. 219–220 (tribunal noted that while it is aware that the arbitral process is not one that is divorced from reality, such grave accusations must be proven and that ‘rumours or innuendo’ will not do).

  46. 46.

    The Rompetrol Group N.V. v. Romania, ICSID Case No. ARB/06/3, Award (6 May 2013) (Rompetrol v. Romania (Award)), para. 178.

  47. 47.

    While there may be a preliminary issue of the law applicable to the standard of proof, most investment tribunals are likely to reason that they are not bound by any specific rule of evidence and may independently determine the applicable standard of proof: see Caprasse and Tecqmenne 2022, pp. 531–532.

  48. 48.

    Nigel Blackaby et al. 2009, p. 387.

  49. 49.

    See Valle and Carvalho 2022, pp. 845–846.

  50. 50.

    Reiner, p. 336.

  51. 51.

    See e.g., In re B (Children) (FC) [2008] UKHL 35, para. 13; Tang Yoke Kheng (trading as Niklex Supply Co) v. Lek Benedict [2005] 3 SLR 263, para. 14.

  52. 52.

    Compare Lao Holdings N.V. v. Lao People’s Democratic Republic, ICSID Case No. ARB(AF)/12/6, Award (6 August 2019), paras. 109–110 (discussed further in Chap. 12 in this volume), Getma International and others v. Republic of Guinea, ICSID Case No. ARB/11/29, Award (16 August 2016) (Getma v. Guinea (Award)), para. 184, EDF v. Romania (Award), para. 221 and Waguih Elie George Siag and Clorinda Vecchi v. The Arab Republic of Egypt, ICSID Case No. ARB/05/15, Award (1 June 2009) (Siag v. Egypt (Award)), paras. 325–326 (requiring ‘clear and convincing evidence’), with Churchill v. Indonesia (Award), para. 244 (discussed further in Chap. 10 in this volume), Libananco Holdings Co Limited v. Republic of Turkey, ICSID Case No. ARB/06/8, Award (2 September 2011) (Libananco v. Turkey (Award)), para. 125, and Rompetrol v. Romania (Award), paras. 180–183 (requiring ‘balance of probabilities’). See also The Republic of Croatia v. MOL Hungarian Oil and Gas PLC, PCA Case No. 2014-15, Final Award (23 December 2016) (Croatia v. MOL (Final Award)), para. 124 (‘it seems clear that there is judicial acceptance of the proposition that the test in matters of corruption should be somewhere between the simple balance of probabilities and absolute certainty’).

  53. 53.

    Metal-Tech Uzbekistan (Award), para. 239. See also Waguih Elie George Siag and Clorinda Vecchi v. The Arab Republic of Egypt, ICSID Case No. ARB/05/15, Dissenting Opinion of Professor Francisco Orrego Vicuña (1 June 2009), p. 4; Niko Resources (Bangladesh) Ltd v. Bangladesh Petroleum Exploration & Production Company Limited (“BAPEX”) and Bangladesh Oil Gas and Mineral Corporation (“PETROBANGLA”), ICSID Case Nos. ARB/10/11 and ARB/10/18, Decision on the Corruption Claim (25 February 2019) (Niko v. B&P (Corruption)), para. 806.

  54. 54.

    See e.g., Perry 2019. (But see Chap. 15 in this volume, where instead the authors argue for the higher standard to be applied so the corruption allegation is examined more closely, otherwise the tribunal might find no corruption and compensation under the award could then be paid only for later domestic proceedings applying the higher standard to find corruption by the investor).

  55. 55.

    See Caprasse and Tecqmenne 2022, p. 541.

  56. 56.

    Nicholls, para. 9.134; see also Perry 2019.

  57. 57.

    See In re Doherty (Original Respondent and Cross-appellant) (Northern Ireland) [2008] UKHL 33.

  58. 58.

    See Churchill v. Indonesia (Award), paras. 240–244; Rompetrol v. Romania (Award), paras. 182–183; Libananco v. Turkey (Award), para. 125. (See further Chap. 7 in this volume, arguing for a differentiated approach more generally regarding allegations of systemic corruption by the host state).

  59. 59.

    Partasides 2010, para. 53.

  60. 60.

    Caprasse and Tecqmenne 2022, p. 543.

  61. 61.

    See e.g., Article 9(1), IBA Rules 2020; Article 27(4), UNCITRAL Arbitration Rules (2013); Rule 36, ICSID Arbitration Rules 2022.

  62. 62.

    See e.g., Spentex v. Uzbekistan (Award), reported in Djanic 2017; Getma v. Guinea (Award), para. 183; Metal-Tech v. Uzbekistan (Award), para. 243; Jan Oostergetel and Theodora Laurentius v. The Slovak Republic, UNCITRAL, Final Award (23 April 2012), para. 303.

  63. 63.

    See e.g., Spentex v. Uzbekistan (Award), reported in Djanic 2017; ICC Case No. 8891 (1998). See also Scherer 2002.

  64. 64.

    See also Valle and Carvalho 2022, pp. 850–853 (providing a list of the most common red flags listed by compliance literature, including the 2019 Basel Toolkit).

  65. 65.

    Lamm et al. p. 704.

  66. 66.

    See Article 9(5), IBA Rules on the Taking of Evidence in International Arbitration (2020). See also Niko v. B&P (Corruption), para. 219.

  67. 67.

    Partasides 2010, paras. 62–77.

  68. 68.

    See e.g., World Duty Free v. Kenya (Award); Metal-Tech v. Uzbekistan (Award); Azpetrol International Holdings B.V. and others v. Republic of Azerbaijan, ICSID Case No. ARB/06/15, Award (8 September 2009).

  69. 69.

    See e.g., Elliott Associates v. South Korea (Award) (Tribunal relied on the factual findings of the South Korean courts with respect to criminal charges); Inceysa v. El Salvador (Award) (investor relied on the Supreme Court of El Salvador’s decision sustaining the legality of certain administrative acts).

  70. 70.

    See e.g., Niko Resources (Bangladesh) Ltd v. People’s Republic of Bangladesh, BAPEX, and PETROBANGLA, ICSID Case Nos. ARB/10/11 and ARB/10/18, Decision on Jurisdiction (19 August 2013) (Niko v. B&P (Jurisdiction)), para. 425 (tribunal acknowledged that national authorities were best placed to investigate and collect proof of corruption relevant to the case).

  71. 71.

    See e.g., EDF v. Romania (Award) (the host state relied on the Romanian Anti-Corruption Authority’s resolutions concluding that there was insufficient evidence to substantiate the claimant’s allegations of corruptions); China Machine v. Jaguar Energy (an international commission’s report was cited by one party in the arbitration as evidence that the other had bribed government officials).

  72. 72.

    See Born 2014, pp. 2310–2311; Brower 1994, p. 48.

  73. 73.

    See e.g., Inceysa v. El Salvador (Award), paras. 209–213 (tribunal held that the state party’s determinations as to the legality or illegality of an investment is not determinative of the issue for the purposes of establishing the tribunal’s jurisdiction under the investment treaty). (See also Chap. 16 in this volume discussing the question whether arbitral tribunals should defer to national court decisions on corruption and illegality).

  74. 74.

    See e.g., Croatia v. MOL (Final Award), para. 85; Niko v. B&P (Jurisdiction), para. 426; TSA Spectrum de Argentina S.A. v. Argentine Republic, ICSID Case No. ARB/05/5, Award (19 December 2008), paras. 174–176. See also China Machine v. Jaguar Energy, para. 227.

  75. 75.

    Malek 2018, para. 28–03.

  76. 76.

    Malek 2018, paras. 43-77–43-79.

  77. 77.

    See e.g., Croatia v. MOL (Final Award), para. 85; Niko v. B&P (Jurisdiction), para. 429.

  78. 78.

    See Born 2014, p. 3774. See also Tethyan Copper v. Pakistan (Dismissal), para. 394 (tribunal ruled out the relevance of a criminal conviction on the ground that it was unrelated to the claimant’s investment).

  79. 79.

    See e.g., Sanum Investments Ltd v. Government of the Lao Peoples Democratic Republic, PCA Case No. 2013–13, Award (6 August 2019) (Sanum v. Laos (Award)), paras. 111, 158 (and Chap. 12 in this volume); Vladislav Kim & Others v. Republic of Uzbekistan, ICSID Case No. ARB/13/6, Decision on Jurisdiction (8 March 2017) (Kim v. Uzbekistan (Jurisdiction)), para. 575; Wena Hotels Ltd v. Arab Republic of Egypt, ICSID Case No. ARB/98/4, Award (8 December 2000) (Wena Hotels v. Egypt (Award)), paras. 111–116; Glencore International AG & CI Prodeco SA v. Republic of Colombia, ICSID Case No. ARB/16/6, Award (27 August 2019), para. 738.

  80. 80.

    EDF v. Romania (Award), para. 221.

  81. 81.

    Chevron Ecuador (Partial Award), para. 8.50.

  82. 82.

    Fraport AG Frankfurt Airport Services Worldwide v. Republic of the Philippines, ICSID Case No. ARB/03/25, Award (16 August 2007) (Fraport v. Philippines (Award)), para. 346. (Discussed further in Chap. 13 in this volume. See also Chap. 7, developing an innovative argument that investors might be able to assert duress in situations of systemic corruption).

  83. 83.

    Metal-Tech v. Uzbekistan (Award), para. 422 (discussed further in Chap. 4 in this volume).

  84. 84.

    See United Nations 2001.

  85. 85.

    See Hobér 2008, p. 550.

  86. 86.

    See Crawford 2010, p. 129; Llamzon 2014, paras. 10.13–10.14. For a recent ISDS example involving allegations of state-only corruption, see Elliott Associates v. South Korea (Award).

  87. 87.

    Article 1, ILC Articles.

  88. 88.

    Article 2, ILC Articles.

  89. 89.

    Article 3, ILC Articles.

  90. 90.

    Article 4(1), ILC Articles. For a recent case in which a corporation was considered to be a de facto state organ, see Elliott Associates v. South Korea (Award), paras. 438–446.

  91. 91.

    See Articles 4–11, ILC Articles.

  92. 92.

    Article 5, ILC Articles.

  93. 93.

    Article 8, ILC Articles.

  94. 94.

    Article 7, ILC Articles.

  95. 95.

    ILC Draft Articles Commentaries, p. 46 (fn. 150).

  96. 96.

    See e.g., Chevron v. Ecuador (Partial Award), paras. 8.43–8.52 (tribunal held that the conduct of a judge in corruptly rendering a judgment against the investor was attributable to the host state).

  97. 97.

    An exception to this would be where the investor was coerced or under duress to participate in the corruption, as this would preclude consent being freely given and act as a vitiating factor (see Commentary (6) to Article 20, ILC Articles).

  98. 98.

    Devendra 2019, pp. 273–275.

  99. 99.

    Llamzon 2014, para. 10.58; cf. Crawford and Mertenskötter 2015, pp. 27–42.

  100. 100.

    See e.g., Metal-Tech Ltd v. Uzbekistan (Award) (critically discussed further in Chap. 4 and generally in Chap. 1 in this volume).

  101. 101.

    Llamzon 2013, p. 76.

  102. 102.

    See Lim 2012, pp. 613–618; cf. Raeschke-Kessler and Gottwald 2008, pp. 596–597and Llamzon 2013. (Compare also the argument about duress developed in Chap. 7 in this volume).

  103. 103.

    Lim 2012, p. 617.

  104. 104.

    See Lim 2012, pp. 635, 639–650.

  105. 105.

    Lim 2012, pp. 652–654.

  106. 106.

    Lim 2012, pp. 654–656.

  107. 107.

    World Duty Free v. Kenya (Award), para. 130 (discussed further in Chap. 4 in this volume).

  108. 108.

    See generally Begic 2005.

  109. 109.

    Newcombe 2011, pp. 192–193.

  110. 110.

    See Desert Line Projects LLC v. Republic of Yemen, ICSID Case No. ARB/05/17, Award (6 February 2008) (Desert Line Yemen (Award)), paras. 104–105. (For examples in some Asian treaty practice, compare Chap. 11 on Japan and Chap. 14 on Korea in this volume).

  111. 111.

    Reichenbach 2022, para. 21.14 (citing Article 14(2), 2016 Iran-Slovakia BIT and Article 17(4), 2016 Morocco-Nigeria BIT).

  112. 112.

    See e.g., Inceysa v. El Salvador (Award) (investor fraudulently misrepresented itself in a bidding process for government contracts); Fraport v. Philippines (Award) (investor knowingly and intentionally circumvented local law on ownership rights of a public utility) (discussed further in Chap. 13 in this volume).

  113. 113.

    Tokios Tokelės v. Ukraine, ICSID Case No. ARB/02/18, Decision on Jurisdiction (29 April 2004), para. 86 (minor administrative defects in the investor’s underlying documents for the registered investments were held to not deprive it of the BIT’s protection); Fraport AG Frankfurt Airport Services Worldwide v. Republic of the Philippines, ICSID Case No. ARB/03/25, Dissenting Opinion of Mr. Bernardo M. Cremades (6 August 2007), para. 37; Hochtief AG v. The Argentine Republic, ICSID Case No. ARB/07/31, Decision on Liability (29 December 2014), para. 199.

  114. 114.

    Lamm et al. pp. 699–731.

  115. 115.

    This refers to corruption that occurs during the performance of the investment that had been lawfully founded (e.g., bribery to accelerate routine paperwork, to expand the investment, or to evade applicable environmental or labour regulations).

  116. 116.

    Reisman 2021, p. 5; Reichenbach 2022, para. 21.12. See e.g., Worley International Services Inc. (USA) v. The Republic of Ecuador, PCA Case No. 2019–15, Final Award (22 December 2023), paras. 421–490 (tribunal majority held that the Claimant's corruption during the operation of its investment rendered its claims inadmissible).

  117. 117.

    Arbitral tribunals are generally empowered to rule on a jurisdictional plea either as a preliminary question or in an award on the merits (see e.g., Article 16(3), UNCITRAL Model Law on International Commercial Arbitration 1985; Article 23(3), UNCITRAL Arbitration Rules 2021; Rule 43(4), ICSID Arbitration Rules 2022).

  118. 118.

    Metal-Tech v. Uzbekistan (Award) (discussed further in Chap. 4 in this volume).

  119. 119.

    Ibid., para. 373.

  120. 120.

    See e.g., David Minnotte & Robert Lewis v. Republic of Poland, ICSID Case No. ARB(AF)/10/1, Award (16 May 2014) (Minnotte v. Poland (Award)), para. 131; SAUR International SA v. Republic of Argentina, ICSID Case No. ARB/04/4, Decision on Jurisdiction and Liability (6 June 2012), para. 308; Álvarez y Marín Corporación S.A. and others v. Republic of Panama, ICSID Case No. ARB/15/14, Final Award (12 September 2018), paras. 135–140.

  121. 121.

    See Reichenbach 2022, paras. 21.31–21.35.

  122. 122.

    See e.g., Spentex v. Uzbekistan (Award), reported in Djanic 2017.

  123. 123.

    Arbitral tribunals are generally empowered to rule on a jurisdictional plea either as a preliminary question or in an award on the merits (see e.g., Article 16(3), UNCITRAL Model Law on International Commercial Arbitration 1985; Article 23(3), UNCITRAL Arbitration Rules 2021; Rule 43(4), ICSID Arbitration Rules 2022).

  124. 124.

    Siag v. Egypt (Award).

  125. 125.

    Empresas Lucchetti, S.A. and Lucchetti Peru, S.A. v. Republic of Peru, ICSID Case No. ARB 03/4, Award (7 February 2005), para. 57.

  126. 126.

    Paulsson 2005, p. 617.

  127. 127.

    See Douglas 2009, paras. 301–312.

  128. 128.

    Lamm et al. pp. 723–236.

  129. 129.

    See e.g., Spentex v. Uzbekistan (Award), reported in Djanic 2017; Churchill v. Indonesia (Award), para. 493 (discussed further in Chap. 10 in this volume); Flughafen Zürich A.G. and Gestión e Ingenería IDC S.A. v. Bolivarian Republic of Venezuela, ICSID Case No. ARB/10/19, Award (18 November 2014), para. 130; Hesham Talaat M Al-Warraq v. Indonesia, UNCITRAL, Final Award (15 December 2014), paras. 645–646; but cf. Niko v. B&P (Jurisdiction), para. 477; Yukos Universal Limited (Isle of Man) v. The Russian Federation, PCA Case No. 2005-04/AA227, Final Award (18 July 2014), paras. 1357–1363; Guyana v. Suriname, Permanent Court of Arbitration, Award (17 September 2007), para. 418.

  130. 130.

    Newcombe 2011, p. 195.

  131. 131.

    Reichenbach 2022, para. 21.40.

  132. 132.

    See e.g., Plama Consortium Limited v. Bulgaria, ICSID Case No. ARB/03/24, Award (27 August 2008) (Plama v. Bulgaria (Award)), para. 144 (tribunal adopted good faith as a free-floating principle of international law that condemns corruption). But cf. criticisms in Reichenbach 2022, para. 21.41.

  133. 133.

    See e.g., Plama v. Bulgaria (Award), para. 143; Khan Resources Inc., et al. v. Government of Mongolia, UNCITRAL, Decision on Jurisdiction (25 July 2012), para. 383; World Duty Free v. Kenya (Award), para. 179.

  134. 134.

    See e.g., Spentex v. Uzbekistan (Award), reported in Djanic 2017 (tribunal found that the respondent’s corruption allegations were made out on the facts and therefore refused to hear the investor’s claims on the basis that the bribery constituted a violation of international public policy); Plama v. Bulgaria (Award), para. 144 (tribunal ruled that the claim was inadmissible because granting treaty protections to the investment would be ‘contrary to the basic notion of international public policy—that a contract obtained by wrongful means (fraudulent misrepresentation) should not be enforced by a tribunal’).

  135. 135.

    World Duty Free v. Kenya (Award), para. 177, citing Tinsley v. Milligan [1994] 1 AC 340 (Tinsley v. Milligan). (See further on this case in Chap. 4 in this volume).

  136. 136.

    See Niko v. B&P (Jurisdiction), paras. 435–446.

  137. 137.

    World Duty Free v. Kenya (Award), para. 186.

  138. 138.

    Patel v. Mirza [2016] UKSC 42 (Patel v. Mirza), para. 120. The UK Supreme Court effectively rejected the rigid illegality doctrine in Tinsley v. Milligan in favour of a broader enquiry (as elaborated in Chap. 4 in this volume).

  139. 139.

    Vail 2016.

  140. 140.

    See e.g., Thunderbird v. Mexico (Award) and Alex Genin and others v. Republic of Estonia, ICSID Case No. ARB/99/2, Award (25 June 2001) (Genin v. Estonia (Award)) (and Chap. 12 on Laos in this volume).

  141. 141.

    See e.g., Minnotte v. Poland (Award), para. 163; Genin v. Estonia (Award), paras. 348–373; Joseph Charles Lemire v. Ukraine, ICSID Case No. ARB/06/18, Decision on Jurisdiction and Liability (14 January 2010).

  142. 142.

    See e.g., MTD Equity Sdn Bhd & MTD Chile SA v. Chile, ICSID Case No. ARB/01/7, Award (25 May 2004), paras. 242–243.

  143. 143.

    See e.g., Cementownia ‘Nowa Huta’ SA v. Turkey (UNCITRAL (NAFTA)), Award (13 August 2009), para. 177 (investor ordered to bear all costs of the arbitration and the host state’s legal fees and expenses due to its abuse of process); Churchill v. Indonesia (Award), paras. 543–557 (investor ordered to pay the fees and expenses of the tribunal, ICSID administrative fees, and 75% of the expenses incurred by the host state for the proceedings due to the proven allegations of forgery and fraud) (discussed further in Chap. 10 in this volume).

  144. 144.

    See e.g., Churchill Mining PLC and Planet Mining Pty Ltd v. Republic of Indonesia, ICSID Case No. ARB/12/14 and 12/40, Procedural Order No. 15 (12 January 2015) (tribunal bifurcated proceedings to deal with forgery allegations first) (discussed further in Chap. 10 in this volume).

  145. 145.

    See e.g., F-W Oil v. Trinidad and Tobago (Award) (investor alleged host state demanded payment of a USD1.5 million bribe as a condition for continuing negotiation of an operating agreement; allegation ultimately abandoned); Rumeli Telekom AS & Telsim Mobil Telekomikasyon Hizmetleri AS v. Republic of Kazakhstan, ICSID Case No. ARB/05/16, Award (29 July 2008) (Rumeli v. Kazakhstan (Award)) (investor alleged systemic corruption of the Kazakhstan judiciary and the solicitation of a bribe by a Kazakh judge); RSM Production Corporation v. Grenada, ICSID Case No. ARB/05/14 (Annulment Proceeding), Decision on RSM Production Corporation’s Application for a Preliminary Ruling of 29 October 2009 (7 December 2009) (investor alleged that Grenada’s denial of an exploration licence and termination of an agreement were motivated by a bribe that was paid or was to be paid to the then Attorney General of Grenada by a third-party corporation); EDF v. Romania (Award) (investor alleged that host state failed to extend contractual arrangements beyond their ten-year term because it had refused to pay a USD2.5 million bribe). (Compare also the Elliott Associates v. South Korea claim discussed in Chap. 14 in this volume).

  146. 146.

    See e.g., the US Model BIT (2012) (contains a ‘minimum standard of treatment clause’ that provides that each party ‘shall accord to covered investments treatment in accordance with customary international law, including fair and equitable treatment’).

  147. 147.

    Haugeneder and Liebscher 2009.

  148. 148.

    Haugeneder and Liebscher 2009.

  149. 149.

    EDF v. Romania (Award), para. 221.

  150. 150.

    Rumeli v. Kazakhstan (Award), paras. 652–653. (On judicial corruption generally, see also Chap. 10 on Indonesia in this volume).

  151. 151.

    Chevron v. Ecuador (Partial Award) at Part VIII.

  152. 152.

    Haugeneder and Liebscher 2009.

  153. 153.

    Blackaby et al. 2009, para. 8.115; see also AAPL v. Sri Lanka (Award), para. 77.

  154. 154.

    See e.g., AAPL v. Sri Lanka (Award); Azurix Corp v. The Argentine Republic, ICSID Case No. ARB/01/12, Award (14 July 2006), para. 408; National Grid PLC v. Argentine Republic, UNCITRAL, Award (3 November 2008), para. 189; Biwater Gauff (Tanzania) Ltd v. United Republic of Tanzania, ICSID Case No. ARB/05/22, Award (24 July 2008), para. 729. But cf. Saluka Investments BV v. The Czech Republic, UNCITRAL, Partial Award (17 March 2006), para. 484.

  155. 155.

    Klaw 2015, p. 82.

  156. 156.

    Case Concerning Elettronica Sicula S.p.A. (ELSI) (United States of America v. Italy) (1989) ICJ 15, para. 128.

  157. 157.

    Antoine Goetz et consorts v. République du Burundi, ICSID Case No. ARB/95/3, Award (10 February 2009), para. 121.

  158. 158.

    See e.g., Article 11.3, 2012 US–South Korea FTA.

  159. 159.

    See e.g., Elliott Associates v. South Korea (Award), paras. 627–668 (investor’s national treatment claim ultimately failed due to the impugned measures being subject to an express reservation to the national treatment clause set out in the treaty).

  160. 160.

    See e.g., Article 6(1), US Model BIT 2012.

  161. 161.

    See generally Lim 2012, pp. 639–650.

  162. 162.

    See e.g., Cheng 1987, pp. 155–158.

  163. 163.

    MacGibbon 1958, p. 512.

  164. 164.

    Cedeño 2003, para. 67.

  165. 165.

    Cedeño 2003, para. 113.

  166. 166.

    See Antunes 2000, p. 33.

  167. 167.

    World Duty Free v. Kenya (Award), para. 130.

  168. 168.

    See e.g., Mutua 1994, pp. 50–56; Wrong 2009.

  169. 169.

    Lim 2012, pp. 662–664.

  170. 170.

    Case Concerning Delimitation of the Maritime Boundary in the Gulf of Maine Area (Canada/United States of America) (1984) ICJ Report 246, para. 130.

  171. 171.

    See e.g., Case Concerning Certain Phosphate Lands in Nauru (Nauru v. Australia) (1992) ICJ Report 240, para. 32 (recognition of possibility of delay resulting in a state’s claim being rendered inadmissible).

  172. 172.

    ILC Articles Commentaries, p. 122 (Commentary on Article 20 at para. 6).

  173. 173.

    Reichenbach 2022, paras. 21.28, 21.57 (citing Wena Hotels v. Egypt (Award)).

  174. 174.

    Llamzon 2014, para. 10.87.

  175. 175.

    Fraport v. Philippines (Award), para. 346.

  176. 176.

    See Southern Pacific Properties v. Arab Republic of Egypt, ICSID Case No. ARB/84/3, Award (20 May 1992), para. 127; Wena Hotels v. Egypt (Award), para. 116.

  177. 177.

    See Kim v. Uzbekistan (Jurisdiction), para. 406.

  178. 178.

    Compare also the argument for duress, in this context, developed in Chap. 7 in this volume.

  179. 179.

    World Duty Free v. Kenya (Award), paras. 180, 184. (Compare the findings in Sanum v. Laos, discussed in Chap. 12 in this volume).

  180. 180.

    Tams 2010, p. 1043.

  181. 181.

    Llamzon 2015, p. 275.

  182. 182.

    See e.g., Article 17(5), 2016 Morocco–Nigeria BIT (‘The States Parties to this Agreement, consistent with their applicable law, shall prosecute and where convicted penalize persons that have breached the applicable law implementing this obligation’).

  183. 183.

    See Reichenbach 2022, para. 21.29.

  184. 184.

    See generally Brown 1996.

  185. 185.

    See Reichenbach 2022, paras. 21.21–21.23.

  186. 186.

    Case Concerning the Temple of Preah Vihear (Cambodia v. Thailand) (1962) ICJ Report 6, p. 32; Pope & Talbot v. The Government of Canada, UNCITRAL, Award on the Merits of Phase 2 (10 April 2001), para. 111.

  187. 187.

    MacGibbon 1958, p. 470.

  188. 188.

    See e.g., Fraport v. Philippines (Award), para. 346; Desert Line v. Yemen (Award), paras. 119–120; Ioannis Kardassopoulos v. The Republic of Georgia, ICSID Case No. ARB/05/18, Decision on Jurisdiction (6 July 2007), paras. 191–193; Niko v. B&P (Jurisdiction), para. 455.

  189. 189.

    See e.g., Kim v. Uzbekistan (Jurisdiction), para. 540 (tribunal’s minority entertained the estoppel argument but found no evidence showing that the host state had the ‘requisite knowledge’ of the corrupt conduct or that the investors had detrimentally relied on a representation or conduct of the respondent).

  190. 190.

    See Reichenbach 2022, paras. 21.66–21.67 (citing the diametrically opposing results in Spentex v. Uzbekistan (Award) and Kim v. Uzbekistan (Award)).

  191. 191.

    Wena Hotels v. Egypt (Award), para. 116.

  192. 192.

    See generally Reichenbach 2022.

  193. 193.

    Metal-Tech v. Uzbekistan (Award), paras. 398, 422; World Duty Free v. Kenya (Award), paras. 190–191.

  194. 194.

    See Spentex v. Uzbekistan (Award), reported in Djanic 2017.

  195. 195.

    See Tussupov 2022, pp. 84–85.

  196. 196.

    Reichenbach 2022, para. 21.62.

  197. 197.

    Cortec Mining Kenya Limited, Cortec (Pty) Limited and Stirling Capital Limited v. Republic of Kenya, ICSID Case No. ARB/15/29, Award (22 October 2018), paras. 390–401.

  198. 198.

    Tethyan Copper Company Pty Limited v. Islamic Republic of Pakistan, ICSID Case No. ARB/12/1, Award (12 July 2019), paras. 1842, 1845, 1853–1854.

  199. 199.

    See Reisman 2021. See also Reichenbach 2022, para. 21.58.

  200. 200.

    Reisman 2021, pp. 14–19.

  201. 201.

    See Partasides 2017, p. 754.

  202. 202.

    Elgueta 2016.

  203. 203.

    See Article 3.3.2(1), UNIDROIT Principles of International Commercial Contracts 2016 (‘Where there has been performance under a contract infringing a mandatory rule under Article 3.3.1, restitution may be granted where this would be reasonable in the circumstances’).

  204. 204.

    Patel v. Mirza, para. 121 (restitution prima facie available even vis-à-vis illegal contracts).