Investment arbitration tribunals have addressed allegations of corruption with asymmetrical and divergent conclusions regarding the consequences for investors and states and the appropriate standard of proof. Victories on grounds of corruption have been granted to states on numerous occasions, and this has become one of the defenses du jour for States. Surveying the last two decades of ISDS awards involving corruption issues, the roadmap for states to avoid liability would appear to be clear: raise corruption allegations at virtually no risk, as even with presumptions or no direct evidence a tribunal may find corruption. In so doing, States can gain effective immunity from liability for acts that may constitute violations of international investment law, with no consequences for the state's own illicit acts. This landscape has raised serious concerns of fairness and has given rise to criticism of the predictability and legitimacy of the system, making apparent the need of substantial reforms.
Areas in Need of Reform
When an investor brings a treaty claim and is accused of corruption in the making or performance of its investment, the investor's claim is usually subject to dismissal.
In the seminal case of World Duty Free v. Kenya, the tribunal found the investor had made a corrupt payment to the president of Kenya, and therefore the investor was "not legally entitled to maintain any of its pleaded claims." World Duty Free Company Limited v. Republic of Kenya, ICSID Case No. ARB/00/07, Award, 4 October 2006, para. 179. When a state defends a treaty claim and is likewise accused of corruption in its treatment of the investor's investment, however, the consequences are nowhere near as drastic–the case goes on, and the state's conduct is at most evaluated as a whole, with clear corruption as but one factor among many. Even the tribunal in World Duty Free found it "highly disturbing" that the state, whose most senior officer was bribed, had advanced a complete defense based on that bribe without ever prosecuting the former president. World Duty Free, Award, para. 180. And yet, the tribunal held that the Kenyan president's receipt of the bribe could not be imputed to the state itself, and the tribunal declined to engage in a "balancing operation reflecting the relative misconduct" between the parties. Word Duty Free, Award, paras. 169, 176–78.
Other tribunals and commentators have recognized that this approach "possibly exonerate[s] defendants that may have themselves been involved in the corrupt acts." Metal-Tech Ltd. v. Republic of Uzbekistan, ICSID Case No. ARB/10/3, Award, 4 October 2013, para. 389. Notwithstanding the recognition of the effects of this approach, the tribunal in Metal-Tech dismissed the investor's claims in full and merely ordered the parties to split the costs of the arbitration and to bear their own legal fees and expenses.
Tribunals have closely adhered to the approach of World Duty Free. Though in some cases, such as Yukos Universal Limited (Isle of Man) v. Russia, PCA Case No. 2005-04/AA227, Final Award, 18 July 2014, para. 1615, tribunals have addressed an investor's economic wrongdoing by applying principles of contributory fault, no such publicly known case involved bribery or similar corruption of state officials. States are thus able to continue escaping responsibility by pointing to corruption even though the State may be fostering the very corruption that it invokes to escape liability for its internationally wrongful conduct.
One area that needs harmonization and reform is the standard of proof that must be satisfied to demonstrate corruption. Tribunals addressing corruption allegations have reached divergent conclusions regarding the applicable standard of proof, also with some asymmetry depending on whether it is the state or the investor who raises allegations of corruption.
In EDF v. Romania, the claimant alleged that the state's officials demanded bribes to extend contractual arrangements involving an airport and, because the claimant refused to do so, the contract was not extended. The claimant alleged that the state's actions in this regard violated the protections for its investments under the treaty. The tribunal required "clear and convincing" evidence that a bribe was requested, stressing the seriousness of accusing government officials of engaging in corruption and concluding that the testimony and other evidence presented by the claimant was not convincing. The tribunal even held that "there is general consensus among international tribunals and commentators regarding the need for a high standard of proof of corruption." EDF (Services) Limited v. Romania, ICSID Case No. ARB/05/13, Award, 8 October 2009, para. 221.
In contrast, several tribunals addressing defenses of corruption by States have applied a much more relaxed standard of proof. In Metal-Tech v. Uzbekistan the tribunal relied on circumstantial evidence, indicia of corrupt activity and adverse inferences to conclude that multiple persons were involved in a bribery scheme in connection with the approval and establishment of the claimant's investment. The tribunal noted that "corruption is by essence difficult to establish and that it is this generally admitted that it can be shown by circumstantial evidence." Metal-Tech Ltd. v. Republic of Uzbekistan, ICSID Case No. ARB/10/03, Award, 4 October 2013, para. 243. The Metal-Tech case is also notable because the tribunal considered "red flags" (indicia) to support its conclusion that a payment to consultants had constituted a bribe by which the claimant procured its investment. Some tribunals have also supported and adopted a "connect the dots" methodology in addressing allegations of corruption by states. Similar to the "red flags" approach, this one is also based on indicia and inferences drawn from connecting "dots" of evidence. For example, in Spentex v. Uzbekistan, the tribunal adopted a "connect the dots" approach and held that the payment made to consultants before the conduction of the public tender gave rise to a presumption that those payments were intended to bribe government officials, even though the respondent had not identified a particular recipient of the alleged bribe. Spentex Netherlands B.V. v. Republic of Uzbekistan, ICSID Case No. ARB/13/26, Final Award, 27 December 2016.
The Spentex tribunal's conclusion that there was corruption based on "connecting the dots" and mere presumptions is controversial, and it may lead tribunals to decline jurisdiction over claims advanced by investors when there is no good evidence to substantiate that the investment was procured via corruption. In Methanex Corporation v. USA–an example of a case in which the tribunal decided that it could "connect the dots" and draw inferences to find corruption–the tribunal actually rejected the claimant's corruption allegations stressing that there was "no credible evidence" of certain specific corrupt acts that the claimants had alleged. Methanex Corporation v. United States of America, PCA, Final Award on Jurisdiction and Merits, 3 August 2005, PART III, Ch. B, para. 60. Similarly, in ECE v. Czech Republic, when applying the "connect the dots" approach in connection with allegations of corruption by an investor, the tribunal expressed its "reservation" as to whether charges of corruption can be advanced "without either some direct evidence or compelling circumstantial evidence." ECE Projektmanagement v. The Czech Republic, UNCITRAL, PCA Case No. 2010-5, Award, 19 September 2013, para. 4.876. These different approaches, including when the claims of corruption are advanced by investors or States, demonstrate that there is far from a consensus on the standard of proof applicable to corruption allegations. This is a fundamental issue that arbitrators and the arbitration community as a whole should be concerned about and efforts are needed to achieve harmonization and fairness and equality in the system.
The simplest way to even out the consequences of corruption is to increase them for states, lessen them for investors, or employ some combination of both. More importantly, however, some reforms may not only level the playing field between investors and states, but may also discourage corruption in the first place.
First, in cases involving corruption by both an investor and a state, tribunals could apply principles of contributory fault, which are reflected in the ILC Articles on State Responsibility. This was the approach taken by the tribunal in the Yukos case, which reduced damages by 25 percent, though the pertinent allegations concerned the investor's abuse of low-tax regimes of Russia and not bribery or direct forms of corruption. In a case involving corruption by both an investor and a state, a tribunal could award damages by a degree reflective of the investor's fault in corrupt actions, relative to the fault of the state, subject to any jurisdictional limitations of applicable investment treaties or arbitral rules.
Second, states whose officials are alleged to be involved in corruption with an investor could be required to show that they took adequate measures to prevent corruption in the first place or address it after the fact. Investors, too, could be allowed to raise a defense that they took adequate measures to prevent and address corruption. This approach would encourage investors and states alike to prevent corruption, and to root out systems that foster corruption lest the investor or state lose its claim or defense. Adopting this approach would mark a departure from World Duty Free, in which the tribunal noted that the Kenyan president involved in the corruption had not been prosecuted, but imposed no consequence for such failure. This approach of allowing States to invoke corruption as a defense while at the same time not penalizing them for their participation in the corruption or for failing to take measures to address corruption is unfortunately prevalent at moment.
Third, it could be made easier for investors to bring claims against states for attempts to extract bribes. Though an unmet demand for a bribe that results in no investment may fall beyond the jurisdiction of an arbitral tribunal limited to actual investments, domestic law may provide some recourse. On the other hand, attempts to extract bribes in the performance of an existing investment can amount to violations under the applicable treaty (e.g., fair and equitable treatment, the obligation to not adopt arbitrary and discriminatory measures against the investment or obstruct its management, development and expansion) and cause significant damages to the investment, and tribunals should not discount them.
Fourth, strict rules of attribution, such as those applied in World Duty Free, should not relieve states of responsibility for maintaining systems that allow corruption to flourish. Corruption is a systemic issue, and a myopic focus on whether an official was acting in a personal capacity as opposed to a governmental one should not allow states to evade liability by raising as a defense the corruption their own systems have fostered or have failed to prevent.
Fifth, tribunals have discretion in awarding costs, and could wield that power to more fairly handle dismissals of investors' claims due to corruption in which states were involved. One tribunal (Spentex v. Uzbekistan) dismissed the investor's claims on ground of corruption and ordered the state to either (1) donate USD 8 million to a United Nations anti-corruption fund, or (2) pay the costs of the proceedings and reimburse 75 percent of the investor's legal fees. Awarding costs to investors whose claims were dismissed for state-involved corruption not only deters states from engaging in corruption but also helps defray the costs of investors whose claims are dismissed for only partial or no fault of their own. Taking a similar approach but raising the stakes by requiring the State to pay a significant portion of the amount invested by the investors whose claim will not be adjudged on the merits due to the finding of corruption would move the scale even more and may help to further diminish corruption.
To implement these reforms it is critical to build consensus among arbitrators and the arbitration community on the consequences that states and investors should expect for cases of corruption and also on the standard of proof applicable to demonstrate such allegations. This is the sort of general consensus that was proclaimed to exist by the tribunal in EDF v. Romania, at least in connection with the need for a high standard of proof of corruption, but that subsequent awards have shown to be perhaps overly optimistic.
Another mechanism to create legal certainty is for treaties to include provisions on these issues, either by joint interpretative statements by states on concluded treaties or by replacing old treaties with modern ones. Investment treaties are silent on the effects of corruption and the applicable standard of proof. This lacuna has led arbitrators to look more to the investor as to its obligations for acts of corruption relying on local law, and to rely on its ample discretion as to the standard of proof to be applied, which has led to more uncertainty and inconsistent decisions. Treaties can refer to international law and best practices dealing with corruption, thus directing tribunals to rely on these in their decisions, including the ILC Articles of State Responsibility and international instruments like the OECD Convention on Combating Bribery of Foreign Public Officials in International Business Transactions and the UN Convention Against Corruption. Tribunals would also be able to rely on best practices as embodied in these instruments or national laws, including on the consequences of the illicit acts. This will help combat corruption and at the same time bring more fairness, predictability and legitimacy to the system.