Kluwer Arbitration Blog

Recent years have seen an uptick in the expansion and enforcement of anti-corruption laws worldwide. In 2017, China amended its Anti-Unfair Competition Law, broadening the scope of bribe recipients covered by the law, and increasing penalties. In 2019, Italy widened its anti-bribery law, No. 3/2019, increasing penalties for both individuals and companies found guilty of bribery. Also, the past few years have been some of the most active for U.S. Foreign Corrupt Practices Act (“FCPA”) litigation. In 2019, the Department of Justice (“DOJ”) and Securities and Exchange Commission (“SEC”) reported more than fifty actions under the Foreign Corrupt Practices Act (“FCPA”), and collected more than US$2.5 billion in total penalties, fines, and disgorgement in FCPA settlements. In the UK, the powers and funding of the Serious Fraud Office (“SFO”) have been gradually increasing, and the appointment of a new Director in 2018 has given new impetus to the SFO’s investigation of bribery and corruption offenses.

Given these global developments, it is unsurprising that corruption defenses have become more common in international investment arbitration, and often these corruption investigations occur in parallel with related arbitration proceedings. Yet, while much has been written about the burden of proof for parties invoking corruption defenses within the context of an arbitration, little has been written about how tribunals treat domestic or international criminal investigations or court rulings regarding the same allegations of a corrupt act at issue in an arbitration. Specifically: How do tribunals adjudicating a corruption defense consider parallel criminal investigations into that alleged act of corruption, a conviction, or even the absence of an investigation entirely? This post addresses these issues in the limited context of international investment arbitration.

Background on Corruption Defenses in International Investment Arbitration

Corruption defenses generally arise in two scenarios: (1) where a State counterclaims that an investor obtained the investment in dispute via corruption; and (2) where an investor alleges they were forced to pay a bribe to ensure continued performance and/or fair treatment by the host State during the life of the investment.

The seminal arbitration award squarely addressing a corruption defense is World Duty Free v. Kenya in 2006. In World Duty Free, an airport concessionaire accused Kenya of breaching a 1989 contract by expropriating its investment. Notably, World Duty filed documents in which its CEO admitted to making covert payments to the former president of Kenya. The tribunal held that World Duty’s expropriation claim based on “contracts obtained by corruption” could not be upheld by the tribunal because such bribery contravened international public policy (paras. 157, 188). World Duty Free established that investors may be entirely precluded from pursuing claims where the investment at issue was obtained via corruption, even where state officers had engaged in the corruption—recognizing that anti-corruption public policy aimed to protect “not the litigating parties but the public” (para. 181) (emphasis added).

Since World Duty Free, corruption defenses have become more common. Indeed, corruption defenses are commonly asserted by States to argue a claim is inadmissible or that a tribunal lacks the jurisdiction to hear a claim due to an investment being tainted by corruption.

In practice, most corruption defenses are dismissed because the tribunals find there to be insufficient evidence for corruption (See, e.g., Croatia v. MOL Hungarian Oil and Gas PLC; see also  Glencore International A.G. and C.I. Prodeco S.A. v. Colombia). Indeed, to date, only a few arbitral claims have been dismissed on the basis of corruption (See World Duty Free; see also Spentex Netherlands, B.V. v. Uzbekistan; Metal-Tech Ltd. v. Uzbekistan).

Logically, domestic corruption findings may affect enforcement of awards or promote settlements, albeit information on such issues is scant. One example is Siemens A.G. v. Argentine Republic, in which an ICSID tribunal concluded that Argentina violated the Argentina-Germany bilateral investment treaty (“BIT”) and ordered Argentina to pay Siemens approximately $208 million, among other payments (Award, para. 403). Later, however, it was revealed in a German judicial proceeding that Siemens had obtained the contract at issue in the arbitration through bribery. Argentina requested the award be revised based on this new evidence. In response to this request for revision and following settlement talks, Siemens withdrew its claim to the award.

Corruption is difficult to prove given the innately covert nature of the crime and often the limited actors involved. Moreover, corruption defenses demand a high level of proof due to the gravity of the allegation and its consequences. As a result of this proof requirement and the limited powers of arbitral tribunals to investigate corruption—contrasted with domestic criminal authorities’ investigatory resources and powers—tribunals often look to domestic findings regarding an alleged corrupt act in assessing corruption defenses.

 

The Intersection of Investment Arbitration Corruption Defenses and Parallel Criminal Investigations and/or Proceedings

Parties invoking corruption defenses bear the burden of proof and may furnish as proof of corruption a conviction or investigatory findings from domestic authorities (See, e.g., Vladislav Kim and others v. Uzbekistan, paras. 417, 557, 614;  EDF (Services) Limited v. Romania, paras. 221, 232, 237, 291).¹⁾Another important consideration in navigating parallel domestic findings in arbitrations is confidentiality, particularly ensuring arbitrators do not risk breaching confidentiality and that counsel cannot access confidential attorney communications in domestic criminal investigations. See Glencore v. Republic of Colombia, ICSID Case No. ARB/16/6, Award, para. 86. A review of arbitrations addressing domestic criminal investigations serves to demonstrate how tribunals have approached domestic findings in adjudicating corruption defenses.

 

Domestic Findings of Corruption Are Not Controlling Case law in Arbitrations

From a procedural perspective, a review of arbitral decisions suggests that a party should seek to furnish more than proof of domestic findings or conviction of corruption to substantiate a corruption defense since a number of investment arbitration tribunals have confirmed that domestic corruption findings or convictions are not binding on arbitral tribunals (See Karkey Karadeniz Elektrik Uretim A.S. v. Pakistan, para. 544; see also Croatia v. MOL Hungarian Oil and Gas PLC). In Glencore International A.G. v. Colombia, Colombia cited pending domestic criminal corruption proceedings for its corruption defense. Although the tribunal recognized that domestic criminal findings “must be considered” when evaluating evidence of corruption, it noted that such findings do not bar tribunals from reaching a contrary conclusion as to corruption because domestic proceedings and arbitrations function in “different legal spheres” and are subject to “diverging standards of proof” (paras. 673–74). Likewise, in Inceysa v. El Salvador, the tribunal rejected the proposition that a domestic court’s denial of a corruption claim was binding and “constitute[ed] res judicata” (paras. 67, 212, 214, 217).

Domestic findings of corruption alone thus appear to be insufficient to prompt a tribunal to find inadmissibility, lack of jurisdiction, or a merits-based finding of corruption. Tribunals will consider the issues themselves and reach an independent decision: Domestic findings are therefore one of multiple factors in their corruption analyses.

 

However, Tribunals Generally Treat Domestic Corruption Findings, Especially Convictions, as Probative Evidence of Corruption

Yet, tribunals generally treat positive findings of corruption by domestic courts, especially convictions, as persuasive evidence, considering such findings in their proof analyses. In Metal-Tech Ltd. v. Uzbekistan, where the tribunal concluded it lacked jurisdiction due to the investment being tainted by corruption, the tribunal considered a domestic criminal investigation and ongoing proceedings as one of multiple factors in its corruption analysis. Along with these domestic actions, the tribunal considered the circumstances surrounding the alleged bribery such as: the amount of the payment, the lack of services rendered for the payment, the claimant’s affiliates’ connections with public officials in charge of the investment, and the claimant’s lack of a plausible reason for the payments made to the government official. Indeed, the tribunal considered the “entire evidentiary record” to find that Uzbekistan established corruption, of which the domestic investigations and proceedings were part (Award, para. 325).

Moreover, in Niko Resources, the tribunal at the jurisdictional stage recognized that Bangladesh established corruption, in part, because a Canadian court convicted the claimant for the corruption following its guilty plea—though it ultimately dismissed respondents’ corruption objection because there was “no link of causation between the established acts of corruption and the conclusion of the agreements and it [was] not alleged that there [was] such a link” (paras. 455, 464). According to the tribunal, the claimant had “committed the acts of corruption which were sanctioned in the Canadian conviction” (para. 423). The Niko Resources tribunal took the understandable position that the conviction of a corruption offence supported a finding of corruption in the arbitration, and may have denied jurisdiction if the respondents had established a sufficient nexus between the established corruption and the conclusion of the agreements. Notably, in a later proceeding where the Respondents requested the tribunal reconsider its determination of its corruption claim, the tribunal rejected the corruption allegations as “unfounded,” and not “instrumental in the procurement of the Agreements.” (paras. 2005–08).

Unsurprisingly, therefore, while positive domestic findings of corruption bolster a corruption defense, negative domestic findings may undermine corruption defenses. In EDF v. Romania, EDF alleged that state officials tried to solicit a bribe, and that when EDF refused, the state unlawfully retaliated. While this arbitration did not concern a corruption defense, it is instructive as to how tribunals may treat negative domestic findings regarding corruption allegations generally. The EDF tribunal considered that the Romanian National Anticorruption Directorate (“DNA”) investigated the claimant’s bribery claim “and twice rejected it,” and that criminal courts “affirmed the DNA’s conclusions that the claimant’s bribery allegations [were] groundless” (para. 228). Given these negative findings and that the claimant’s witness testimony as to the corruption was found to be “of doubtful value,” the tribunal found that the claimant failed to establish its bribery allegation (paras. 231, 237).

 

When Faced with Pending Investigations, Tribunals May Conduct an Independent Analysis of the Corruption Allegations, Notwithstanding the Fact that Investigations Have Not Concluded

Sometimes, a domestic investigation is still pending at the time of an arbitration. Although there are not many publicly available examples of tribunals grappling with this situation, it is clear that such pending investigations bear evaluation by tribunals. Those decisions that are available suggest that the pending status of national investigations does not preclude tribunals from making independent determinations as to whether a party has adduced sufficient proof of corruption. Indeed, this appears to be the favored approach. In some instances, even if no evidence of a finding of corruption is submitted, tribunals have considered public factors such as the length of an investigation or the existence (or lack) of public proceedings as factors relevant to adjudicating an allegation of corruption. Moreover, while tribunals may consider information in pending investigations as such investigations develop, if appropriate, they are unlikely to postpone rendering an award based solely on the existence of a pending investigation.

For example, the Niko Resources tribunal rejected certain corruption allegations where investigations had “not led to any trial, let alone conviction” for corruption or where “other regulatory investigations appear[ed] to remain outstanding” (Decision on Jurisdiction, paras. 391 426–29). Despite the ongoing status of these investigations, the tribunal considered the length of the investigations and absence of a consequent trial or conviction to weigh against a finding of corruption.

In TSA Spectrum de Argentina S.A. v. Argentina, the tribunal acknowledged ongoing criminal investigations into the alleged bribery. By the time of the award, the investigation had been ongoing for seven years, and public officials and persons affiliated with TSA had been charged. Despite the ongoing nature of these investigations and there being no final judgment, the tribunal found that it was “not unable to determine the TSA’s investment was not made in accordance with Argentine law” (para. 167). Conducting an independent analysis of the record, the tribunal concluded that Argentina failed to adduce proof to substantiate its bribery defense. Notably, the tribunal dismissed the underlying claim of the arbitration on another jurisdictional ground, thus the corruption issue was not outcome determinative. The tribunal acknowledged that had there been no other jurisdictional bar, it may very well have joined the corruption defense to the merits of the arbitration.

There exists the possibility that a tribunal may postpone decisions based on the status of a national investigation, and decide on jurisdiction or liability once it deems the record sufficient. If national proceedings are sufficiently far advanced, this may appear to be an attractive approach for a tribunal to take. However, a tribunal may be unwilling to postpone decisions in an international arbitration indefinitely based on the pendency of national proceeding or investigations. In Fraport AG v. the Republic of the Philippines, the Philippines requested the tribunal postpone its decision, as criminal investigations in Germany, Hong Kong, and the United States were ongoing. The Philippines argued the pending status of these investigations “impeded its ability to fully present its case” (para. 27). Yet the tribunal—which had already postponed proceedings before due to these ongoing investigations—found the record to be sufficient to make a determination as to jurisdiction and liability and denied the postponement request. Notably, the tribunal previously ordered the parties to update it on ongoing developments in the pending Hong Kong and German investigations, supporting the conclusion that a tribunal may consider evidence in pending investigations as it develops (paras. 57, 59, 61).

 

The Absence of Domestic Findings of or Action Against an Alleged Corrupt Act Undermines a Party’s Corruption Allegations

The absence of a domestic investigation, charge, or conviction decreases the likelihood of an arbitral corruption finding. In Lao Holdings, the tribunal held that Laos failed to establish its corruption defense to an expropriation claim. In doing so, the tribunal found it “disturbing” that no prosecutions were ever initiated against “any persons alleged to have accepted bribes,” nor was there any evidence of a “serious criminal investigation of anyone other than the principals of the Claimant” (paras. 111–12) (emphasis in original). The tribunal described such “omissions” as “relevant to the credibility of the Government’s allegations” (para. 112).

Similar to the Lao Holdings tribunal, the Glencore tribunal considered the lack of any investigation or indictment by the state regarding the corrupt act at issue in the arbitration to undermine the State’s corruption defense. Although the absence of an investigation or indictment did not “preclude a hypothetical finding by [the] Tribunal that corruption has occurred,” such omissions by domestic authorities, “which have a much higher capacity of investigation than [the] Arbitral Tribunal, is one of various elements that must be considered when evaluating the available evidence” (Award, paras. 673–74). Thus, just as domestic positive findings of corruption are probative evidence, so too are absences of domestic action targeting the corrupt act.

Also potentially pertinent to tribunals’ analyses of domestic corruption action (or lack thereof) are the perceived motivations behind the domestic action. For example, in Getma International v. Guinea (II), the tribunal emphasized repeatedly that Guinea did not launch an investigation into the alleged corruption until long after a complaint regarding the corruption was filed. The tribunal found that the initiation of the investigation was likely motivated by a desire to prevail on its corruption defense in the arbitration. The tribunal concluded that Respondent’s “flagrant absence of any follow-up whatsoever given to the complaint,” along with its failure to produce evidence from the investigation, evinced that the respondent “gave priority to the grounds of the defense which corruption constitutes, rather than to the prosecution of the corrupted parties” (para. 221). This holding supports that tribunals may consider not just the findings of domestic investigations, but motivations and circumstances surrounding the launching of such investigations.

 

Conclusion

Arbitral corruption defenses are still largely uncharted terrain since there are very few awards dealing head on with such issues. As a result, how such defenses operate in tandem with parallel domestic investigations and proceedings is still a developing area of the law. The same is likely true in the context of international commercial arbitration – while corruption issues are frequently raised in commercial arbitrations and there are often issues with parallel domestic investigations, this is largely left obscured from the public domain (unlike investment arbitration awards, many of which are published).²⁾For example, in Concesionaria Ruta Del Sol S.A.S. v. Agencia Nacional de Infraestructura of Colombia (p. 491), the tribunal cited multiple other proceedings against Concesionaria, from U.S. criminal cases to Colombian criminal decisions. The tribunal found these domestic findings to be probative evidence of corruption and went so far as to expressly accord these findings full “probative value.” Considering the increase in the publication of international commercial arbitration awards in recent years (as part of the greater trend towards transparency in international commercial arbitration), more commercial awards dealing with corruption allegations may become known. The existing jurisprudence on corruption in investment arbitration set forth in this note will likely, in turn, inform party submissions in commercial arbitrations. These issues are therefore ones to watch, as parties more commonly invoke corruption defenses and countries begin more aggressively to enact and enforce anti-corruption laws and policies.

Notwithstanding the developing nature of this area of the law, it is clear that international arbitration tribunals do consider parallel domestic corruption findings—or the absence thereof—in adjudicating whether a party established a corruption defense. Accordingly, parties on both sides (invoking and defending) of corruption defenses should be proactive in tracking parallel domestic investigations into and proceedings regarding alleged corruption.

 

________________________

To make sure you do not miss out on regular updates from the Kluwer Arbitration Blog, please subscribe here. To submit a proposal for a blog post, please consult our Editorial Guidelines.