As David Attanasio (Co-Chair of the ILA American Branch Investment Committee; Associate, Dechert) set out in his opening remarks, the conference addressed the resolution of corruption allegations in international investment arbitration following the Metal-Tech and Spentex awards. In the aftermath of those awards, the field of investment arbitration has had to grapple with a set of questions regarding the proof of corruption and response to findings of corruption. Those awards combined flexible evidentiary techniques for assessing corruption allegations with the outright dismissal of the arbitration upon finding corruption. The conference addressed whether and to what degree investment arbitration should follow such approaches to corruption allegations.
This blog post will discuss several (of many) important contributions from the conference, focusing on two principal threads: proof of corruption, and the proper response when corruption is found.
- What is sufficient proof of corruption?
The first panel, moderated by Susan D. Franck (Professor of Law, American University), focused on the question of the proof of corruption. This issue has become increasingly heated against the background of the Metal-Tech tribunal’s invocation of red flags to find corruption (see a previous discussion of Metal-Tech on the Blog) and the Spentex tribunal’s finding of corruption on the basis of “connecting the dots.”
To stimulate discussion, Prof. Franck put to the panelists a 2014 empirical study, carried out during the biennial Congress for the International Council for Commercial Arbitration (ICCA) and presented in the article “International Arbitration: Demographics, Precision and Justice”. The study concluded that practitioners consider that the burden of proof is frequently outcome determinative in international arbitrations, but that it is only occasionally or never identified in advance.
In this regard, Aloysius Llamzon (Senior Associate, King & Spalding) observed that the failure to identify the applicable standard of proof in advance is a common flaw in the adjudication of corruption allegations. Jason Yackee (Professor of Law, University of Wisconsin) too was of the opinion that parties should know what standard of proof they will be judged by, given that it may be outcome determinative.
There is a question, however, as to the degree to which knowing the standard of proof in advance would significantly alter party behavior, since parties might present whatever evidence they have in order to support their allegations regardless. This is true even if the standard of proof would be relevant to the tribunal’s analysis of the case.
Nevertheless, the panel reported that the field has splintered in its views on the applicable standard of proof for such allegations. One division, highlighted by Prof. Yackee, is the difference between the civil law standard of proof of intimate conviction and the probabilistic standards used in the US. A second division, noted by Mr. Llamzon, is that regarding the stringency of the standards of proof, where there are two main camps: one advocating a higher standard, versus the other advocating an ordinary standard of proof.
As Mr. Llamzon observed, much of the difference is derived from national conceptions of fraud and corruption. In his view, these different national conceptions are likely to lead to disagreement regarding the standard to adopt when a tribunal is comprised of arbitrators from both civil and common law traditions.
Prof. Yackee observed that, when a probabilistic standard of proof is employed, one analytic approach to setting the standard is to compare the costs of a false positive (i.e., an erroneous finding of corruption) with the costs of a false negative (i.e., an erroneous finding against corruption). In this regard, a higher standard of proof may be required if the costs of a false positive (for example, the denial of the forum) are considered to be higher than the costs of a false negative.
A major further question is whether the applicable standard of proof can be satisfied by identifying so-called red flags of corruption—an increasingly common tactic by parties following the Metal-Tech award. Mr. Llamzon took a skeptical view, observing that the concept of red flags comes from the world of compliance where it is used to assess, ex ante, the risks of entering into an agreement with a third party, not for the evidentiary purpose of assessing, ex post, the existence of corruption. By contrast, in Prof. Yackee’s view, red flags of corruption could go into the “bucket” of evidence, albeit taking into account the specific evidentiary weight of a given red flag.
Nevertheless, the question remains as to whether red flags should in fact constitute evidence and how strong that evidence might be. This is a question that tribunals will continue to confront in light of the contrast between the seriousness of allegations of corruption, on the one hand, and the limitations on the tribunal’s evidence-gathering powers, on the other.
Meriam Al-Rashid (Partner, Dentons) noted that, whatever standard of proof is ultimately adopted, in accordance with the principle of equality of arms, arbitrators have a duty to apply the same standard of proof to allegations of corruption made by an investor against the state as it applies to allegations of corruption made by the state against an investor.
- What is the right response when corruption is found?
The second panel, moderated by Jan Paulsson (Professor of Law, University of Miami School of Law; Partner, Three Crowns), addressed the appropriate response from an investment tribunal following findings of corruption. This issue too has become increasingly challenging given that some investment tribunals are inclined to take a more flexible evidentiary approach to finding corruption and it is increasingly recognized that “it takes two to tango”—i.e., alleged corruption often involves both the state and the investor.
The panel had doubts as to whether a binary response to corruption—i.e., either ignore the corruption or dismiss the arbitration entirely—is appropriate. Lucinda Low (Partner, Steptoe) noted that the binary response incentivizes the respondent state not to investigate allegations of corruption.
Arif H. Ali (Partner, Dechert) considered the binary response problematic at its core: a tribunal’s role is not to mete out punishment for corruption. However, according to him, refusing to address the legality or the economics of the situation on the moral grounds that some tribunals have invoked is an abdication of the arbitrator’s function. This could be the case, for example, when the tribunal relies on standards of public policy as did the tribunal in World Duty Free.
Further, Mr. Ali noted that arbitrators usually do not examine questions of fact in the same detail and depth as, for example, domestic courts do in a criminal trial, and the procedural forum is far too limited in its evidence-gathering to accommodate the evidentiary challenges of corruption allegations.
Two potential alternatives to the binary response emerged from the panel.
Mr. Ali proposed that the concept of contributory fault could be employed to balance the pertinent considerations of law, morality, and economics. The corruption could be taken into account (if relevant) in determining the compensation due to the investor for the state action at issue in the investment arbitration. In this case, the compensation could be reduced based on the investor’s contribution to its own loss through its participation in the corruption. This is an approach that some investment tribunals, such as the MTD tribunal, have adopted, albeit not in connection with corruption.
By contrast, Ms. Low set out—albeit for provisional consideration only—a proportionality approach. Such an approach might ensure that the state has proper incentives to eliminate corruption, consistent with obligations assumed under international anti-corruption treaties. Under this approach, tribunals would look to a set of relevant factors to determine the appropriate remedy for its findings of corruption, but would not automatically dismiss the arbitration simply because corruption is found. Among the factors that a tribunal might consider for this purpose:
- Was the public sector involved in the investment or the corruption?
- Did the investor freely offer the alleged bribe, or did a host state official extort it from the investor?
- Did both the investor and the state comply with their obligations to prevent or investigate the corruption?
A third new option, suggested by these panelist comments, could be to apply a merged version of these two approaches. For example, tribunals might consider some of the factors identified by Ms. Low in order to determine each party’s fault in the case and, thus, the compensation that should be awarded to the investor.
The conference concluded with closing remarks from Malika Aggarwal (Georgetown International Arbitration Society).