As has been explored on this blog and elsewhere, corruption is a controversial topic in investor-state arbitration disputes. First emerging as a defense by states seeking to avoid liability, multiple tribunals have refused to enforce arbitration contracts tainted by corruption (see World Duty Free v Kenya and Plama Consortium v Bulgaria). Corruption has also been used as a cause of action by investors claiming unfair treatment (see Yukos v Russian Federation and here). The unclear incentive effects of corruption in arbitration proceedings have been analyzed from different angles—whether it provides countries with perverse incentives that might encourage corruption or instead buttresses anticorruption principles and promotes accountability.
Unfortunately, less attention has been paid to the procedural step at which tribunals discuss corruption. In the past ten years, an increasing number of tribunals are evaluating evidence of corruption at the jurisdictional stage of arbitration rather than at the merits stage. Those readers who are not lawyers (and even those who are), may be wondering, “Who cares? Why does it matter if corruption is treated as a ‘jurisdictional’ issue as opposed to a ‘merits’ issue?”
Actually, it matters a lot.
To understand why it matters, some preliminary explanation of the jurisdiction-merits distinction may be helpful. In general, when an investor wants to bring a claim before an international arbitral tribunal for violation of the terms of a bilateral investment treaty (BIT), the claimant must first convince the arbitral panel that the panel has “jurisdiction” over the dispute. In order for the panel to have jurisdiction, the aggrieved investor must show that the dispute concerns an “investment” as defined by the BIT. This makes sense in the abstract – you don’t want an international arbitral tribunal adjudicating issues that it has not been granted the power to adjudicate. Historically, panels simply compared the characteristics of the property or contract at issue with those listed under the investment definition in the BIT and called it a day. But in a series of unexpected rulings, arbitral panels have recently begun to reinterpret standard BIT language—so-called “Accordance Clauses” that define the “investments” over which the tribunal has jurisdiction as investments “in accordance with host state laws”—to mean that investments made in violation of the host state’s laws (or even in violation of “international law norms”) do not qualify as “investments” under the BIT. This in turn means that the tribunal has no jurisdiction over the dispute (see Niko Resources v Bangladesh). Recently, arbitration panels have gone even further and found a lack of jurisdiction for an “illegal investment” in cases where there was no Accordance Clause (see Inceysa v El Salvador).
Treating corruption as a jurisdictional issue has three negative consequences:
- Treating corruption in the formation of the investment contract as a jurisdictional bar, rather than a possible defense at the merits stage, denies the investor the opportunity to be heard in a neutral forum (the very reason for an investor-state provision in a BIT). Investors value access to arbitration very highly, and the cursory dismissal of their claims on the basis of judicial economy (as in cases like Phoenix v. Czech Republic) undermines the investment incentives provided by BITs. Particularly in countries where it is impossible to make an investment without engaging in corruption, the exclusion of any “unclean” investments from arbitral protection drastically narrows the coverage of a BIT. On the one hand, this might be a good thing if it sufficiently increases the costs of corruption for countries seeking foreign direct investment (FDI) that it encourages them to crack down on corruption. On the other hand, though, unilaterally decreasing the protections available to investors who are forced to engage in corruption may result only in a net reduction in FDI.
- Jurisdictional analysis of corruption is unfair because although a state actor is necessarily the counter-party to the corrupt act, the state actor unilaterally benefits from analyzing corruption at the jurisdictional stage. Unlike at the merits stage, where there is an opportunity for the investor to argue that the government should be estopped (precluded) from using corruption as a defense to liability, denying jurisdiction over the dispute results in a total victory for the host state. Although some panels have treated corruption as an absolute defense even at the merits stage (as in World Duty Free), others have been willing to embrace estoppel arguments in similar contexts (see Desert Projects v Yemen). Moreover, analyzing the issue at the jurisdictional stage precludes the possibility of evaluating comparative fault when it comes to damages – an important and highly contested area in traditional investor-state arbitration that has been relevant to recent disputes involving corruption (see, for example, the 25% reduction in damages awarded to the claimants in Yukos).
- Dismissing the dispute at an earlier stage of the arbitration proceedings might preclude the fact-finding necessary to paint a more complete picture of responsibility between the parties. Jurisdictional decision-making lacks the necessary tools to ensure thorough and even-handed analysis (such as discovery, cross-examination, and so forth). This is important because one of the goals of investor-state arbitration, like any adjudicative process, is to get the full story, figure out what went wrong, and determine who is responsible.
If panels are unwilling to forgo Accordance Clause analysis, or more generally unwilling to cabin their jurisdictional analysis, another partial solution to these problems is for arbiters to increase the use of two doctrines:
- The Minor Breaches Exception: This doctrine has been used by a few arbitral panels to avoid jurisdictionally excluding an investment “on the basis of minor errors” (see Tokios Tokeles v Ukraine). Although it has not been used in the context of corruption yet, minor acts of corruption might be relevant at the merits stage (say, for comparative fault determinations) but should not be preclusive at the jurisdictional stage.
- The Prima Facie Corruption Requirement: At the jurisdictional stage, one tribunal has required “a high standard of proof” for corruption allegations (see EDF (Services) Limited v Romania). Because the jurisdictional inquiry is a threshold question, the standard of proof should be high and allegations of corruption should only be dispositive if there is clear and uncontested evidence that an investment was conceived corruptly.
Ultimately, evidence of corruption should be analyzed like any other fact-dependent component of arbitration – at the merits stage. This affords host states the opportunity to raise corruption as a defense (and leverage the compelling public policy arguments against enforceability) while permitting investors to argue that a facially enforceable contract formed after some corrupt acts is not void, but merely voidable. Equitable considerations under these circumstances are important: Did the state fail to prosecute the alleged corruption? Were admissions made by either party? Did the investor engage in bad faith behavior? To what extent should the panel attribute constructive knowledge of the corrupt acts to the host state? If both parties engaged in corruption, should neither be able to seek redress (through arbitration, expropriation, etc.)? Because all of these questions merit the panel’s full attention, rather than investigate at the jurisdictional stage whether an investment was tainted by corruption, saving this analysis for the merits stage of a proceeding makes sense for both legal and practical reasons.