The 30th Annual ITA Workshop on Multiple Proceedings, Multiple Parties, and International Arbitration: What a Tangled Web We Weave, took place in Dallas, Texas on 20-22 June 2018. Co-chairing were Erica Stein (Dechert), Jean-Christophe Honlet (Dentons), and Frédéric G. Sourgens (Washburn University). The workshop, a lead event of the ITA, was dedicated to an in-depth exploration of the increasing procedural complexity in contemporary international dispute resolution.
The keynote speech from Prof. Emmanuel Gaillard (Shearman & Sterling, Institut d’études politiques de Paris) identified the major problem for the workshop: how to control the potential chaos resulting from a system where multiple proceedings regularly arise from the same matter. This blog post will describe several (of many) important contributions from the ITA Workshop, focusing on two principal threads: how arbitrators may coordinate multiple proceedings arising from the same facts and how domestic courts may coordinate multiple set aside and enforcement proceedings arising from the same award.
- How to Coordinate Multiple Arbitral Proceedings
One of, if not the, preeminent issues of the workshop was how to coordinate multiple proceedings brought by related entities concerning the same basic facts. This issue frequently arises in commercial and investment arbitration, when disputes arising from the same or related facts are submitted to multiple fora.
Prof. Hanno Wehland (Lenz & Staehelin) captured the basic problem of multiple proceedings as it pertains to investment arbitration: international investment agreements often extend protection to both direct and indirect investors, so multiple entities in a single corporate structure are often protected investors (and potential claimants) for the same investment. Such circumstances arose in, among others, the well-known Lauder v. Czech Republic and CME v. Czech Republic cases and the recent Orascom TMT v. Algeria and Orascom Telecom v. Algeria cases. And, in addition to multiple entities qualifying as investors, one or more may also be parties to contracts concerning the investment, and eligible to bring arbitration on that basis as well.
This potential for multiple claims by related companies concerning the same facts creates serious risks of double recovery and of repeat litigation. The workshop focused extensively on whether and how arbitral tribunals can coordinate or manage these risks.
The Traditional Doctrine of Res Judicata
Perhaps the most traditional response to the problem is to analyze it through the lens of res judicata (or related doctrines). Prof. Chiara Giorgetti (University of Richmond) observed that the application of res judicata has the potential to avoid repetition and strengthen the rule of law in the face of repeat proceedings.
However, against the background of multiple arbitrations under different instruments, a tribunal may have to answer a number of difficult questions when invoking the doctrine of res judicata.
What is the governing law for res judicata? It could be the law of the first arbitration’s seat, that of the second, the national law of one or the other of the parties, or transnational principles. Prof. Gaillard proposed that transnational principles are most appropriately applied, even if they are not fully developed. However, Prof. Pierre Mayer emphasized the lack of any transnational consensus on res judicata, as common law and civil law have divergent principles.
What is the scope of res judicata? The workshop participants evinced agreement that res judicata indeed applies among strictly identical claims (per the traditional triple identity test). However, beyond that, there were divergent views on the appropriate scope for res judicata.
Prof. Gaillard noted, without endorsement, that the Apotex Holding v. USA tribunal considered that issue estoppel had sufficient recognition to be a genuinely transnational principle. This would give binding effect not only to an award’s dispositive but also to the legal and factual reasoning supporting that dispositive. Prof. Giorgetti went a step further and endorsed the broad Apotex approach as better calibrated to the objectives of avoiding repetition and coordinating multiple arbitral proceedings.
By contrast, Prof. Mayer took a more skeptical view. While he accepted that the dispositive part of an award is binding on subsequent tribunals, he proposed that the nature of arbitration warrants against granting any binding effect to an award’s reasoning in a subsequent arbitration. The cornerstone of arbitration, in his view, is the parties’ ability to appoint at least one member of the arbitral tribunal, so those tribunals should not abdicate their power of judgment on the dispute entrusted to them.
A Turn toward Abuse of Process?
A more recent response to repeat proceedings is to analyze them for abuse of process, following the model of Orascom TMT. Prof. Gaillard, who acted as respondent’s counsel in that arbitration, recommended this approach in his keynote speech. He emphasized that the Orascom TMT decision shows that tribunals can look at the effects on the arbitral system as a whole and not solely on their particular dispute in isolation.
The Orascom TMT tribunal (Gabrielle Kaufmann-Kohler presiding) found the submitted claims inadmissible on the grounds of an abuse of process. Three similarly-framed disputes, all concerning the same state actions, had been brought at multiple levels of a single corporate chain. Under the facts of the dispute, the tribunal reasoned that the claimant could not claim for the same harm as in the other arbitrations and, indeed, committed an abuse of right because it sought “to impugn the same host state measures and claims for the same harm at various levels of the chain in reliance on several investment treaties concluded by the host state.”
Ricky Diwan QC (Essex Court Chambers) predicted that the Orascom TMT award will increase future scrutiny of the Lauder and CME approach to the issue of multiple proceedings. The well-known Lauder and CME disputes concerned the same underlying factual allegations but were brought by different entities in the same corporate chain under distinct investment treaties. Both the Lauder and the CME tribunals rejected the relevance of the parallel arbitration on the grounds that they concerned formally different parties bringing different causes of action (i.e., under different investment treaties). In doing so, both tribunals specifically rejected any relevance of the fact that the two claims were for the same injury and both concluded that there was no abuse of process.
Although Orascom TMT identifies a new solution to the problem of multiple proceedings, Mr. Diwan identified multiple open questions that tribunals will have to resolve when following that example.
- First, at what moment does it become abusive to bring parallel disputes? Is it at the time when both claims are initiated or when both tribunals find that they have jurisdiction? If, possibly only one tribunal has jurisdiction, then the parallel proceedings might seek to ensure nothing more than that the dispute may find a competent forum.
- Second, to what degree must two entities be related for parallel disputes to become abusive? Diwan observed that, in Eskosol v. Italy, both an 80% shareholder and the subsidiary company commenced proceedings against Italy. The Eskosol tribunal concluded that the interests of the shareholder and the company were not fully aligned, so the parallel proceedings were not abusive.
Identifying the Investor through its Active Investment
Instead of coordinating multiple potential proceedings by related entities through procedural approaches like res judicata or abuse of process, it may be possible to select the proper claimant on substantive grounds. One approach sometimes adopted in investment arbitration is to identify the proper claimant as the legal entity in a corporate structure with an active investment role.
As Prof. Ursula Kriebaum (University of Vienna) observed, several tribunals have picked out the proper claimant in a multilevel company structure on such grounds. She noted that, among others, Standard Chartered Bank v. Tanzania, Caratube v. Kazakhstan, and KT Asia v. Kazakhstan had followed such an approach. Standard Chartered Bank, for example, stated that, to benefit from the investment treaty’s arbitration provision, “a claimant must demonstrate that the investment was made at the claimant’s direction, that the claimant funded the investment or that the claimant controlled the investment in an active and direct manner.”
Prof. Kriebaum critiqued this approach to the problem of multiple investors. She observed that an active investment requirement would deny protection to an investment if it is ever transferred, as the recipient company would not have made an active investment in the host state.
However, it is unclear whether Prof. Kriebaum intended to apply this critique to all the elements of an active investment requirement. Standard Chartered Bank proposed several alternative ways to have an active investment. If active investment requires making the investment at the claimant’s direction, then perhaps the purchase of an investment would not suffice. But, if active investment requires only that the relevant entity actively and directly control the investment, it is less obvious that the failure to make the initial investment would be disqualifying.
Assisting Arbitrators with Treaty Provisions
Treaty provisions, both present and future, may also assist arbitrators in coordinating repeat proceedings by precluding such proceedings. Jean Kalicki proposed that a new generation of treaties would be advisable to provide further assistance with the problem of consolidating proceedings.
However, as Prof. Gaillard noted, we have already seen four generations of treaty provisions designed to address the problem. The first generation established a fork in the road clause, where the choice of one forum precludes the submission of the same dispute to any other forum. The second generation, exemplified by NAFTA, required both the foreign investor and any local operating company to waive the right to submit the same dispute elsewhere. The third generation added a prohibition on parallel claims by the local operating company as well as by both direct and indirect shareholders.
The fourth, current, generation has sought to require waiver from entities both below and above the level of the investor in the corporate chain. The objective is to preclude comprehensively parallel claims. However, in Prof. Gaillard’s view, it is unclear how an investor could force its shareholders to waive potential claims, given that the investor does not have control over its shareholders. He did not address whether the decision to waive could be left to shareholder discretion discretion (as opposed to being placed under investor control); the shareholders could, potentially, voluntarily elect to waive if they wish the arbitration to proceed.
- How to Coordinate Multiple Centers of Control
After an award is rendered, the struggle to enforce that award is often fought across multiple court proceedings in multiple jurisdictions. These courts can potentially arrive at inconsistent results, or at least results that are in tension with one another. Thus, a number of participants focused on whether and when a court should defer to a prior court judgment on an award from a different jurisdiction. The central issue is which courts should exercise control over arbitral awards. As Mr. Diwan observed, courts could opt for sole control in the place of enforcement, shared control between the place of enforcement and the place of the seat, or sole control in the place of the seat.
Deference to Enforcement Judgments
Prof. Gaillard described a new trend of courts deferring to prior enforcement judgments, rather than making an independent evaluation of the award. He drew attention to examples from the U.S. and the U.K. where courts considering set aside or enforcement suits have deferred to foreign enforcement judgments. (Yasmine Lahlou (Chaffetz Lindsey) later suggested that these cases are not representative in the U.S.)
It was Prof. Gaillard’s view that this trend of deference to enforcement judgments is a recipe for chaos. He considers that it is contrary to the fundamental assumptions of the New York Convention. This treaty contemplated that each court would independently consider the award and reach its decision, regardless of the views of neighboring courts.
Mr. Diwan agreed that it is a mistake for courts to defer to foreign enforcement judgments. He asked why, for example, an English court should defer to an Austrian court judgment that applies an incorrect analysis under the New York Convention? He suggested that this is to apply English private law principles concerning foreign judgments to a subject for which they are not suited. Ms. Lahlou echoed this sentiment that there is no good reason to apply general legal principles concerning foreign judgments to ancillary enforcement judgments.
Deference to Set Aside Judgments
Mr. Diwan raised the further question of whether enforcing courts should defer to judgments from the courts of the seat on set aside actions. He observed that the English courts have adopted a policy of deference on the basis of private law principles, where that judgment is given effect unless it violates principles of honesty, natural justice, or public policy. This approach, in Mr. Diwan’s view, is contrary to the New York Convention’s basic policy that sole control of the award rests at the place of enforcement.
As Ms. Lahlou noted, U.S. courts similarly will not enforce awards set aside at the seat except in extraordinary circumstances.
This attempt to create a new system of coordination, as Mr. Diwan argued, could lead to chaos. The courts of a single state cannot sensibly attempt to create a system of coordination on a unilateral basis and it is problematic to seek a new consensus among the many member states of the New York Convention. The legal principles underlying the English approach, most notably that of issue estoppel are not accepted in civil law countries; so the French courts, for example, will not follow the same approach as the English courts.
Prof. Gaillard added that the English approach recreates the very requirement of double exequatur for award enforcement that the New York Convention was designed to eliminate. He considered that control over the award has to remain with the enforcement courts. Thus, he believes that the exercise of control over the award elsewhere should be suppressed for the sake of order.
In response, Prof. Mayer took the view that the judgment of the set aside court should indeed have effect. If the award is set aside, then a second arbitration proceeding will often produce a second award. By failing to give effect to the judgment setting aside the first award, a court may well recognize that award and refuse to recognize the second award. But the state of the seat and practically all other states would consider the second award to be the only valid award.
The ITA Workshop did not provide definitive answers to the questions of how to coordinate the complex proceedings that emerge in modern arbitral practice. But it identified out the chief issues that confront arbitrators and practitioners and provided key direction for the further evolution of the system.