Today I will attempt to bring some clarity to American federal law concerning enforcement of international arbitration clauses by non-signatories.  I will discuss two recent cases, one in Texas and the other in New York, in each of which a non-signatory sought to compel a signatory to arbitrate claims on which the signatory had commenced litigation. 

In the Texas case, a US company (“Licensee”) had an arbitration agreement with a Dutch company (“Licensor”), contained in a technology license agreement. After Licensee rejected a takeover bid from Licensor, the US subsidiary of the Dutch company (“Licensor Sub”), according to Licensee, conspired with another US company, to poach Licensee’s largest customer. Licensee sued Licensor Sub in a Texas court; Licensor Sub removed the action under FAA Section 205 and moved to compel arbitration; and the motion was denied by a federal district judge in Houston.  (QPro, Inc. v. RTD Quality Services USA, Inc., 2011 U.S. Dist. LEXIS 438 (S.D. Tex. Jan. 4, 2011).  The US Fifth Circuit Court of Appeals, whose precedents governed, has recognized “equitable estoppel” as a basis for a non-signatory to enforce an arbitration clause against a signatory in two scenarios: first, when the signatory, in bringing an action against the non-signatory, must rely upon the contract containing the arbitration clause as the basis for its claims; second, where the signatory’s lawsuit alleges substantially interdependent and concerted action between the non-signatory defendant and the signatory.    

Why did the motion to compel arbitration fail in this case?  First, the cause of action had nothing to do with the license agreement containing the arbitration clause.  The suit alleged tortious interference with Licensee’s relationship with its largest customer, not with the technology license. Second, Licensor (non-signatory’s parent company) was not alleged to be a concerted actor in the tortious interference; no arbitration claim based on the same facts had been brought against Licensor; and Licensor was not named as a defendant. The complaint alleged that Licensor had been motivated to direct Licensee’s tortious conduct as retaliation for Licensee’s rejection of its takeover bid, but Licensor was not alleged to have participated in execution of the tortious interference scheme.

Fast backward (by approximately three weeks) to the New York case — and read on with caution, as your Arbitration Commentator was counsel for the movant/ prevailing party on the motion to compel arbitration.  On the eve of the hearing on the merits in a pending arbitration between my Chinese corporate client and its US customer, arising from a troubled software development project, the US customer filed suit in the federal district court in New York against the CEO and largest shareholder of my corporate client — a signatory only in his corporate capacity but not individually to the software development contract at issue in the arbitration.  The complaint against the CEO parroted the counterclaims made in the arbitration – alleging fraudulent inducement of the contract and tortious interference with the customer’s capital-raising efforts – and alleged that the CEO as the alter ego of the corporation was entirely responsible individually for the corporation’s prospective liability on the counterclaims.

The arbitral equitable estoppel principles in the jurisprudence of the US Second Circuit Court of Appeals, concerning when a non-signatory may compel a signatory to arbitrate, are not materially different from those of the Fifth Circuit discussed in the Texas case. But the plaintiff here sought to escape application of those principles by arguing (1) that the corporation’s alleged misconduct would be determined in the arbitration, and would collaterally estop the CEO; and (2) therefore, only the alter ego facts would have to be determined in the lawsuit, and these facts were independent of the software contract and independent of the corporation’s alleged misconduct.

The New York federal judge flatly rejected the second argument, finding persuasive the movant’s position that the alter ego claims were completely intertwined with the corporation’s conduct that formed the basis for the counterclaims (because under New York alter ego law, domination and control, and fraud or other misconduct by the dominating person, would have to be demonstrated in connection with transactions that caused the harm for which recovery was sought).   (Charity Folks, Inc. v. Kim, Civ. No. 1:10-cv-08765 (S.D.N.Y. Dec. 13, 2010), unpublished Memorandum Opinion obtainable with subscriber password on the Southern District of New York’s website, www.nysd.uscourts.gov, or by contacting this writer. I will attempt to establish a link here shortly).

These arbitral equitable estoppel principles seem fairly straightforward and not particularly difficult for courts to apply, once the facts and pleadings and procedural posture and claims in the related arbitration, if one is pending, are understood.  However, these rules do seem to have become somewhat disconnected from the principles of federal arbitration law from whence they sprung.  The main point is that the arbitration agreement justifies a reasonable expectation on the part of the each signatory that disputes with the counterparty signatory, commenced by that other signatory, and involving the subject matter of the contract, will be arbitrable, whether the dispute is framed as one involving the liability of the signatory, or that of the signatory’s parent, subsidiary, officer, director, employee, etc.  This is the element of reasonable reliance that underlies the “estoppel”  notion. Were the rule otherwise, the assumed benefits of agreeing to arbitration would be lost so long as the signatory claimant can fashion of theory of liability that involves the non-signatory.  Further, if the rule were otherwise, signatory parties who at the time the dispute arises would prefer to resolve the dispute using litigation procedures, would have large incentives to use the discovery tools available in litigation, obtain judgment, and then claim collateral estoppel effect in an ensuing arbitration against a signatory who was in privity with the litigation defendant.

Thus, arbitral equitable estoppel is the doctrine by which the rights gained from choosing arbitration are preserved against erosion threatened by “party-shopping” an arbitrable dispute into a judicial forum.        

Before we part ways in flight to family hearths, groomed pistes, and pristine sandy beaches nearer to the Equator, let us return briefly to one of the favorite topics of Arbitration Commentaries and its readers:  arbitrability with non-signatories, and specifically the choice-of-law governing arbitrability in regard to non-signatories.

A war story begins today’s installment. My Hong Kong client contracted to sell software to a New York company and in the contract provided for Hong Kong governing law and the resolution of disputes by arbitration. The contract named no arbitral institution, nor any rules, place of arbitration, or method to appoint arbitrators.  A commercial dispute arose, and an agreed solution to overcome the pathology of the arbitration clause was achieved: an Arbitration Submission Agreement that called for ICDR Arbitration in New York (with the witnesses from China permitted to appear by teleconference).  Ten days prior to the hearing on the merits, the US company, arbitral respondent and counterclaimant, filed suit against the Hong Kong-resident CEO of my client, in the Southern District of New York. I filed a motion to compel arbitration, on behalf of the CEO, relying upon US arbitrability law. The motion was granted in short order, on the grounds and under the law asserted in my motion.

But wait.  Did I lead the court astray?  Should US arbitrability law have been applied in preference to Hong Kong law? Or was this a “default” application of US law because my adversary chose to assume its applicability?

The murky state of US law on the question of what law applies to arbitrability determinations involving non-signatories in international arbitration cases, is reflected in an even more recent decision by a different judge in the Southern District of New York.  (FR8 Singapore Pte. Ltd. v. Albacore Maritime  Inc., 2010 U.S. Dist LEXIS 132212 (S.D.N.Y. Dec. 14, 2010)).   In Albacore, Plaintiff Singapore company brought the action (under Section 4 of the FAA) to require non-signatory defendants, Marshall Islands companies with offices in Greece, to arbitrate a dispute arising from the purchase and sale of a ship. Plaintiff alleged that the non-signatories were the real parties in interest, and that the signatory seller was merely a shell entity, an alter ego.  Plaintiff’s commercial contract with the signatory seller provided for arbitration in London and English governing law.  The question: What law should be applied to determine whether the alter ego allegations are sufficient to require the non-signatories to arbitrate at the insistence of the counterparty signatory: Marshall Islands law? English law? or US law?

To answer the question (and settle finally upon English law), the Court looked mainly to three Second Circuit decisions. In the first, Smith/Enron Cogeneration Limited Partnership, Inc. v. Smith Cogeneration Int’l, Inc., 198 F.3d 88 (2d Cir. 1999), the Second Circuit rejected the use of the forum’s choice-of-law principles (New York) to address the non-signatories issue, on the basis that such “parochialism” would “subvert the goal of simplifying and unifying international arbitration law.”  On this basis, the Albacore Court held that while New York choice-of-law principles would indicate applying the alter ego law of the place of incorporation of the allegedly dominated entity, that approach was improper and so application of Marshall Islands law could not be justified.

The second case addressed in Albacore was Motorola Credit Corp. v. Uzan, 388 F.3d 39 (2d Cir. 2004).  In Motorola, the US manufacturer sought to litigate its claims against the non-signatory Turkish shareholder of the Turkish company with which it had contracted, providing therein for Swiss law and arbitration in Switzerland under Zurich Chamber of Commerce Rules.  The Turkish non-signatory sought in Motorola’s US District Court action to compel arbitration under American equitable estoppel principles. He failed. The Second Circuit held that the “choice-of-law clause” in the contract had to be respected by the non-signatory defendant seeking to arbitrate rather than litigate, and that under Swiss arbitration law a non-signatory could not require the signatory to arbitrate.

Let us pause here, and return for a moment to my war story.  If the foreign shareholder could not invoke the arbitration clause in Motorola to compel arbitration, because of a choice of foreign law clause, why was I able to obtain an order compelling arbitration on behalf of my foreign shareholder-CEO client under American arbitral equitable estoppel principles despite a Hong Kong choice of law clause in the contract?   Were we merely the lucky beneficiaries of the choice-of-law issue having been bypassed?   I do not think so – because what the Second Circuit really meant in Motorola, and should have said much more clearly, was that it was giving effect to the contract’s choice of Swiss arbitration law, a choice made not by reason of the general selection of Swiss contract law, but rather by the lex arbitri choice that was implied-in-law from the agreement to arbitrate in Switzerland. In my case, the Arbitration Submission Agreement provided for ICDR Arbitration in New York, and therefore, despite the general Hong Kong choice of law clause in the underlying contract, US arbitration law was the law applicable to the arbitration clause my client sought to invoke. The fact that the party seeking to dodge arbitration and sue my Chinese client in a US court was the US party to the contract and the Submission Agreement is an additional, but I think unnecessary, reason to apply the contractual American lex arbitri – and indeed the nationality of the parties played no part in the analysis by the Second Circuit in Motorola.  

But the lack of precision in Motorola’s reasoning became a difficulty for the District Court in Albacore. The Court took at face value the Motorola Court’s statement that it was respecting the contract’s “choice-of law” clause.   And it found that holding to be in possible conflict with another Second Circuit decision, Sarhank Group v. Oracle Corp., 404 F.3d 657 (2d Cir. 2005).  Oracle, the US software maker, had set up a foreign subsidiary to sell in Egypt and through that subsidiary had contracted with Sarhank and subjected the contract to Egyptian contract law and arbitration in Egypt under Cairo Arbitration Centre Rules.  Sarhank named Oracle as a party to the arbitration, and the Tribunal over Oracle’s objections found it had jurisdiction over Oracle and entered an award against Oracle, which Sarhank then sought to have confirmed in New York under the New York Convention.  Oracle presented its non-arbitrability position as a basis to refuse confirmation under Article V of the Convention, and when the case reached the Second Circuit the Court decided to apply US not  Egyptian arbitrability law to decide the issue.  The Second Circuit reasoned that Oracle had set up the foreign subsidiary precisely to avoid exposing itself to adjudication in a foreign tribunal under foreign law, and so the arbitration agreement signed by its subsidiary furnished no basis to apply the contractual choice of Egyptian arbitration law, or to give deference to the Tribunal’s jurisdiction ruling.

Reading these cases, the District Court in Albacore asked aloud how it should resolve the seeming conflict – the contractual choice of arbitration law having been respected in Motorola, but cast aside in preference to US arbitrability law in Sarhank.   The Court chose to follow Motorola, and initially suggested it might distinguish Sarhank on the basis of its procedural posture, i.e. that the non-signatory was opposing confirmation under Article V of the Convention.  That distinction seems unconvincing however; there should be no reason for the US Court to reach one conclusion about arbitrability if the issue is raised on a motion to compel (or stay) arbitration, and another if it arises at the confirmation stage.  And the Court, evidently unsatisfied, went further in search of meaningful reconciliation. The Albacore Court examined whether Motorola and Sarhank could be persuasively distinguished on the basis that in the former, the non-signatory was seeking to enforce the arbitration clause against a signatory party bound by it, whereas in the latter it was the signatory seeking to extend the clause to a non-signatory That basis for distinguishing the cases had been adopted at least once before by Southern District of New York judge, in Republic of Ecuador v. ChevronTexaco Corp., 376 F. Supp.2d 334 (S.D.N.Y. 2005).  But this was found unsatisfactory, oversimplifying matters because the signatory-nonsignatory dichotomy fails to account for the possibility that a nonsignatory seeking to compel arbitration might seek the benefit of a contractual lex arbitri that is more liberal on arbitrability than US arbitration law, and a nonsignatory seeking to resist arbitration might seek the benefit of contractual lex arbitri that is more restrictive on arbitrability with nonsignatories than US law. A blanket rule that the non-signatory’s opposition to arbitration shall be governed by US law in a US court, the Court reasoned, would motivate the signatory seeking to compel arbitration for forum shop based on whether the US law or the contractual lex arbitri is more favorable to its position. Avoidance of such forum shopping, the Court held, is promoted by adhering to the Motorola approach – which  required the non-signatory alleged real party in interest who was resisting arbitration in a US Court to invoke the contractual lex arbitri.

Distillation of these cases into a handful of applicable principles is a formidable challenge.  But I shall try.  It seems that if a party is seeking relief to avoid arbitration or its consequences (whether by filing a lawsuit to litigate the merits and opposing a motion to compel arbitration, moving for an anti-arbitration stay/injunction, or seeking to vacate or prevent confirmation of an award), that party, whether signatory or non-signatory, must plead its case under the contractual lex arbitri if it has an involvement, prima facie, with the performance of the contract. (A parent company like Oracle in Sarhank, whose corporate veil is not sought to be pierced, therefore is not bound by its subsidiary’s contractual lex arbitri, but controlling shareholders like the defendants in Motorola and Albacore, alleged to be the dominant forces behind a “shell” contracting entities, are so bound).  And if a party is seeking relief to require arbitration or confirm the results of arbitration, whether signatory or non-signatory, it must invoke the contractual lex arbitri in support of its arbitrability position, as part and parcel of its invocation of the agreement to arbitrate.

We can reasonably expect the Second Circuit to revisit these issues and seek to achieve some analytical consistency in the not too distant future.  

The functus officio doctrine returned to center stage this week, with a new decision in a New York Convention case from the U.S. Fourth Circuit Court of Appeals. Here, a partial award determined that a contract had been breached, but the subsequent final award determined that the same contract was unenforceable. In the circumstances, the Court held, the final award was enforceable and confirmation of the partial award could be (and properly was) refused under the Convention by the district court. (AO Techsnabexport v. Globe Nuclear Servs. & Supply, Inc., 2010 U.S. App. LEXIS 25640 (4th Cir. Dec. 15, 2010)).

The arbitral procedure that led to this outcome was a bit out of the ordinary. At issue was a contract involving purchase and sale of uranium. The seller, a Russian company, was the appointee of the Russian Federation to manage its supplies of nuclear material. The buyer, incorporated in Maryland, was a self-styled trader in uranium used in the production of nuclear fuel.

At an early stage, the Tribunal was informed of U.S. criminal indictments related to alleged fraudulent acquisition of a controlling interest in the buyer company by former officials of the Russian Federation, and was told of a parallel criminal investigation in the Russian General Prosecutor’s Office. (The case report does not say, but one assumes, that the Russian sales agent concluded that the buyer was not a secure and trustworthy repository of nuclear materials, and terminated the contract for that reason).

The Tribunal elected to divide the arbitration into three phases: breach of contract liability, damages, and then potentially a third phase addressing the legality of the contract under the applicable Swedish law based on the evidence that might be developed in the Russian criminal investigation. After Phases One and Two, the Tribunal issued a partial award finding breach of contract and awarded nearly $1 billion of damages to the buyer. But in Phase Three the buyer submitted nearly 500 exhibits derived from the Russian criminal investigation, including extensive witness interview transcripts. Based on that evidence, the Tribunal concluded that the contract should be denied enforcement under a provision of the Swedish Contracts Act that in essence permitted rescission based on fraudulent inducement (here, evidently, a scheme to conceal the disreputable character of persons who had secretly conspired to obtain controlling interest in the buyer).

The Fourth Circuit rejected the argument that the Tribunal under the functus officio doctrine lacked power to revisit the conclusion in the partial award that the contract was enforceable. And its rationale for doing so makes good sense: the Tribunal from the outset had made clear that any determinations it might make, prior to any potential consideration of the Russian criminal investigation evidence, were subject to change if that evidence were later presented and it supported (as it did) the conclusion that the contract should be deemed not enforceable. It would not be appropriate in such a case, the Court held, to refuse to confirm the final award under Convention Article V(1)(c) based on the Tribunal having allegedly exceeded its powers.

The Court’s opinion does not permit arbitral procedure-watchers a clear understanding of why the Tribunal elected to proceed as it did. One can imagine the Tribunal had highly imperfect information about the status of the Russian Prosecutor’s work, its timing, its focus, or the merits of the charges under inquiry. One can imagine the Tribunal having been reluctant to stay its proceedings for an indeterminate time. Further, the issuance of the partial award appears to have been quite proper, as at that point it could not be foreseen that the Russian award-loser would be able to present, as it later did, such a compelling case of fraud based on the evidence gathered by the Prosecutor. Thus it appears that well-designed arbitral procedure combined with meticulous application of the Convention and the functus officio doctrine by the U.S. Court yielded a sensible solution to a complex problem increasingly faced by international arbitral tribunals — the arbitral adjudication of issues related to alleged criminality of participants in the arbitration.

The efforts of Chevron Corporation to obtain evidence under 28 U.S.C. § 1782 for use in litigation in Ecuador’s courts  and in a BIT arbitration against the Government of Ecuador has given U.S. District courts in a number of jurisdictions the opportunity to consider the discretionary factors affecting the availability of section 1782 discovery. Such discretionary factors were identified by the U.S. Supreme Court in the Intel case. (Intel Corp. v. Advanced Microdevices, Inc., 542 U.S. 241 (2004)).

Chevron is the defendant in private environmental litigation in Ecuador, and contends that the Government of Ecuador has improperly colluded with the Plaintiffs in that case. That collusion underlies Chevron’s BIT arbitration claims against Ecuador.

Federal district judges in Boston and Baltimore have granted Chevron permission in recent weeks to conduct depositions of damages experts who submitted reports for the Ecuador civil plaintiffs. (Chevron Corp. v. Sheffitz, 2010 U.S. Dist. LEXIS 129540 (D. Mass. Dec. 7, 2010); In re Chevron Corp., 2010 U.S. Dist. LEXIS 124897 (D. Md. Nov. 24 , 2010)).

Each court paused to consider whether the foreign court was receptive to the proposed discovery — and on which side the “receptivity” discretionary factor should impose a burden of persuasion. The Maryland judge was inclined to the view that the Respondent-witness bears the burden to demonstrate that the foreign tribunal is not receptive to the proposed evidence-gathering.  The Massachusetts judge took a more moderate view on this issue, finding that the burden is not clearly imposed on one party or the other. But each judge concluded that even though there was no clear indication that the Ecuador court wished to consider the proposed evidence, the fact that the Ecuador court stated that it was ethically-bound the make the evidence part of the official court file was sufficient to make the “receptivity” factor tip slightly in Chevron’s favor.

The two courts were more nearly consistent in the treatment of the third Intel discretionary factor: whether the proposed discovery would circumvent the procedures of the foreign tribunal. For the Maryland judge, it was sufficient to re-state the settled principle that the proposed evidence need not be discoverable in the foreign proceeding. The Massachusetts judge went a step further, to point out that the issue is properly viewed as one if bad faith, i.e. whether the evidence is sought in defiance of a preference of the foreign tribunal that the evidence not be obtained. 

In both cases, Chevron’s petitions were granted.

Observers of the relationship between the New York Convention and U.S. bankruptcy proceedings will take interest in a recent decision of a federal district court in Pittsburgh, in which the Court rejected the bankruptcy trustee’s attempt to collaterally attack an ICC arbitration award and the judgment entered on that award. G&G Investments, Inc. v. Buschmeier, 2010 U.S. Dist. LEXIS 125902 (W.D. Pa. Nov. 30, 2010).   The bankruptcy debtor in G&G Investments had: contracted to purchase a controlling interest in a German company; attempted to rescind and recover its initial payment; and then lost an ICC arbitration in which it was held fully liable to the German company for the purchase price. Judgment on the award was entered by the federal district court in Pittsburgh several years prior to the bankruptcy filing, in a confirmation proceeding under the New York Convention and Chapter Two of the Federal Arbitration Act in which allegations that the award had been procured by fraud on the tribunal were raised without success. In the bankruptcy case, the controlling shareholder of the German company filed a claim based on the unsatisfied judgment, and the trustee filed an objection to the claim, citing new evidence of fraud based on the report of a German private investigator the trustee had hired to review the documentary record in the arbitration and to interview certain of the witnesses.  The bankruptcy court refused to accept jurisdiction over the trustee’s objection, based on res judicata. That decision was upheld in this appeal to the District Court, which found no excuse for the debtor’s lack of diligence in gathering the evidence of fraud, and no reason to relief the trustee of the res judicata effect of the enforcement proceeding in which  the fraud on the tribunal claim had been raised to no avail.