“International comity” is the term that first springs to mind when the subject under discussion is a foreign anti-suit injunctions (which, for purposes of this commentary, means an order from a US court enjoining a party before it from proceeding with foreign litigation).  Numerous US federal appellate cases admonish district courts to permit foreign anti-suit injunctions sparingly because they interfere, indirectly but tangibly, with the exercise of jurisdiction by the judiciary of a foreign sovereign state.

What should become of that admonition, and what remains of respect for the sovereignty of a foreign state’s judiciary, when the foreign litigation involves a dispute that the US court — enforcing the commands of Article II (3) of the New York Convention – decides must be resolved in arbitration?  

A decision last week by a federal district judge in Manhattan granting a pro-arbitration foreign anti-suit injunction against proceedings in Brazil. suggests upon close analysis that foreign anti-suit injunctions in arbitrable disputes are to be granted almost as a matter of course in view of the strong federal policy favoring arbitration of international commercial disputes. (Stolt Tankers BV v. Allianz Seguros, S.A., 2011 WL 2436662 (S.D.N.Y. June 16, 2011)).  But even granted the importance of this pro-arbitration policy, does not “comity” require that, before granting the injunction, there should be at least some indication that the foreign court, and not the litigants alone, is complicit in violation of that policy? And should not “comity” require that, before granting the injunction, the US court should be satisfied that it would be futile for the movant to present the injunction issue to the foreign court itself, or that there would be irreparable injury to the movant from the proceedings in the foreign court before the time when the foreign court could decide whether or not to stay its own proceedings?

In Stolt Tankers, the contract (charter party) between ship owner and charterer provided for arbitration of disputes in New York.  The charter party was incorporated by reference in bills of lading between the ship owner Stolt and the consignees in Brazil of certain shipments of caustic soda.  The goods were damaged in transit; the consignees obtained insurance recoveries; and the insurers based on subrogation brought suit against Stolt in Brazil.  Stolt then demanded arbitration, but the insurers refused to nominate an arbitrator. Stolt then petitioned the federal court to compel arbitration and to enjoin the Brazil litigation.

Only the Court’s grounds for granting the anti-suit injunction concern us here: first, that the Brazil lawsuit would “frustrate the general federal policy of promoting arbitration”; second, that the Brazilian court would not apply the Carriage of Goods by Sea Act, so that the outcomes “could be inconsistent”; third, equitable considerations such as deterring forum shopping favor the injunction; and fourth “adjudication of the same issues in two separate actions would result in inconvenience, inconsistency, and a possible race to judgment.”

But there is nothing in the Court’s rationale that persuasively justifies the conclusion that the proper forum to decide a motion to enjoin or stay the foreign litigation should in the first instance have been the US rather than the foreign court.  Brazil as a party to the New York Convention has an international obligation toward the United States and other member States to enforce written arbitration agreements, and nothing in the record here suggests that Brazil’s courts intended to violate those obligations.  And if there was no imminent stage of the Brazil proceedings in which Stolt’s rights to arbitral determination according to arbitral procedure would be impaired, would it not have been more sound to deny the injunction without prejudice, and to direct the movant to present the order compelling arbitration to the foreign court in support of an application to stay or dismiss? If that application were the made and denied, or not decided, by the foreign court, the granting of the anti-suit injunction by the US Court would then be supported by the very convincing reason that Brazil would be in violation of its international obligations under the New York Convention.  To deny Brazil’s courts the opportunity to “do the right thing,” does in the circumstances seem to offend international comity more than necessary; foregoing this intermediate step where there is no clear indication that the foreign court will disregard the arbitration clause arguably fuels a perception abroad of US courts as imperialist over-extenders of their judicial hegemony.     

The Stolt Tankers  decision may be usefully contrasted with another pro-arbitration foreign anti-suit injunction case in New York last year:  Amaprop Ltd. v. Indiabulls Financial Services Ltd., 2010 WL 1050988 (S.D.N.Y. Mar. 23, 2010).  In Amaprop, the record before the US Court that supported issuance of an anti-suit injunction included, crucially, the fact that the party seeking to litigate in India rather than arbitrate in New York had obtained an anti-arbitration injunction, initially on an ex parte basis, from a first instance court in Mumbai.  That party was also guilty of some duplicity, having represented to the ICDR that it intended to participate fully in the arbitration, when in fact it was in the process of obtaining the Mumbai anti-arbitration injunction. There was also expert evidence that civil litigation in India’s courts can be expected to continue for a decade or more without resolution. But the critical element, I submit, was that the foreign court was part of the problem, and therefore could not reasonably be expected to be part of the solution.  India, also a member State of the New York Convention, was in violation of its international obligations under that treaty, and therefore there was no reason for the US court first to direct the movant to take the order compelling arbitration to the Mumbai court and to ask that court to stay or dismiss its own proceedings.

It will be useful here briefly to juxtapose this sampling of recent US anti-suit injunction case law with the situation in Europe. (There is now a vast literature on this. My summary draws only on an article, recently published in the American Review of International Arbitration Volume 21, by the French jurist Dominique Hascher.  Any misunderstandings resulting from my lack of grounding in EU law are apologized for, in advance, and are certainly not attributable to Judge Hascher). The European position starts with a famous case called West Tankers, whose facts were not very different from Stolt Tankers, except that the subrogee insurers (including in both instances Allianz) sought to litigate in Italy and the pro-arbitration foreign anti-suit injunction was obtained from the High Court in London. On appeal, the House of Lords asked the European Court of Justice for an advisory ruling on whether the High Court’s order was consistent with Brussels Regulation 44 of 2001, which generally prohibits anti-suit injunctions from one EU Member State’s courts that impact proceedings in courts of another EU Member State.  The ECJ ruled that the High Court’s injunction was improper, and that it was an interference with the Italian court’s right to decide whether its jurisdiction was ousted or affected by the alleged agreement to arbitrate.

Comity, I suggest, should lead to the same result, presumptively but not categorically, between the United States and other countries that are member States of the New York Convention, even without a formal regulated union such as exists among the European countries. There should a presumption that the foreign court in which the litigation is pending has the right to consider in the first instance the effect of the arbitration agreement on its own jurisdiction (but not to reconsider the validity or existence of the arbitration agreement if the US court properly had jurisdiction to decide that question). That presumption may be rebutted by evidence that the foreign court has ignored or intends to ignore an agreement to arbitrate that is clearly sufficient, under the New York Convention, to impose on the foreign court an obligation to refer the parties to arbitration.  This is an approach that strikes a reasonable balance between American arbitration policy and respect for the jurisdictional sovereignty of foreign courts. 

Dear Readers

It is not usually the function of Arbitration Commentaries merely to report news of the day, but while a new post on anti-suit injunctions in aid of international arbitration is in preparation, I invite your attention to an article in today’s New York Times reporting that a disbarred former lawyer, having completed his prison term for paying kickbacks to secure clients, is seeking engagements as an arbitrator and mediator.   Read further at www.lexmarc.us under Legal Developments, where there is a link to the article itself, and if you wish, post your comments here. 

Warm Regards.

Marc Goldstein

Dear Readers:

On the general website of my law practice, www.lexmarc.us, you will find an essay on recent developments in US litigation under the Alien Tort Statute (“ATS”). 

MJG

A recent US Second Circuit Court of Appeals decision re-affirmed some well-settled principles about the appealability of district court orders dealing with arbitration. The Court held that an order of the district court refusing to enjoin a pending arbitration was not appealable while the arbitration was in progress. (Accenture, Inc. v. Spreng, 2011 U.S. App. LEXIS 10933 (2d Cir. May 27, 2011)). But what seems more interesting about the case, for the practicing arbitrator, is the arbitral procedural order that led to this costly spasm of collateral litigation.

Accenture was named a Respondent in a breach of contract arbitration by the former owner of a company Accenture had purchased – the owner having been given an employment contract that Accenture later terminated. The case was on an accelerated timetable, commenced in June 2010 and scheduled for merits hearings beginning October 19, 2010. In mid-September, Claimant received documents from Accenture, responsive to discovery requests, that led Claimant to seek permission to add a claim for fraudulent inducement of the contract. But Claimant made his request to amend only one week before the first hearing date, and the arbitrator denied leave to amend.

Claimant responded by filing with the AAA a “without prejudice” withdrawal of the pending arbitration, and a new arbitration demand encompassing the original contract claim and the new fraud claim. Accenture asked the arbitrator in the first case to determine that the withdrawal of claim was “with prejudice,” but was informed by the AAA that the arbitrator was functus officio. That decision launched Accenture’s litigation initiative, seeking to enjoin the second arbitration in favor of the first.

The case report does not indicate whether the second arbitration progressed during the litigation. But whereas the district court decided the motion to enjoin on December 23, 2010 and the Second Circuit agreed to hear and decide the case on an accelerated timetable (while denying a stay of the arbitration pending appeal), it seems fair to assume that arbitration #2 was on hold, whether by stipulation or voluntary decision of the arbitrator or indulgence of the AAA, until the litigation ran its course.

Looking back on this history today, and considering that seven months were lost to litigation that resolved no aspect of the pending dispute, it seems fair to ask whether the arbitrator might have made a more prudent decision. Claimant’s willingness to start anew in order to arbitrate the fraud and contract claims in one proceeding should have been anticipated. If the arbitrator was not aware that the AAA would permit “without prejudice” withdrawal and recommencement in the circumstances, it would seem that sound case management by the AAA and the arbitrator should have led to the arbitrator being so informed. Further, Accenture’s responses, before the AAA and the courts, were also foreseeable.

An order granting leave to amend obviously would have required postponement of then-imminent hearings and a new round of discovery and pre-hearing submissions. But if Claimant deserved to suffer some procedural adverse consequence for having (apparently) been dilatory in proposing the new fraud claim — notably allowing nearly four weeks to elapse after obtaining the critical new documents, a period that ended only a week before the hearing date — perhaps the better approach, consistent with the objectives to optimize decision costs and streamline proceedings, would have been to grant leave to amend, to require completion of pre-hearing activity on the fraud claim on a timetable that placed burdens on the parties commensurate with their respective responsibility for the late emergence of the new claim, and to reschedule the merits hearings perhaps 60 or 90 days forward.

It might be counter-argued that in a civil litigation in the same posture, most judges would deny leave to amend a pleading one week before the trial date. But to approve of the arbitrator’s decision on this basis overlooks two fundamental differences between arbitration and civil litigation. First, litigation consumes scarce public resources (judges, clerks, courtrooms, etc.), and rules and judicial doctrines concerning amendments, withdrawal of proceedings, etc. reflect to a significant extent concerns about the systemic, public costs of ill-timed strategic corrections by the parties. An arbitrator hired by the parties and being well-paid for her time has far less reason to consider her own inconvenience resulting from a belated proposed schedule change. Second, judicial rules for pleadings, discovery, pre-trial orders, etc. are relatively inflexible due to the sheer volume of cases the courts must process. The arbitrator’s discretion to short-circuit formalities of pleadings and to truncate pre-hearing proceedings is, within the confines of due process, one of the principal strengths of arbitral procedure.

Much of the foregoing will seem obvious to the experienced arbitrator. But the ranks of arbitrators include, often by choice of the parties, many retired judges and senior status lawyers arriving from careers focused in other disciplines without particular exposure to the arbitral process. Arbitral institutions that welcome, as they should, such newly-minted arbitrators to their rosters, bear a large responsibility for their exposure to shared norms and best practices in the arbitrator community.

Last week the US Second Circuit Court of Appeals wrote what should be the final chapter in one of the largest international arbitrations to emerge from the US financial crisis of 2007-2008.

The case was an “international” arbitration only in the sense that Claimant was a Swiss company while Respondent was a US affiliate of a Swiss investment bank. The arbitration took place in New York under the arbitration rules of the Financial Institutions Regulatory Authority (FINRA), and resulted in an award issued by a three-member tribunal, without reasons as is customary in FINRA arbitrations, in favor of Claimant for more than $400 million of investment losses. After two years of post-award litigation focused on the bank’s claims of inadequate disclosure by the non-lawyer arbitrator, the judgment confirming the award has been upheld and the matter presumably has reached a conclusion. (STMicroelectronics, N.V. v. Credit Suisse Securities (USA) LLC, 2011 U.S. App. LEXIS 11116 (2d Cir. June 2, 2011)).

My topic for discussion is the Second Circuit’s rejection of the bank’s claim of misconduct by the arbitrator in regard to his disclosures during the appointment process. But for financial crisis aficionados I offer this brief recap of the essential facts: The bank, having assured the customer that it would abide the latter’s instructions to invest its funds only in federally-guaranteed securities, mainly pools of student loans, instead put the client’s funds in risky derivatives bearing no federal guarantees, and the bank consistently mis-identified the investments in the e-mail confirmations so that the client would think its instructions were being heeded. When the scheme was uncovered, the customer was left with worthless, unmarketable securities. Two of the bank’s investment managers were eventually prosecuted for securities fraud, convicted, and sentenced to jail terms.

The bank’s challenge to the non-lawyer arbitrator — a retired finance professional acting frequently as an expert witness and consultant in customer-bank disputes —  claimed the arbitrator made misleading and inadequate disclosures that concealed the fact that he was engaged mainly by customers and therefore was pre-disposed toward the customer position on issues germane to the dispute.  The Second Circuit, in agreement with the District Court, held that the bank had failed even to establish its factual premise that the arbitrator worked predominantly for customers — and so the Court had no occasion to reach what it said were difficult and unresolved issues of whether the type of predisposition alleged, when concealed by non-disclosure, could support the vacatur of an award on grounds of “misconduct” by the arbitrator resulting in “prejudice” to the party. (FAA Section 10(a)(3)).

What I find to be an especially curious element of the Court’s analysis of the bank’s position is its observation (with a decided note of criticism) that the bank had failed even to request use of the discovery methods under the Federal Rules of Civil Procedure to secure details about the challenged arbitrator’s expert witness engagements. The Court stated that while it had shown reluctance in several cases to allow discovery to be used to mount challenges to awards based on arbitrator bias, it had not entirely foreclosed this type of use of court-annexed evidence-gathering.

But if, as the Court seems to suggest, this would have been a suitable occasion for discovery in aid of a post-award inquisition on an arbitrator, what are the criteria, inferable from this case, against which such requests will be measured in the future?

The first element would appear to be that the applicant for discovery has a substantial non-speculative concern about the arbitrator based upon concrete (but insufficient) evidence already in its possession. Here the bank had the formal written disclosure that said the arbitrator had had prior expert engagements on “both sides” of bank customer disputes. But the bank’s investigation, triggered by an offhand remark by the arbitrator during the hearings, revealed that he had apparently testified mainly for customers. 

Second, it would appear that the discovery applicant should have at least a credible legal position that, if the facts turn out to be as the applicant alleges, vacatur of the award under the FAA would be appropriate. In STMicro v. Credit Suisse, the bank’s position was that if indeed the arbitrator had been an expert/consultant mainly for customers against banks, but had not so indicated in the written disclosures, this would constitute misconduct prejudicial to the bank warranting vacatur. Neither the parties’ briefs nor the Court’s own research uncovered case law pointing in either direction on this issue.  

Third, the applicant should have a plausible position that the alleged arbitrator misconduct had a material impact on the outcome. This is technically part of the second element above (i.e. “prejudice”). But it deserves to be stated and considered separately from whether the alleged conduct was in fact misconduct at all.  In ST Micro v. Credit Suisse, the arbitral tribunal had issued an unreasoned award giving money damages in excess of $400 million; the award did not even indicate which of several causes of action (fraud, contract, negligence, fiduciary duty, etc.) had been sustained. In such circumstances, it cannot be excluded that the challenged arbitrator was a particularly effective advocate of the customer’s position in the deliberations of the Tribunal. Applicants for post-award discovery on arbitrator conduct may face a higher burden where there is a reasoned, unanimous award – suggesting that any misconduct or bias of one arbitrator, who was not the presiding arbitrator, was inconsequential.  A similarly higher burden may be imposed even when the unanimous award is unreasoned but is made by a tribunal that includes party-appointed arbitrators.  If the challenged arbitrator is the other party’s appointee, but the challenging party’s appointee subscribes to the award, the court may doubt that the applicant has a prima facie case of prejudice. In the Second Circuit case under discussion, all three arbitrators had been appointed by FINRA based on the parties’ rankings of a list of candidates.

Fourth, the court must consider that the discovery request is made in a proceeding that likely started with a petition to confirm the award – and was met, in response, with a cross-motion to vacate. Confirmation proceedings are supposed to be streamlined and expedited, according to the Supreme Court. And the grounds for vacatur are to be narrowly construed. Such discovery requests must be viewed in the context of federal arbitration policy as reflected in FAA jurisprudence.

To answer tentatively the question posed in the title of this post, the Second Circuit in STMicro v. Credit Suisse can scarcely be said to have opened the floodgates for discovery designed to mount an award challenge based on arbitrator bias or misconduct. But parties seeking such discovery will surely find STMicro v Credit Suisse in their research. And so it is useful to consider, as this post has attempted, precisely what are or should be the factors guiding judicial discretion in addressing such discovery requests.

(This is a condensed Arbitration Commentaries version, without citations, of what may become a longer and more formal article on joinder of new parties in an ongoing international arbitration. The source material for this Commentary consists of the various reports of the UNCITRAL Working Group and the notes of the UNCITRAL Secretariat, which may be found on the UNCITRAL website.)

The joinder of persons as new parties to an ongoing international arbitration, before an arbitral tribunal already appointed by the original parties or confirmed by an appointing authority, has presented difficulties that have not yet yielded to a unifying transnational practice principle.

The problems unique to joinder are not exclusively those associated with arbitration involving non-signatories to the arbitration agreement. Those challenges are present even if the non-signatory is named in the original Request for Arbitration.  The unique problems arise from the conceptual difficulty of requiring New Party C to arbitrate before a tribunal that has been selected only by Original Parties A and B.

UNCITRAL Rule 17(5), an entirely new provision in the 2010 UNCITRAL Rules,  is the product of several years of debate in the UNCITRAL Working Group that was charged with giving birth to the 2010 Rules. The 1976 version of the Rules had no provision on joinder, so the adoption of any measure was an important new development.  As first introduced as a proposal in 2006, the text would have read:

The arbitral tribunal may, on the application of any party, allow one or more third persons to be joined in the arbitration as a party and, provided such a third person and the applicant party have consented, make an award in respect of all parties involved in the arbitration.

But some questioned whether this was a new development at all. After all, the 1976 Rules did not expressly forbid joinder. If the new party and the existing parties consented, joinder could and presumably did occur.

The challenge was to have a rule that would permit a third person to be joined, and for the joined person to be regarded as having consented that its rights and duties should determined by the award of an arbitral tribunal that it had no role in selecting.

To meet this goal, new text of the proposed Rule emerged:

The arbitral tribunal may, at the request of any party, allow one or more third persons to be joined in the arbitration as a party provided such person is a party to the arbitration agreement, unless the arbitral tribunal finds, after giving all parties, including the person or persons to be joined, the opportunity to be heard, that joinder should not be permitted because of prejudice to any of those parties.  The arbitral tribunal may make a single award or several awards in respect of all parties so involved in the arbitration

Proponents of this text urged the view that if the newly-joined person were already a party to the arbitration agreement, that person could be said to have consented in advance to the prospect of being joined to an arbitration already pending between other parties to the same agreement, before a tribunal chosen by or for those original parties, provided of course that the arbitration to which the third person was being joined was pursuant to that agreement.

Opponents of this text argued that an award against the joined person would be vulnerable to challenge under Article V(1)(d) of the New York Convention as having been made according to procedures that varied from the agreement of the parties. They sought to re-insert the requirement that all parties should consent to the joinder at the time of joinder.

But the proposed reinsertion of such specific consent language was rejected. Consent, said the proponents of 17(5) as adopted, was adequately addressed by the requirement that the joined person be a party to the arbitration agreement.

After nearly four years of gestation, including comments elicited from major arbitral institutions, the proponents prevailed and this major new development in international arbitration procedure came into force in the 2010 UNCITRAL Rules.

To my knowledge no court has yet been required to decide whether an award against a party joined under Rule 17(5) violates the New York Convention. But the case for such a violation would seem to be unpersuasive. The Convention does not elevate equality of the parties in selection of the Tribunal to the status of an independent principle. That principle is enshrined to an extent in virtually all modern arbitration rules and laws, but is sometimes qualified due to the need to serve other important interests.

For example, the ICC Rules do not provide for strict equality of all original parties in selecting a tribunal in all cases. If a Claimant names four Respondents in the Request for Arbitration, and there will be three arbitrators, the Claimant is expected to nominate one arbitrator and the Respondents are to nominate one arbitrator jointly. The four Respondents are only treated equally with Claimant if they fail to make a joint nomination, whereupon the ICC Court will appoint the entire tribunal. (ICC Rule 10).

Of course, joinder under UNCITRAL Rule 17(5) might not save the award against the joined party from vacatur in the courts at a seat of arbitration whose arbitration law does enshrine a principle of strict party equality in choosing the tribunal. But that is not the case in UNCITRAL Model Law countries, where vacatur standards are co-extensive with grounds for refusal of enforcement under the Convention. Nor is it the case under the US Federal Arbitration Act.

Where issues of consent are most likely to arise in joinder practice under Rule 17(5) is in the realm of non-signatories to the arbitration agreement. The drafters of Rule 17(5) obviously selected the phrase “party to the arbitration agreement” with considerable deliberation. Perhaps a person that is required to arbitrate because a signatory signed the contract as her agent is a “party.” But what about a person against whom the arbitration clause may be enforced under the doctrine of arbitral equitable estoppel? Whether that person is a 17(5) “party” to the arbitration agreement probably depends on the view under law applicable to determine who is required to arbitrate under the agreement. No guide to a drafters’ interpretation of “party to the agreement” is to be found in the UNCITRAL drafting history (travaux préparatoires). So joinder practice will still be marked by some variation depending upon the applicable arbitrability law.

That said, Rule 17(5) is a major accomplishment. It may become the transnational standard to which major arbitral institutions gravitate as they review and update their rules on joinder.  Such a rules review at the ICC is in progress (although this writer has not involvement or knowledge of the position the revised ICC Rules will take). An important advance in the efficiency of arbitral proceedings should be the result if Rule 17(5) is widely emulated, and this should be seen as wan improvement that does not give serious offense to the consensual basis for arbitration.