Non-Party Discovery in Arbitration: The Second Circuit Weighs In
By Marc J. Goldstein

It is an often-overlooked fact that the Federal Arbitration Act (“FAA”) is now a very “old” statute. Enacted in 1925, and not notably amended since then (except to add Chapters 2 and 3, governing matters under the New York and Panama Conventions, respectively), the FAA is 13 years older than the Federal Rules of Civil Procedure.

The 83-year-old first chapter of the FAA says either nothing or very little – depending on one’s reading of Section 7 — about pre-hearing discovery in arbitrations subject to the Act. That is understandable, because the FAA was motivated mainly by a concern about the enforceability of arbitration agreements and the enforcement of arbitral awards, and not by a desire to establish procedures to be followed by arbitrators.

Section 7 is the singular exception to this. It is concerned with the subpoena power of the arbitrator. It was and is a special provision vesting in arbitrators some of the powers to secure evidence by compulsory process that were vested in federal district judges under rules governing subpoenas that predated Rule 45 of the Federal Rules of Civil Procedure.

Section 7 provides in relevant part that the arbitrators may subpoena persons to “appear before them to give testimony” and “to bring with them” relevant documents. Over the years, arbitrators have acted with varying degrees of adherence to these seemingly unambiguous words, and courts have drawn conflicting conclusions on whether Section 7 permits subpoenas for pre-hearing discovery. In November 2008, the U. S. Court of Appeals for the Second Circuit, embracing a Third Circuit decision in 2004 (written by a notable strict statutory constructionist, then-Circuit Judge now-Justice Samuel Alito) , held that Section 7 does not permit issuance of arbitral subpoenas for pre-hearing discovery from non-parties. Life Receivables Trust v. Lloyd’s of London Syndicate 102, 549 F.3d 210 (2d Cir. Nov. 25, 2008).

The Life Receivables Trust Decision

The Life Receivables decision arose from arbitration over an obscure insurance product. Claimant (“Trust”) and an affiliate (“Peartree”) had a business buying life insurance policies from the elderly, for cash payments at a discount to policy values. Peartree would buy the policies and transferred them to Trust. Trust then paid the premiums to keep the policies in force until the insureds died. Trust elected to insure with Respondent, a Lloyd’s of London Syndicate (“Syndicate”), the risk that the named insured on the purchased life policy would outlive his or her calculated life expectancy. Trust thus paid Syndicate premiums, in exchange for Syndicate’s commitment to pay Trust the net death benefit of the underlying life policy if the insured outlived his or her life expectancy by two years or more.

Arbitration ensued when Syndicate denied payment on a Trust claim. Syndicate sought documents from Trust, including Peartree documents. Trust said it had no “control” over Peartree, and could not produce Peartree documents. Syndicate responded by having the arbitral tribunal issue a subpoena to Peartree for pre-hearing documents discovery, which Trust did not oppose. Peartree moved in District Court to quash the subpoena; Syndicate cross-moved to compel compliance; and the District Court granted the motion to compel. Peartree complied with the subpoena, but nevertheless appealed from the District Court order. The Second Circuit concluded that the dispute was not mooted by the compliance, and proceeded to decision.

The holding of the Second Circuit is in fact quite narrow, although some effort is required to discern the holding and its limitations. And, as if often the case, even what the Court states to be the “holding” cannot be understood without reference to other amplifying remarks. At one point, the Court states that, under the clear language of Section 7 , “[d]ocuments are only discoverable in an arbitration when brought before arbitrators by a testifying witness.” 549 F.3d at __. This is in fact quite close to being what the case holds, even though at another point a few sentences later the Court purports to “join the Third Circuit in holding . . . that section 7 of the FAA does not authorize arbitrators to compel pre-hearing document discovery from entities not party to the arbitration proceedings. “ Id. at __.

But the Court was careful to avoid having its decision interpreted as prohibiting all document disclosure from non-parties that occurs at a stage of the proceedings earlier than the “hearing on the merits,” i.e. the hearings scheduled, normally after submission of documentary exhibits and legal briefs, for the presentation of oral testimony, upon direct and cross-examination and under questioning by the Tribunal. Noting that Section 7 “’does not leave arbitrators powerless’” to order document production from non-parties, the Court stated that “arbitrators may, consistent with Section 7, order ‘any person’ to produce documents so long as that person is called as a witness at a hearing.” (Id. at __, emphasis supplied). And, referring to a 2005 Second Circuit decision — in which this Section 7 issue was raised, but not decided — the Court noted that “section 7 authority is not limited to witnesses at merits hearings, but extends to hearings covering a variety of preliminary matters.” Id. Essentially, according to the Second Circuit, section 7 imposes “a presence requirement. . .forc[ing] the party seeking the non-party discovery – and the arbitrators authorizing it – to consider whether production is truly necessary.” Id.

Implications for Practice

What are the practical consequences for practitioners and arbitrators conducting arbitrations in New York after the Second Circuit’s decision?

Certainly in many cases a party to the arbitration will be mainly interested in obtaining documents from a non-party, reviewing those documents in advance of merits hearings, and deciding whether any oral testimony from the producing party or person is needed. The Second and Third Circuit decisions do not interpret the word “witness” in Section 7 in a limiting fashion. Further neither Section 7 nor these decisions imposes a requirement that the non-party appear as a witness at a particular
stage of the case. Therefore the arbitrator seeking accommodate a party’s desire to obtain a third party’s documents before the hearings on the merits could lawfully issue a subpoena to a records custodian directing the custodian to appear to give testimony on a particular date and to bring the documents on that date. Nothing in the Act prevents the testimony from being confined to an examination of the custodian concerning, for example, the efforts made to search for documents, and document retention policies. This is not a circumvention of Section 7, which creates only a modest requirement for the presence of a witness and an arbitrator at the time of document production, and arguably requires that the testimony and the documents be made part of the arbitral record.

But perhaps this does not mean that Section 7 always satisfied when at least one of the arbitrators is present, and perfunctory testimony is presented, at the time of a non-party document production. Section 7 states that the documents obtainable are those that “may be deemed material as evidence in the case.” These words appear to define a scope more narrow than the Federal Rule of Civil Procedure standard of “might lead to discovery of admissible evidence.” It would appear that the requesting party must show with specificity that the documents requested could be material evidence. Thus, while arbitrators sitting in the United States would have discretion to apply the Federal Rules standard to document production by the parties — unless the parties have otherwise agreed — in considering a non-party subpoena the arbitrator are required to consider whether the requested information “may be deemed material as evidence.” There is little if any guidance from the courts on the meaning of this standard, but there does appear to be more room than in federal litigation for a party opposing the subpoena, or the non-party, to protest a “fishing expedition.”

In some cases, arbitrators may wish to streamline proceedings and limit the involvement of non-parties by issuing a non-party subpoena for the appearance of the witness during the same hearings on the merits at which other testimony will be heard. This substantially eliminates any “pre-hearing discovery” element of obtaining the non-party’s evidence. Such limitations are not required by Section 7, as the Second Circuit made clear in the Stolt-Nielsen case , but an arbitrator acts well within her discretion in imposing such limits (bearing in mind that Section 7 creates no right of a party to obtain the issuance of an arbitral subpoena).

Further, even if the Section 7 non-party “witness” might be a records custodian, the arbitrator might elect to permit a subpoena for records only in connection with material fact testimony of a non-party fact witness. Section 7 certainly supports, even if does not require, an approach that gives primacy to the testimony of the non-party witness and warrants issuance of a document subpoena only “in a proper case” – which the arbitrator might limit to a case where the party cannot effectively present its case without the documents known to be or reasonably believed to be in the non-party’s possession.

Unlike a subpoena issued in litigation by a party’s attorney, compliance with the arbitral subpoena cannot be waived or compromised by fewer than all the parties to the arbitration. If all parties agree with the subpoenaed non-party to waive appearance as a witness if the documents are produced in advance of hearings, Section 7 is not violated. Section 7 addresses only what the arbitrators may compel, not what the parties might agree with the witness under the influence of that compulsion. But what should happen if the witness contacts the arbitrator to ask that pre-hearing production without testimony be accepted, and all parties do not agree? Nothing in the FAA empowers the arbitrator to accept such a request; this in effect modifies the subpoena to require only pre-hearing disclosure outside the presence of any arbitrator, and thus violates Section 7 according to the construction given by the Second and Third Circuits.

Arbitrators should also apply Section 7 in a fashion the respects the parties’ ability to cross-examine the witness. Several measures may be taken. One step would be to require by procedural order that if one party receives documents from the subpoenaed non-party in advance of the return date of the subpoena, those documents must be disclosed immediately to all parties. Another feasible measure is to treat the non-party as a witness under the control of a party, if this relationship is considered to exist based upon facts presented to the arbitrators. In that case, any general requirement that witnesses provide witness statements in advance of hearings, and that parties disclose their exhibits prior to the hearings, may be extended to the testimony and documents of a non-party. This approach prevents a party from manipulating Section 7 to shield a cooperating non-party witness from the rules imposed by the arbitrator concerning voluntary witnesses. A related point is that, especially in the case of a non-party that might be more friendly to one side than the other, the arbitrator in setting limits on the scope of the subpoena for records should afford both sides reasonable scope for obtaining documents useful as direct evidence and as impeachment. Finally, where records are in fact brought to the hearing at which the non-party’s fact testimony is given, and the documents cannot be assimilated by the parties quickly enough for them to prepare questions immediately, reasonable regulation of the timing of the testimony (e.g. asking the witness to return on a later date in a multi-day hearing) should not offend Section 7 by making the subpoena for “pre-hearing” discovery.

In complex cases, where the challenge of digesting large volumes of non-party documents before testimony is foreseeable, the scheduling of appearances by non-party custodial witnesses as a separate, relatively early procedural phase of the case, should meet the needs of the parties and the requirement of Section 7.

Conclusion

Arbitral practice has varied widely on the involvement of non-parties through the compulsion of arbitral subpoenas. The Second Circuit’s decision holding that all disclosure by non-parties pursuant to subpoena must, to comply with the FAA, take place in the presence of one or more arbitrators in connection with testimony by the witness, is faithful to the text of the FAA and the context in which it was enacted. The decision is evidence of a growing consensus that under existing federal law there are distinctive differences between litigation and arbitration in the ability to obtain and present evidence from non-parties.

Hall Street Associates v. Mattel, Inc.
and the Uncertain Future of Manifest Disregard of the Law

By Marc J. Goldstein

The decision of the US Supreme Court in Hall Street Associates v Mattel, Inc. , in March 2008, was possibly the Court’s most important ruling on federal arbitration law in this nine-year-old milennium.

The Court granted certiorari to decide, and did decide, an unsettled important question: whether the scope of federal judicial review of arbitral awards may be expanded, by contractual agreement, beyond what the Federal Arbitration Act (“FAA”) provides.

That question was answered in the negative. But the provocative and potentially transformational aspect of Hall Street is not its narrow holding that contractual expansion of judicial review is unlawful — that was surely to be expected — but rather is the Court’s sweeping and emphatic disavowal of more than a half-century of judicial embellishment of the review powers conferred on courts by the FAA. That embellishment, of course, is the famous, some might say notorious, doctrine of “manifest disregard of the law.”

This doctrine had evolved since the 1950s as a federal common law principle permitting judges to set aside arbitration awards without apparent regard for the limited grounds for vacating awards that are provided in the federal arbitration statute, the FAA.

And while the incidence of awards actually being set aside on this ground was exceedingly small, well-financed arbitration losers have been unable to resist the roulette game. And so, for a half-century, the federal courts have been a bustling appellate casino in which the “house” (that is, the court as guardian of federal arbitration law) nearly always wins. But in that process, arbitration has suffered, as the prospect of long and costly judicial process has robbed arbitration of much of its promise as a quicker and less expensive option for getting closure on business disputes.

Now the Supreme Court has proclaimed that the statutory grounds for setting aside awards, in Section 10 of the FAA, are “exclusive.” But that proclamation was issued without a decisive rejection of the “manifest disregard of the law” doctrine. And so, at least temporarily, the federal courts have been left to navigate between pre-Hall Street “manifest disregard” case law, and the Court’s critical but not necessarily fatal assault on that case law.

*

The manifest disregard doctrine had its roots in a now-discredited 1953 Supreme Court case, Wilko v Swan . The Court’s decision in Wilko effectively prohibited arbitration of federal securities fraud claims, until it was overruled in 1985 . Wilko held that an agreement to arbitrate such claims was a “stipulation … waiving compliance” with the 1933 Securities Act, because the securities purchaser waived the right conferred by the Act to bring suit in a state or federal court.

The majority’s rationale in Wilko was unabashedly hostile to arbitration: that arbitrators could not be trusted to apply scrupulously a statute as important as the Securities Act, and their feared failures to apply the Act, while in principle subject to reversal under the FAA as exceeding their powers, would in practice go unchecked because the FAA affords no avenue for review of arbitrators’ interpretations of the law. Thus, in a sentence that followed immediately after the Court’s statement that arbitral failure to apply the Securities Act would be reversible, under section 10 of the Arbitration Act, as exceeding the powers of the arbitrator, the Wilko Court stated: “…. [T]he interpretations of the law by arbitrators in contrast to manifest disregard are not subject, in the federal courts, to judicial review for error in interpretation.”

In the Court’s dicta in Wilko, it is virtually impossible to discern any intention to energize a new principle of judicial review of awards by arbitrators. The solution imposed by the Court in Wilko, for the perceived practical inadequacy of judicial review under the FAA, was to make claims under the Securities Act non-arbitrable, and to deny enforcement of contracts to arbitrate such claims. The only evident purpose for the use of the term “manifest disregard” in Wilko was to contrast the available statutory review when arbitrators exceed their powers by refusing to apply governing law, with the unavailability of judicial review when arbitrators make mistakes when trying to apply the law. Nevertheless, from this tiny rhetorical seed was harvested a judicial doctrine permitting review of arbitral awards that was eventually accepted by every federal judicial circuit.

**
The majority opinion in Hall Street written by Justice Souter is remarkable for the Court’s unabashed desire to place its imprimatur on arbitration issues related to, but not directly involved in, the specific question presented for review.

The opening sentences of the opinion reveal this: “The question here is whether statutory grounds for prompt vacatur and modification may be supplemented by contract. We hold that the statutory grounds are exclusive.”

Why did the Court elect not to answer the narrow question in comparably narrow terms? The Court could easily have said “We hold that such expansion of review by contract is contrary to the FAA.” After all this formulation is, in conventional jurisprudential terms, the “holding” of Hall Street.

The explanation for this, I suggest, is that the main issue on the minds of the Justices was not the question of expansion by contract — a rare occurrence. The Justices evidently were concerned, as are arbitration practitioners around the world, over what many regard as a pernicious and uniquely American judicial doctrine: manifest disregard of the law. That was the “hot-button” issue because the lower federal courts in important arbitration venues have been over-burdened with “manifest disregard” litigation. And the unfortunate reality is that “manifest disregard” is the garb in which arbitration losers cloak requests for judicial review of alleged mistakes of law (or fact) by arbitrators.

By declaring, as the “holding” of the case, that the statutory criteria for review of awards are exclusive, the Court appears by necessary implication to have declared the demise of manifest disregard as a judicially-created non-statutory ground for review – although there is language in the opinion that some readers might consider to have left even this issue still open. But it seems nearly certain that manifest disregard as a judicial creation has been interred, and the urgent question today is whether the Court intended to go further, to signal the eventual complete demise of the manifest disregard doctrine, even as a judicial gloss upon or interpretation of Section 10 of the FAA.

So, the heart of the majority opinion in Hall Street — the language arbitration lawyers read again and again — is the Court’s ponderous revisiting of Wilko. And the main vein of that discussion, of course, was the Wilko Court’s doctrine-provoking remark that “interpretations of the law by arbitrators in contrast to manifest disregard are not subject, in the federal courts, to judicial review for error in interpretation.”

Here it is useful to quote the Hall Street majority opinion at some length:

Hall Street reads this statement as recognizing “manifest disregard of the law” as a further ground for vacatur on top of those listed in § 10, and some Circuits have read it the same way. . . . Hall Street sees this supposed addition to § 10 as the camel’s nose: if judges can add grounds to vacate (or modify), so can contracting parties.
But this is too much for Wilko to bear. Quite apart from its leap from a supposed judicial expansion by interpretation to a private expansion by contract, Hall Street overlooks the fact that the statement it relies on expressly rejects just what Hall Street asks for here, general review for an arbitrator’s legal errors.
Then there is the vagueness of Wilko’s phrasing. Maybe the term “manifest disregard” was meant to name a new ground for review, but maybe it merely referred to the § 10 grounds collectively, rather than adding to them. . . . Or, as some courts have thought, “manifest disregard” may have been shorthand for § 10 (a)(3) or § 10(a)(4), the subsections authorizing vacatur when the arbitrators were “guilty of misconduct” or “exceeded their powers.” . . . We, when speaking as a Court, have merely taken the Wilko language as we found it, without embellishment, . . . and now that its meaning is implicated, we see no reason to accord it the significance that Hall Street urges.

The decision thus appears to resolve no issues with certainty, other than the one immediately at hand concerning expansion by contract. And it thus leaves, at least temporarily, considerable room for federal courts to continue to apply their pre-Hall Street manifest disregard case law, particularly by accepting the Court’s invitation to treat “manifest disregard” as a term equivalent to the statutory bases to set aside awards.

The Second Circuit for example held Stolt-Nielsen SA v. Animalfeeds International Corp. , that when an arbitrator fails to interpret the contract at all, or wilfully refuses to apply controlling law presented by the parties, she commits manifest disregard of the law as understood in the Second Circuit’s pre-Hall Street case law, and exceeds her powers under Section 10(a)(4) of the FAA. Thus, in the view of the Second Circuit, manifest disregard and Section 10(a)(4) have essentially the same scope, and manifest disregard may be applied as a judicial gloss or interpretation of Section 10(a)(4).

This approach has a veneer of logic to it, but on closer inspection it is troublesome. If the conduct of arbitrators that constitutes manifest disregard falls squarely within the statutory vacatur category of “exceed(ing)…powers,” then why was there ever a need perceived by the courts for a judicially-crafted additional non-statutory ground for vacatur? And why have the courts stated repeatedly over the last several decades that manifest disregard is a non-statutory ground for vacatur, nearly always without suggesting that arbitral conduct in manifest disregard of the law qualifies for vacatur under Section 10 of the FAA?

The problem with the reasoning of Stolt-Nielsen is that the Court effectively performs a reverse-engineering of 50 years of federal common law adjudication, reshaping the non-statutory doctrine of manifest disregard into a construction of the FAA itself, when the FAA’s stringent limitations on vacating awards were the original motivation for creating a supplemental non-statutory doctrine. This is an extraordinary approach to statutory construction, one that entails not the least effort to discern what was the actual intent of Congress when the FAA was enacted.

Only three other federal circuits have remarked upon the issue since Hall Street; none held conclusively that Hall Street mandates the demise of manifest disregard. Only one published decision by a federal district court in Colorado addressed the issue; the court declined to decide upon the impact of Hall Street on the manifest disregard doctrine, stating that “there remains some question as to whether Hall Street did indeed eliminate all judicially-created grounds for vacatur.”
What does seem appropriate, after Hall Street, is for courts to apply a careful analytic approach to determine what powers were conferred on the arbitrator by the parties’ agreement, and by any rules of arbitration or rules of law that the parties agreed would govern the arbitrators’ conduct. Thus if the parties had agreed that there would be no pre-hearing depositions, and the arbitrator ordered such depositions, there would be an exceeding of powers. If the parties had agreed that Colorado law would govern, and the arbitrator elected to apply a New York rule that was the opposite of a clearly applicable controlling rule of Colorado law, she would exceed her powers, as she had no power derived from the arbitration agreement to apply substantive rules of law other than those of Colorado.
***

Whether or not manifest disregard survives in any form, the systemic problem of federal appeals from arbitration awards will remain. Whether the governing legal standard under Section 10 is the pre-Hall Street “manifest disregard” doctrine, or some other approach to deciding what arbitral conduct exceeds arbitral powers, arbitration losers will still present their arguments to the District Courts, hoping to find an arbitration-hostile audience that will bend the law to overrule a questionable award. Only the remote possibility of sanctions for bad faith litigation is available as a deterrent, save in those cases where an arbitration clause provides for the prevailing party to recover post-award litigation fees.

Both statistics and observation suggest that the vast majority of manifest disregard appeals are meritless. In a footnote of the Stolt-Nielsen decision, the Second Circuit cited 18 manifest disregard decisions in the Second Circuit since 2003. In 15 of them, all relief was denied, and in ten of those cases the Court ruled by summary order. In the other three cases, one award was vacated and two were remanded for clarification. The inescapable conclusion is that nearly all manifest disregard litigation is a costly exercise in futility, which serves to deprive arbitration victors of the full value of their success – especially in lesser-value cases where the ratio between legal fees and money damages is quite high.

What can be done about this? The most appealing solution, I believe, is to amend the FAA to provide for recovery of attorneys’ fees by the prevailing party in judicial proceedings in which vacatur of an award is sought, unless the parties have expressly agreed that each party shall bear its own costs in any such proceedings. The victor in such litigation vindicates an important element of federal arbitration policy, and as such should be entitled to recover fees as a reward for vindicating federal law and policy. Moreover, the prospect of a fee award should serve as a deterrent to many, if not most, arbitral appeals which are essentially challenges for legal or factual error by the arbitrator that are destined to fail. Such an amendment to the Act would go far toward fulfillment of the fundamental objectives of the FAA, which, per the Supreme Court in Hall Street, “substantiat(es) a national policy favoring arbitration with just the limited review needed to maintain arbitration’s essential virtue of resolving disputes straightaway.” Such an amendment would address the essential flaw in our current system — that risk-free merits appeals from arbitral awards (meaning appeals that entail no adverse consequence to the petitioner/appellant, other than losing and bearing its own costs), clothed in “manifest disregard” or “exceeding powers” attire, serve, in the words of Hall Street, to “bring arbitration theory to grief in post-arbitration process.”