Mayhew-Hite Report        Managing the Risk of Legal Error in Arbitration

Stephen L. Sepinuck


Arbitrators sometimes make mistakes. Sometimes spectacular mistakes. Consider the case of Martin Evans. He and Craig Nielsen purchased several H&R Block franchises, with each franchise owned by a separate limited liability company. Nielsen provided financing for Evans, who signed a promissory note for the amount due. The note provided that, upon default, Nielsen was authorized “to charge or setoff all sums owing on the debt” against Evans’ interests in the LLCs. Evans did default and Nielsen proposed to keep the LLC interests in full satisfaction of the debt. Evans objected and brought an action seeking a declaration that Nielsen’s seizure of the LLC interests was ineffective and that Evans remained a member of the LLCs.

The matter was referred to arbitration pursuant to the parties’ agreement. The arbitrator ruled that Article 9 of the UCC did not apply because § 9-109(d)(10) generally excludes recoupment and setoff from the scope of the Article.1 The arbitrator then added that, even if Article 9 did apply and even if Nielsen had failed to comply with § 9-620 by not obtaining Evans’ consent to Nielsen’s acceptance of the collateral in satisfaction of the debt, the acceptance was effective and Evans’ only right was to recover damages for the loss of a surplus. 2

Both rulings are patently wrong. Setoff is a mechanism for netting mutual debts. A security interest, on the other hand, is an interest in personal property that secures a debt.3 Evans’ interests in the LLCs were his personal property, not debts. Thus, the note provided for a security interest, not setoff. The fact that the note described Nielsen’s right as a “setoff” is immaterial: Article 9 applies to any transaction, “regardless of its form,” that creates a security interest.4 The arbitrator’s ruling, if applied generally, would allow people to avoid application of Article 9 simply by labeling the creditor’s rights as a “setoff.” It is therefore bad policy in addition to being clearly erroneous.

Article 9 is equally clear that acceptance of collateral in satisfaction of the secured obligation is  ineffective, not wrongful, unless the debtor consents after default.5 Thus, if Evans timely objected to Nielsen’s proposal, as he apparently did, Evans remained the owner of his LLC interests.

Despite these errors, a Utah trial court confirmed the arbitration decision. In March, that ruling was affirmed on appeal.6 The appellate court noted that the judiciary’s role is not to review an arbitrator’s award for legal error, but merely to determine whether the arbitrator exceeded his authority. The court then concluded that the decision was not so without foundation as to justify refusing to enforce it based on irrationality or manifest disregard for the law.7

For litigators and transactional attorneys alike, this case should be troubling. Arbitration is frequently touted not only as speedier, less expensive, and more confidential than litigation, but also as less prone to error because of the expertise and experience of the arbitrators. But if even flagrant errors of law cannot be corrected, two questions naturally follow: (i) how does the risk of legal error in arbitration differ from the risk in litigation; and (ii) how should transactional attorneys manage that risk to protect their client’s interests?


Quantitative Risk of Error

There are some reasons to believe that the risk of error is lower in arbitration than in litigation. First, arbitrators can be screened and chosen for their expertise in the legal issues in dispute or for their familiarity with the parties’ industry. Such was apparently not done in Evans v. Nielsen, however. The arbitrator in that case was a professor of clinical law experienced in dispute resolution, but with no apparent expertise in commercial law.

Second, arbitrators have a bit more freedom to confer and consult with third parties before rendering a decision. For example, under the rules of the American Arbitration Association, an arbitrator may obtain help from an associate, a research assistant or other person if the arbitrator informs the parties and the person providing help agrees to be bound by the confidentiality rule that binds the arbitrator.8 In contrast, a judge may obtain written advice from a disinterested expert on the law only after giving advance notice to the parties and affording them the opportunity both to object and to respond to the advice received.9 Despite this greater freedom afforded arbitrators, it is not clear that either arbitrators or judges avail themselves of this authority in anything other than an exceptional case.

On the other hand, there is one reason to think that the frequency of error in litigation might be less than in arbitration. Judicial decisions are a matter of public record. No judge likes to be wrong and most take the time to inform themselves about the law that applies to the dispute before them. Presumably, arbitrators too want to get the law correct – indeed their selection as arbitrator in future cases probably depends on their reputation – but the confidentiality of their decisions makes it difficult to assess their competence and correctness. Secrecy might beget sloth.

These countervailing considerations leave us with little guidance; each is merely an untested hypothesis about which process is more prone to an erroneous decision. Moreover, for several reasons, it is unlikely there will ever be a reliable, empirical study of the comparative frequency of error in arbitration and litigation. First, there is the normative or epistemological problem of determining which decisions are wrong. While the arbitrator’s decision in Evans v. Nielsen was unquestionably wrong, such blatant errors are, one can hope, relatively rare. Instead, many errors will concern matters about which reasonable minds could disagree. It seems likely, therefore, that we could never have widespread consensus on which decisions were in fact erroneous. Second, not all errors are equivalent. Thus, even if we could quantify the rate of error, it is doubtful we could objectively determine the significance of those errors. Finally, too much of each data set is unavailable. Many judicial rulings at the trial court level never lead to a reported decision or even to an unreported decision available on Lexis or Westlaw. Arbitrators are often not required to explain their reasons and the great bulk of arbitrations rulings are confidential. Moreover, there is no way to ensure that any sampling of either data set would be representative.

Qualitative Risk

While the relative quantum of error might never be known, in at least one respect the risk of error in arbitration seems qualitatively different – and greater – than the risk of error in litigation. That is because an arbitrator’s decision may be based on notions of justice and equity; it need not be consistent with the law.10 Thus, for example, an action barred by the applicable statute of limitations might nevertheless lead to an arbitration award.11

Indeed, one distinguished commercial lawyer recently reported that he received arbitrator training several years ago from a national arbitration service. During that training, the group of prospective arbitrators was given a hypothetical involving an effort to collect a usurious loan and told that the penalty under applicable law for charging usurious interest was a forfeiture of the right to all interest. When asked how they would rule in the case, approximately one-third stated they would apply the law and prohibit the lender from recovering any interest. Approximately one-third said they would reduce the interest rate to the highest non-usurious amount. The remaining one-third stated that they would enforce the agreement as written despite the prohibition on usury. The trainers did not indicate that any of these ruling would be improper.

For these reasons, some arbitration decisions will be contrary to what the law requires. This is, of course, also true with respect to litigation,12 but appellate review of judicial decisions provides an opportunity to correct legal errors by the trial judge. In contrast, judicial review of an arbitration decision – at least in federal court – is restricted to evidence that the award was procured by fraud or corruption, the arbitrator was patently partial to one side, the arbitrator’s misconduct prejudiced one party’s rights, or the arbitrator exceeded his or her authority.13 Manifest disregard of the law might be an additional basis for a federal court to refuse to enforce an arbitration award, but that point remains in doubt.14 What is clear is that parties cannot by agreement expand the bases for federal judicial review, such as by authorizing courts to review for any legal error.15 The risk of legal error is, simply put, “the price of agreeing to arbitration.”16

It is important to understand, however, that these rules apply only to review in federal court. A few states, such as New Hampshire, provide for a more expansive judicial review of arbitration decisions, including review for  plain  error.17  In addition, at least one state S California S authorizes parties to provide in their arbitration agreement for judicial review of arbitration decisions based on legal error.18 Nevertheless, most states, particularly those that have enacted the Revised Uniform Arbitration Act,19 do not permit judicial review for legal error, even if the parties provide for it.20


If we accept the proposition that the risk of legal error in arbitration is greater than the risk of legal error in litigation – or if we simply want to reduce or manage that risk for whatever reason – there are four different strategies the transactional lawyer could employ.

Require the Arbitrator to Follow the Substantive Law. Contracting parties that expect their counterparts to strictly comply with their contractual and legal duties might choose to circumscribe an arbitrator’s otherwise wide discretion to render a decision based on notions of justice and equity. They can do this by including in the agreement a requirement that the arbitrator’s decision be based upon and consistent with the parties’ legal rights. This should prevent an arbitrator from willfully disregarding the law in favor of some sense of justice or equity. However, it is doubtful that this approach would have any effect on unintentional error, such as apparently occurred in Evans v. Nielsen.

Provide for Appeal to a Panel of Arbitrators. The parties can provide for non-judicial review of an arbitration decision. For example, they can permit an appeal of the arbitrator’s findings of fact and conclusions of law to an appellate arbitrator or to a panel of appellate arbitrators. Indeed, the rules of some arbitration organizations expressly envision an appealsprocess while those of others, such as the AAA, implicitly permit it.21 If such review is desired, the arbitration clause in the parties’ agreement should: (i) require the initial arbitrator to apply the law; (ii) require the initial arbitrator to state in writing the basis for the arbitrator’s decision;22  and specify the grounds for reversal on appeal, the standard of review, and the procedures to be

Provide for Judicial Review. If applicable state law provides for judicial review based on legal error, the parties could require the arbitrator to follow the law and choose the state courts in that state as the exclusive forum for  enforcing  or  challenging  an  arbitration  award.  Alternatively, if applicable state law permits parties to expand the scope of judicial review to include legal error, the parties could require the arbitrator to follow the law, provide for judicial review based on legal error, and choose the state courts in that state as the exclusive forum for enforcing or challenging an arbitration award.

Unfortunately, providing for appeal to either a panel of arbitrators or a court undermines two of the principal benefits that arbitration purports to have: speed and lower cost. The process might still be a bit faster and less expensive than litigation due to less discovery and motion practice, but the parties have to pay the arbitrator whereas they do not pay a judge. Moreover, if the parties arrange for judicial review for legal error, a third principal benefit of arbitration – confidentiality – is also compromised. For these reasons, parties concerned about error might wish to rethink the decision to arbitrate at all.

Reconsider Whether and What to Arbitrate. There are some types of contracts and disputes for which arbitration might be particularly desirable. For example, a business that provides goods or services to consumers might want to require arbitration to avoid  class proceedings or juries. Arbitration might also be appropriate with respect to a transaction involving trade secrets or confidential information which, if disclosed or made available publicly, might prompt or affect other litigation.

Finally, some businesses might want to arbitrate to help insulate themselves from – that is, to evade – the law. For example, a client who tends to charge usurious interest in a state with a significant penalty for doing so might wish to provide for mandatory arbitration in the hope that the arbitrator will not be inclined to enforce those penalties. Similarly, a buyer of structured settlement payments might attempt to use arbitration to bypass the statutory procedures designed to protect individuals from this somewhat predatory practice – although one somewhat notorious lender found that this did not work.23 Of course, evading the law might not be the most legitimate reason to provide for arbitration and an attorney drafting an agreement that skirts the law should carefully  consider whether doing so violates applicable ethical rules.24

On the other hand, there are several other situations for which parties might wish to avoid arbitration. One pair of arbitration advocates identified three:

  1. high stakes (“bet-the-company”) disputes, in which the parties may fear an aberrational arbitration award subject only to limited judicial review
  2. disputes in which the parties anticipate needing emergency relief, which arbitration is ill-suited to provide; and
  3. disputes in areas with clear and well developed law and contract terms, because the industry expertise of arbitrators is of less value and the limited judicial review in arbitration more [25]

Transactional lawyers should carefully consider this list. The first item in particular seems predicated on the risk of legal error in arbitration. More generally though, transactional lawyers should consider that parties enter into written agreements in large measure to detail their legal rights. An agreement to arbitrate is, to some degree, an agreement to surrender those rights and allow an arbitrator to render a decision contrary to the law. Query if that is what the client really wants.

1 Interim Award at 4 (Oct. 31, 2011).
2 Id.
3 See § 1-201(b)(35).
4 § 9-109(a)(1).
5 See § 9-602(a)(1), (b)(2), (c).
6 Evans v. Nielsen, 2015 WL 1325540 (Utah Ct. App. 2015).
7 Id. at *3-6.
8 AAA Code of Ethics for Arbitrators in Commercial Disputes, Canon VI(b) (2004).
9 Code of Judicial Conduct for United States Judges, Canon 3(A)(4)(c); ABA Model Code of Judicial Conduct,  Rule 9(a)(2).
10 See AAA, Commercial Arbitration Rules and Mediation Procedures, Rule 47(a) (2013) (“The arbitrator may grant any remedy or relief that the arbitrator deems just and equitable and within the scope of the agreement of the parties”).
11 Broom Morgan Stanley DW Inc., 236 P.3d 182 (Wash. 2010)
12 See Bayer Cropscience LP Texana Rice Mill Ltd, 2015 WL 1474393 (E.D. Mo. 2015) (citing to UCC 9-322(a)(1) as the applicable priority rule but improperly describing it as a first-to-perfect rule, rather than a first-to-file-or-perfect rule, and reaching the wrong result).
13 See 9 S.C. § 10(a) (applicable only to federal courts). See also Cal. Civ. Pro. § 1286.2 (specifying the exclusive bases for a California court to vacate an arbitration award); General Mills, Inc. v. BCTGM Local 316G, 2014 WL 5100650 (N.D. Ill. 2014) (arbitrator exceeded her authority in case concerning employer’s discharge of an employee by awarding damages to the union).
14 Compare Stolt-Nielsen A. v. AnimalFeeds Int’l Corp., 559 U.S. 662, 672 n.3 (2010) (suggesting that an arbitration decision may be reversed for a manifest disregard of the law, which “requir[es] a showing that the arbitrators knew of the relevant [legal] principle, appreciated that this principle controlled the outcome of the disputed issue, and nonetheless willfully flouted the governing law by refusing to apply it”); LaTour v. Citigroup Global Markets, Inc., 544 F. App’x 748 (9th Cir. 2013) (upholding  an arbitration award after reviewing it for manifest disregard of the law); Wachovia Sec. LLC v. Brand, 671 F.3d 472, 480-83 (4th Cir. 2012) (concluding after Stolt-Nielsen that manifest disregard of the law remains a basis for vacating an arbitration award); Schwartz v. Merrill Lynch & Co., Inc., 665 F.3d 444, 451-52 (2d Cir. 2011) (same); Broom v. Morgan Stanley DW Inc., 236 P.3d 182 (Wash. 2010) (a facial legal error is a basis for vacating an arbitral award because it indicates that the arbitrators exceeded their powers), with Med. Shoppe Int’l, Inc. v. Turner Invs., Inc., 614 F.3d  485, 489 (8th Cir. 2010) (manifest disregard of the law is no longer a basis for vacating a arbitration award); Frazier v. CitiFinancial Corp., 604 F.3d 1313, 1324 (11th Cir. 2010) (same); Citigroup Global Mkts., Inc. v. Bacon,  562 F.3d 349, 355 (5th Cir. 2009) (same). See also Michael H. LeRoy, Are Arbitrators Above the Law? The “Manifest Disregard of the Law” Standard, 52 B.C. LAW REV. 137, 180-81 (2011).
15 See Hall Street v. Mattel, Inc., 552 U.S. 576 (2008).
16 Oxford Health Plans LLC Sutter, 133 S. Ct. 2064, 2070-71 (2013).
17 See, g., N.H. Rev. Stat. § 542:8 (permitting courts to correct or modify an award for “plain mistake”). See also Stephen K. Huber, State Regulation of Arbitration Proceedings: Judicial Review of Arbitration Awards by State Courts, 10 CARDOZO J. CONFLICT  RESOL. 509 (2009); Hall Street Assocs. LLC v. Mattel, Inc., 552 U.S.  576, 590 (2008) (suggesting that the scope of judicial review might be different under state law); Schmidt v. UBS Fin. Servs., Inc., 10 N.E.3d 1145 (Mass. Ct. App. 2014) (§ 10 of the FAA does not apply in state court).
18 See Cable Connection, v. DIRECTV, Inc., 190 P.3d 586 (Cal. 2008).
19 Eighteen S. jurisdictions have adopted the Revised Uniform Arbitration Act of 2000 (Alaska, Arizona, Arkansas, Colorado, D.C., Florida, Hawaii, Michigan, Minnesota, Nevada, New Jersey, New Mexico, North Carolina, North Dakota, Oklahoma, Oregon, Utah, and Washington). Other states have a different statutory scheme, some modeled after the previous Uniform Arbitration Act. E.g., Del. Code tit. 10, ch. 57.
20 See, g., HL 1, LLC v. Riverwalk, LLC, 15 A.3d 725, 735 n.11 (Me. 2011).
21 See AAA, Optional Appellate Arbitration Rules, Rule A-1 (2013); Paul Marrow, A Practical Approach to Affording Review of Commercial Arbitration Awards: Using an Appellate Arbitrator, 60 DISP. RES. J. 10 (2005) (referring to rules of the International Institute for Conflict Prevention and Resolution, the Judicial Arbitration and Mediation Services, and the National Arbitration Forum).
22 AAA, Commercial Arbitration Rules and Mediation Procedures, Rule 46(b) (requiring a reasoned award if both parties request one).
23 See Symetra Nat’l Life Co. v. Rapid Settlements, Ltd., 2009 WL 1057339 (Tex. Ct. App. 2009) (assignee of payments from structured settlement violated public policy by using arbitration scheme to bypass state statutory requirement of judicial approval for the assignment); Symetra Nat’l Life Ins. Co. v. Rapid Settlements, Ltd., 657 F. Supp. 2d 795 (S.D. Tex. 2009) (factor was permanently enjoined from effectuating a transfer of rights to payment under structured settlements through the use of arbitration rather than by complying with state statutes on transfer); Symetra Life Ins. Co. v. Rapid Settlements, Ltd., 2011 WL 4807901 (S.D. Tex. 2011) (obligor on structured settlements had claim for tortious interference with contractual relations against assignee of structured settlement payments that attempted to use arbitration to avoid state statutes requiring court approval of the transfers because the assignee had no colorable argument that arbitration could be used in such a manner). See also Symetra Life Ins. Co. v. Rapid Settlements, Ltd., 599 F. Supp. 2d 809 (S.D. Tex. 2008); Symetra Life Ins. Co. v. Rapid Settlements, Ltd., 2007 WL 114497 (S.D. Tex. 2007); Fidelity and Guaranty Life Ins. Co. v. Harrod, 2007 WL 2847966 (Md. 2007); R & Q Reinsurance Co. v. Rapid Settlements, Ltd., 2007 WL  2330899 (S.D. Fla. 2007); Symetra Life Ins. Co. v. Rapid Settlements, Ltd., 2007 WL 1643211 (S.D. Tex. 2007); Allstate Settlement Corp. v. Rapid Settlements, Ltd.,  2007  WL  1377667  (E.D.  Pa.  2007);  In  re  Rapid Settlements, Ltd., 2007 WL 925698 (Tex. Ct. App. 2007) (refusing to order arbitration against issuer of annuity for structured settlement).
24 See Y. Ethics Op. 584 (1987); Alaska Ethics Op. 84-4 (both distinguishing between an illegal clause or contract and an unenforceable clause or contract). See also Greg M. Duhl, The Ethics of Contract Drafting, 14 LEWIS & CLARK L. REV. 989, 1012-17 (2010).
25 Christopher Drahozal and Stephen J. Ware, Why Do Businesses Use (or Not Use) Arbitration Clauses? 25 OHIO STATE J. ON DISP. RESOL. 433, 437(2010).