October 17, 2019 Feature

An Unhappy Bargain: The Downsides of Arbitration

The benefits of arbitration come at a price that can make it an unhappy bargain.

Katie Burghardt Kramer and Amiad Kushner

Download a printable PDF of this article.

Be careful what you wish for, lest it come true. And be forewarned: Arbitration is not the shimmering nirvana for parties in dispute. Yet, clients and their lawyers contract for it all the time and sometimes choose to submit to it even absent an agreement requiring that they do so.

While many have championed the cause of arbitration, arguing that it offers a streamlined, compelling, and economical alternative to litigation, that is true only to some extent. There are also some sneaky downsides and underappreciated risks, often ignored in the rush to embrace something other than the prospect of slogging through years of discovery and seemingly infinite pages of documents, with all of the resultant tedium and protracted motion practice.

We’ve all heard that one benefit of arbitration is that it is subject to only extremely narrow judicial review. In a word, it’s final.

That benefit carries a big price. There is no mechanism to reverse even the most egregious errors of fact or law by arbitrators. When you commit to arbitration, you’re agreeing to be bound by what may turn out to be—in the words of various courts—an “ugly,” “whacky,” or flat-out wrong decision.

As compared with a trial court decision that may be reversed by one or more levels of appellate courts, an arbitration award typically can’t be challenged on the merits. In fact, numerous courts have concluded that they were obligated to confirm arbitration awards that are simply wrong.

When your client agrees to arbitration, it is agreeing to accept that any decision, even an incorrect one, is final. As the U.S. Supreme Court has stated, “[t]he potential for . . . mistakes is the price of agreeing to arbitration” and thus, “[t]he arbitrator’s construction holds, however good, bad, or ugly.” Oxford Health Plans LLC v. Sutter, 569 U.S. 564, 572–73 (2013).

The Court of Appeals for the Seventh Circuit noted that “[f]actual or legal errors by arbitrators—even clear or gross errors—do not authorize courts to annul awards.” Gingiss Int’l, Inc. v. Bormet, 58 F.3d 328, 333 (7th Cir. 1995) (internal citations omitted). In a later decision, the Seventh Circuit boldly declared: “We will not overturn an award because an arbitrator committed serious error, or the decision is incorrect or even whacky.” Johnson Controls, Inc. v. Edman Controls, Inc., 712 F.3d 1021, 1025 (7th Cir. 2013) (internal citations and quotations omitted).

Be sure that you and your client understand that risk when you start down the road toward arbitration. As the courts have held for nearly a century, part of the bargained-for “value” of arbitration is that everyone is bound by the decision, right or wrong. See Matter of Pine St. Realty Co., Inc. v. Coutroulos, 233 A.D. 404, 407, 253 N.Y.S. 174 (1st Dep’t 1931) (“Errors, mistakes, departures from strict legal rules, are all included in the arbitration risk”).

In recent years, those principles have been consistently affirmed. For example, in 2018, the New York Appellate Division, First Department, reversed a lower court order that substantially vacated an award on the basis that the arbitrators had manifestly disregarded applicable law and misconstrued the procedural record. The court held that, regardless of whether such errors were made by the arbitrators, the trial court lacked authority to vacate the award. See Matter of Daesang Corp. v. NutraSweet Co., 167 A.D.3d 1, 22 (1st Dep’t 2018) (“A court is not empowered . . . to review the arbitrators’ procedural findings, any more than it is empowered to review the arbitrators’ determinations of law or fact.”). So long as the arbitrators’ decision was at least “barely colorable,” it must stand. Id. at 24.

It’s worth a quick review of the statutory standard for vacating an arbitration award. Under the Federal Arbitration Act (FAA), arbitration awards can be vacated only in extremely limited circumstances:

  1. where the award was procured by corruption, fraud, or undue means;
  2. where there was evident partiality or corruption in the arbitrators, or either of them;
  3. where the arbitrators were guilty of misconduct in refusing to postpone the hearing, upon sufficient cause shown, or in refusing to hear evidence pertinent and material to the controversy; or of any other misbehavior by which the rights of any party have been prejudiced; or
  4. where the arbitrators exceeded their powers, or so imperfectly executed them that a mutual, final, and definite award upon the subject matter submitted was not made.

9 U.S.C. § 10(a).

All of those grounds are narrow and basically procedural, rather than substantive or merits-based. Moreover, the courts have construed each of them narrowly, given the strong federal policy favoring arbitration. See, e.g., Rai v. Barclays Capital, Inc., 739 F. Supp. 2d 364, 371 (S.D.N.Y. 2010) (holding that the grounds for vacating an arbitration award have been “narrowly construed so as not to impinge on the broad discretion afforded to arbitrators”).

It’s likely that the error rate of arbitrators is at least as high as that of judges. Of course, because courts typically will not delve into the record to determine whether such errors occurred, even when merits-based arguments for vacating an award have been raised, there is no reliable record of material factual or legal errors committed by arbitrators.

But comparable litigation statistics suggest that the error rate in arbitration is probably substantial. For example, in the 12 months ending on June 30, 2018, about 12 percent of private civil actions ruled on by the U.S. Courts of Appeals were reversed. See U.S. Courts, Statistical Tables for the Federal Judiciary, U.S. Courts of Appeals tbl. B-5 (June 30, 2018) (www.uscourts.gov/statistics/table/b-5/statistical-tables-federal-judiciary/2018/06/30).

Arbitrators are not inherently less likely than judges to make “reversible” errors of fact or law. If anything, the quick and dirty procedures of arbitration might well lead to a higher error rate than in the court system. But unlike a judicial decision, which can be challenged on appeal, an arbitrator’s award almost always stands even if it is completely wrong on the law or the facts.

In subtle and often underappreciated ways, treating arbitration awards as sacrosanct leads to perverse outcomes. Arbitrators are well aware of the courts’ extreme deference to arbitration awards and thus, not surprisingly, have no incentive to produce thorough and well-supported awards. Consider that recent NutraSweet decision. There, the trial court determined that the arbitrators had manifestly disregarded New York law in holding that claims based on allegedly false representations and warranties were contractual in nature and thus could not be used as a basis for fraudulent inducement claims. See NutraSweet 167 A.D.3d at 17–19.

On appeal, the First Department noted that the arbitrators had relied heavily on a statement by the Second Circuit in Merrill Lynch & Co. Inc. v. Allegheny Energy, Inc., that “under New York law, parallel fraud and contract claims may be brought if the plaintiff . . . points to a fraudulent misrepresentation that is collateral or extraneous to the contract.” Id. at 17 (quoting Merrill Lynch, 500 F.3d 171, 183 (2d Cir. 2007)). The First Department noted that “[t]he meaning of the rule that an alleged misrepresentation is actionable as fraud if it is ‘collateral or extraneous to the contract’ . . . is not necessarily transparent from the quoted words alone and must be drawn out through detailed analysis of cases in which the rule has been applied.” Id. at 19 (emphasis added).

The court concluded that regardless of whether the arbitrators had misapplied that rule, they “did not manifestly disregard the law by according to the [Second Circuit’s language in Merrill Lynch] what may have reasonably seemed to them its natural meaning.” Id. (emphasis added). The clear implication is that the arbitrators did not conduct “a detailed analysis of cases in which the rule has been applied”—that is, they did not analyze precedents in the manner that a court applying New York law would be required to do.

Arbitration decisions generally are shorter and thinner than comparable court decisions. To be sure, many arbitrators take their jobs very seriously and devote extensive time and energy to rendering a decision. But arbitrators have no obligation to produce rigorous and well-supported decisions, and “[an] arbitration award must be upheld when the arbitrator offers even a barely colorable justification for the outcome reached.” Wien & Malkin LLP v. Helmsley-Spear, Inc., 6 N.Y.3d 471, 479–80 (2006) (citations, brackets, and internal quotation marks omitted).

Given the overwhelming weight of judicial policies favoring arbitration, arbitrators face no incentive to produce anything beyond rudimentary cutting and pasting of snippets of case law. As a result, when you reach the end of the arbitration road, the result might be a perfunctory and error-laden decision for which you have essentially no recourse.

Limited Discovery

Limited discovery in arbitration also creates particular perils. Although the prospect of limited discovery can seem enticing, often the ultimate costs are not worth the benefit. Not only do arbitration rules typically allow for only very narrow discovery, especially as compared with the broad discovery generally available in U.S. courts, but also some expedited arbitration procedures scrap discovery altogether.

In practice, such limits can result in the denial of access to critical discovery. In a recent proceeding, the arbitration provision in the parties’ agreement called for an expedited process that did not provide for any discovery. Although we suspected there were valuable documents we hadn’t seen, we had no effective means of accessing them. In a related court proceeding after the arbitration had concluded, we obtained documents that almost certainly would have improved our chances of success in the arbitration. Without robust discovery during the arbitration, we and our clients were handicapped significantly.

Even more worrisome, discovery limits in arbitration actually give parties an incentive to manipulate the process by engaging in strategic nondisclosure or selective disclosure. Unscrupulous parties can calculate that material information can be withheld or concealed without adverse consequences. Chances are low that you’ll ever know what you weren’t given.

In addition, the law provides just a short time to move to vacate an arbitration award after it issues. Under the FAA, for example, a party has only three months to move to vacate an award, hardly enough for any post-arbitration sleuthing. See 9 U.S.C. § 12. Similar time limits apply under most state laws. See, e.g., N.Y. C.P.L.R. § 7511(a) (90-day time period for moving to vacate).

The difference between typical arbitration discovery rules and the Federal Rules of Civil Procedure is stark. For example, under the Commercial Arbitration Rules of the American Arbitration Association (AAA), an arbitrator may—but is not required to—permit parties to propound requests for documents “reasonably believed by the party seeking the documents to exist and to be relevant and material to the outcome of disputed issues.” Rule 22(b)(iii).

Depositions are very rare as well. Rule L-3(f) of the AAA’s Procedures for Large, Complex Commercial Disputes specifically limits depositions to “exceptional cases . . . upon good cause shown and consistent with the expedited nature of arbitration.” By comparison, in U.S. courts, parties are generally entitled to obtain material documents and depositions without a court order.

While tedious at times, the litigation discovery process gives parties the opportunity to unearth “unknown unknowns,” to borrow from Donald Rumsfeld’s famous phrase. By contrast, in arbitration proceedings with no or severely limited discovery, the parties don’t have the chance to uncover unknown, yet potentially relevant, issues or documents.

The decision in Bain Cotton Co. v. Chesnutt Cotton Co., No. 5:11-CV-189-C, 2012 U.S. Dist. LEXIS 192977 (N.D. Tex. Oct. 17, 2012), highlights those risks. That dispute centered on contracts for the provision of cotton from particular acreage. The plaintiff alleged that the defendant breached the agreement by selling some cotton on the open market instead of giving it to the plaintiff. In the ensuing arbitration, the plaintiff requested that the defendant produce relevant records about the acres and crops that the defendant had planted. The defendant refused to produce that information. The arbitrators issued an award in favor of the defendant.

The plaintiff sought to vacate the award in Texas federal court on the basis that the arbitrators had failed to ensure that the plaintiff could obtain material discovery. The district court rejected the argument, noting “the Court is disinclined to disturb an arbitration award even if the parties did not receive the full measure of discovery and procedure as would have been obtained in a court setting.” Id. at *3.

The kicker came on appeal. The Fifth Circuit affirmed, holding that while the lack of discovery likely would have warranted reversal in a civil case, the plaintiff was out of luck in the arbitration context: “Had this discovery dispute arisen in and been ruled on by the district court, it is not unlikely that the denial of [plaintiff’s] pleas [for discovery] would have led to reversal; however, under the strong federal policy favoring arbitration, judicial review of an arbitration award is extremely narrow.” Bain Cotton Co. v. Chesnutt Cotton Co., 531 F. App’x 500, 500–501 (5th Cir. 2013) (emphasis added; quotations omitted).

Limited Ability to Assert New Claims

A related downside of arbitration is the limited ability to assert new claims or to implead third parties. That is in direct contrast to the liberal amendment standard in court. In fact, in federal and some state courts, even after the pleading stage, courts must “freely give” parties leave to amend their pleadings when justice requires, such as when material new facts or evidence is revealed.

There are generally no analogues in arbitration. For example, under Rule 6(b) of the AAA rules, once an arbitrator is appointed, “no new or different claim may be submitted except with the arbitrator’s consent.” Rule 10 of JAMS’ Comprehensive Arbitration Rules & Procedures is essentially the same: “After the Arbitrator is appointed, no new or different claim may be submitted, except with the Arbitrator’s approval.” Further, arbitration rules generally do not allow any claims to be asserted against third parties who did not sign the applicable arbitration agreement. Collectively, those provisions severely limit a party’s ability to assert new claims or defenses or implead third parties, even if new facts are uncovered that give rise to material new claims.

Indeed, an arbitrator lacks power to resolve a dispute that the parties have not submitted. As the Second Circuit has stated, “if arbitrators rule on issues not presented to them by the parties, they have exceeded their authority and the award must be vacated.” Fahnestock & Co., Inc. v. Waltman, 935 F.2d 512, 514 (2d Cir. 1991). An arbitrator who permits an amendment over the other party’s objection runs the risk that a reviewing court will question the arbitrator’s authority to decide the claim and, thus, the enforceability of any award.

In one arbitration, evidence produced during arbitration discovery revealed a massive fraud by one party that was unknown to the other parties when the arbitration commenced. That evidence was produced only after all the parties had agreed on the issues to be determined by the arbitrators. Thus, there was no procedural mechanism to assert new claims based on the alleged fraud. The issue never was presented to the arbitrators—and could not be—despite piles of uncovered evidence.

For an advocate, the relative difficulty of asserting new claims or defenses is an important disadvantage. Litigants don’t necessarily know all of their claims or defenses at the outset of a dispute, before any discovery has been had. Despite the touted flexibility of arbitration, in this area arbitration is decidedly less fluid and accommodating than litigation.


Often, arbitration clauses are intended to keep confidential potentially damaging disputes. In practice, however, that promise of confidentiality is hollow. Treat it as such.

Almost always, one can find a plausible basis to initiate court proceedings to compel or stay arbitration or seek other judicial relief. That court proceeding will make the confidential dispute public, threatening or causing embarrassment, harm to reputation, or other collateral damage. Whoosh—there go the benefits of confidentiality.

The recent dispute between a law firm and several of its former partners vividly illustrates the point. In April 2018, Quinn Emanuel Urquhart & Sullivan, LLP, initiated arbitration in California, under an arbitration provision in Quinn’s partnership agreement, against a group of former partners who had left the firm and formed a new litigation boutique. Quinn sought to recover a portion of fees collected from Quinn clients by the new firm.

The arbitration clause provided that “[a]ll matters pertaining to the arbitration shall be kept strictly confidential.” In May 2018, however, the former Quinn partners filed a petition in New York state court, seeking to stay and enjoin the arbitration on the basis that the relief sought in the arbitration violated New York attorney ethics rules. The petition attached a copy of Quinn’s arbitration demand and publicly disclosed internal communications related to the dispute.

Just like that, the illusion of confidentiality was shattered. As one would expect in a juicy litigator-versus-litigator battle, the lawsuit was widely covered in legal media. Any pretense of keeping the dispute confidential was thus lost. Though the New York court ultimately compelled the matter back into arbitration, the cat was out of the bag.

As the Quinn case demonstrates, there often are colorable arguments that would permit a party to seek relief in court, notwithstanding a provision requiring confidential arbitration or even the pendency of a confidential arbitration. And even if no one ever files suit, the mere existence of a plausible basis for publicly airing the dispute might impose greater pressure to settle. When a party really, really does not want the dispute made public, the myth of arbitration confidentiality plays an important role. Even if a court likely will compel arbitration or enforce an award, the threat of publicity may prompt early settlement.


So then, is arbitration still truly less expensive than litigation, or is that a fiction too? Depending on the dispute, perhaps. But in several essential ways, arbitration has an inflated cost structure and includes some hidden costs. The assumed money savings may not be as significant as expected.

First, while a judge is free of charge, parties to an arbitration must cover the arbitrator’s fees and expenses. If there’s a panel, there are fees and expenses for the multiple arbitrators. Often those amounts must be paid in advance or in periodic payments during the pendency of the arbitration.

Arbitrators can be expensive. In complex commercial cases, they may charge upwards of $1,000 per hour. That doesn’t include travel expenses or other incidental costs, and arbitral bodies such as AAA or JAMS charge significant filing and administrative fees. Those inherent arbitration costs alone can drive the parties toward settlement.

Second, in arbitration, the winning party often receives an award to recover its fees and costs, which may also include its legal fees. That potential for such fee shifting dramatically alters the cost calculus, particularly for parties accustomed to litigation in U.S. courts, where it’s not the norm.

Third, arbitration proceedings don’t always conclude once an award is issued. While some losing parties simply pay the arbitration award, an arbitration award is not enforceable unless it is converted to a court judgment, so even in the best of circumstances, ancillary court proceedings may well arise, carrying additional costs and compelling further attorney fees. Thus, proceedings to confirm or vacate an arbitration award often follow, setting in motion a whole new legal action. Petitions to confirm or vacate an arbitration award can be time-consuming to prepare and costly to initiate or oppose.

All those additional proceedings generate further legal fees and associated expenses. At every step, the client must fund the prosecution or defense. Typically, the risk or prospect of post-award proceedings is overlooked in calculating the relative cost of arbitration compared with litigation. Don’t fall victim to that myopic thinking.

In addition, your client may face exhausting litigation over the threshold question of whether the dispute should even be sent to arbitration. In Henry Schein, Inc. v. Archer & White Sales, Inc., 139 S. Ct. 524 (2019), the U.S. Supreme Court recently ruled on the narrow question of who properly decides that preliminary question. There, the dispute made its way through three levels of court review. Ultimately, the Supreme Court sent the matter back to square one, holding that, under the contract provision, it was the arbitrator—not the court—who had authority to decide whether the dispute was arbitrable.

Efficient? Not really at all.

When attorneys and commentators discuss the downsides of arbitration, they frequently focus on consumer or employment contexts. In those settings, individuals often are forced into arbitration to pursue their rights, and access to the public court system is denied to them. Those situations definitely raise important systemic considerations about balancing the rights of individuals against freedom of contract, while taking into account the perceived or portrayed efficiency of arbitration.

Yet even in the context of commercial arbitration between well-matched parties, there are significant and underappreciated risks and downsides. Consider and discuss them openly when weighing options and advising clients. In some cases, the very features of arbitration that are cited as attractive may lead to devastating outcomes. Given that, a clear-eyed assessment of arbitration risks is critical. Remember, be careful what you wish for. q

Katie Burghardt Kramer and Amiad Kushner

Katie Burghardt Kramer is the founder of KBK Law and of counsel to Dai & Associates. Amiad Kushner is a partner with Dai & Associates, New York City.

Copyright © 2019, American Bar Association. All rights reserved. This information or any portion thereof may not be copied or disseminated in any form or by any means or downloaded or stored in an electronic database or retrieval system without the express written consent of the American Bar Association. The views expressed in this article are those of the author(s) and do not necessarily reflect the positions or policies of the American Bar Association, the Section of Litigation, this committee, or the employer(s) of the author(s).