Just in time for the Halloween season, the Oklahoma Supreme Court gives us a scary tale about buying a new car.  In Sutton v. David Stanley Chevrolet, Inc., 2020 OK 87, ¶ 1 the Court finds that an arbitration clause in a consumer contract was induced by fraud because the structure of the transaction was misleading.


No, seriously.  It is a little scary, at least for any businesses that use form contracts.

Without too many spoilers, I think of the impact of this case in terms of a quote the Court uses:

The law is not designed to protect the vigilant, or tolerably vigilant, alone, although it rather favors them, but is intended as a protection to even the foolishly credulous, as against the machinations of the designedly wicked.

Adhesive contracting, the Court comes very, very, very close to saying amounts to “the machinations of the designedly wicked.”


Act I: Background

The relevant background is eerily familiar for those of us who’ve recently gone car shopping.  The plaintiff traded one car and purchased another.  Eighty-six pages later, he’d signed a bunch of times including four times on a two-page purchase agreement, which contained an arbitration clause.

Surprise, surprise, he didn’t read anything!  But I get ahead of myself.

Confusion and shenanigans ensued.  I won’t belabor you with the facts inciting the lawsuit, as honestly, they don’t make a lick of sense to me.  But basically, the trade-in car was never appropriately paid off, then it was stolen, plaintiff had to keep making payments on the stolen car, and eventually the plaintiff returned the new car that he’d purchased.  It all sounds very intense and messy.

Plaintiff sued and the dealership moved to compel arbitration.

Cue the creepy organ music!

Act II: The Fraud

According to the plaintiff, the arbitration agreement was induced by fraud.  There was no delegation provision. So, the separability doctrine wasn’t a problem because the plaintiff concentrated his allegations on the arbitration clause itself rather than the container contract.

What, you might ask, were the nefarious misdeeds of the dealership?  Well, the dealership didn’t outright lie to the plaintiff.  Moreover, the plaintiff conceded that all the signatures on the relevant documents were his and “he was not forced into executing [the purchase agreement].” And, the plaintiff admitted that he didn’t read the purchase agreement.

The creepy music crescendos!

The plaintiff maintained that the conduct of dealership’s finance manager combined with the very structure of the transaction created a false impression.

The creepy music abruptly ends with a kazoo noise.

Well, what did the finance manager do?  According to the plaintiff, he stated that a purpose of the purchase agreement was for verifying the plaintiff’s personal information, the vehicle information on both vehicles, and how much plaintiff would be paying.  He then pointed out the four places where the plaintiff was to sign: under a section entitled “vehicle purchased description,” under a section entitled “purchase price disclosure,” under a section entitled “security agreement,” and after two sections entitled “trade-in vehicle” and “dispute resolution clause.”  That’s it.

The arbitration agreement was in the dispute resolution clause.  It’s also worth noting that the dispute resolution clause was called out in red font, though the majority describes it as being “small.”  The dissents describe the provision a little differently.  One of them says that it was “the only provision in red ink, and it was located in the middle of the agreement.  The heading in all capital letters stated ‘DISPUTE RESOLUTION CLAUSE.’”

In any event, based on these facts, the Oklahoma Supreme Court sided with plaintiff.  It found that, in combination, the finance manager’s conduct and the structure of the transaction “created a false impression that the purpose of [plaintiff’s] signature was to only verify information concerning his trade-in vehicle. He surely was not under the impression he was agreeing to waive his right to a jury trial and obligating himself to pay a share of the costs of arbitration when he signed underneath the trade-in vehicle section of the purchase agreement.”

It was no defense, the Court said, that plaintiff didn’t read the purchase agreement.  “Every man or woman, even though illiterate, is presumed to know the contents of a written instrument signed by him; but no presumption of knowledge will stand in the way of a charge of fraud made in regard to the contents of the writing.”

Act III: The Significance of the Case

Holly wow!  I mean.  Seriously.  Wow.  All joking aside, this is a game-changer, at least in Oklahoma.  If other states borrow this sort of reasoning, it could revolutionize our approach to adhesive contracting.

Just to recap and generalize: a business presented a two-page, pre-printed form contract containing a variety of provisions, including an arbitration clause.  The business representative focused the customer on a few salient transactional provisions.  The business representative did not misrepresent any other provisions of the two-page document, but he also didn’t call out any other provisions.  The customer didn’t read the document but signed it in four different places.  Under these circumstances, the court found that the business had misled the customer.

If the Court is right, then pretty much every provision in a consumer transaction that a business representative does not specifically point out is vulnerable.

The Court tries to suggest that its ruling is narrower than all that.  But I don’t believe the Court.

It’s worth thinking about how the Court distinguished the security agreement.  Like the dispute resolution clause, the security agreement wasn’t discussed by the finance manager.  So, under the plaintiff’s logic, as the Court of Appeals below had noted, the security agreement could also be invalidated on the basis of fraud.

The Court brushed this point aside with two quick answers.  First, it said that the plaintiffs weren’t contending that the security agreement was induced by fraud.  That strikes me as a weak-sauce retort.  While true, in this case, it doesn’t change the fact that, if the plaintiffs had had some sort of reason to want to avoid the security agreement, they could have relied on the precise same logic.

Second, the Court says that the security agreement section had its own, stand-alone, signature line, which distinguished it from the dispute resolution clause.  The dispute resolution clause  shared a signature line with the trade-in car description.

So, does that mean that, if everything else was the same, the dispute resolution clause would have been enforceable if only it had its own stand-alone signature line?  Maybe, I suppose.  But that outcome seems pretty arbitrary.

How and why is a stand-alone signature line any better than the red-ink call out of the dispute resolution clause?  Both mechanisms are designed to focus the consumer’s attention on relevant details.  And neither mechanism actually worked.  The consumer didn’t read the signature lined sections that he signed.  He also didn’t read the only red-inked section.  He didn’t read anything.  So why make fine-grained distinctions between mechanisms that equally failed to call attention to a relevant provision?

Ultimately, this case is a spooky cautionary tale for lawyers creating standard form contracts for their business clients.

Welcome back to ArbitrationNation after a pandemic and protests hiatus.  I hope that you and your families are safe and that you’re confronting and coping with the injustices of our world.

I’m glad to have a good reason to write about arbitration again.  I’ve got a boatload of arbitration developments and cases to catch up on in the coming weeks.  But there’s no better day to get rolling than on a day that SCOTUS grants cert on a new(ish) arbitration case—Henry Schein, Inc. v. Archer and While Sales, Inc. Part II: Revenge of the Wholly Groundless Doctrine’s Zombie.

Yeah, okay, it’s a good thing I’m a law professor and not someone naming sequels.

Before getting to the new case, let’s have a quick refresher on Henry Schein Part I.  Henry Schein Part I focused on the wholly groundless doctrine.  The wholly groundless doctrine, you might recall, was basically a smell test for arbitrability.  It gave courts the right to regulate dubious arbitration agreements even if those agreements included a delegation provision.  A unanimous Supreme Court sounded the death knell on the doctrine.

But SCOTUS “express[ed] no view about whether the [particular] contract at issue in th[e] case in fact delegated the arbitrability question to an arbitrator.”  Accordingly, the Court remanded the case to the Fifth Circuit.  As I mentioned here, the Fifth Circuit on remand doubled down on its original conclusion.  It held that the contract at issue did not, in fact, assign arbitrability to the arbitrator.  See Archer and White Sales, Inc. v. Henry Schein, Inc., 2019 WL 3812352 (5th Cir. Aug. 14, 2019).

The court agreed that there was a valid delegation clause (through the AAA rules – Rule 7(a)).  But the claimant sought, at least partially, injunctive relief, and the arbitration clause carved out “actions seeking injunctive relief.”  The Fifth Circuit determined that the delegation provision did not “clearly and unmistakably” assign questions about the arbitrability of injunctive relief to the arbitrator, so a court got to decide if the claimant had to arbitrate.  It decided the claimant did not have to arbitrate.

Today, SCOTUS took the case up again.  Essentially, the issue presented is whether an arbitration clause containing a carve-out provision necessarily requires a court to determine whether the claims fall within the scope of the arbitration agreement before following the complying with a delegation provision.  To date, courts have split on this issue.

Delaware and the Second Circuit take the view that a carve-out generally requires a court to decide whether the dispute falls within the carve-out or not.  The delegation provision only becomes relevant after this threshold court determination.  See James & Jackson, LLC v. Willie Gary, LLC, 906 A.2d 76 (Del. 2006) (In the presence of a carve-out, “something other than the incorporation of the AAA rules” was “needed to establish that the parties intended to submit arbitrability questions to an arbitrator.”); NASDAQ OMX Group, Inc. v. UBS Securities, LLC, 770 F.3d 1010 (2d. Cir. 2014) (“The presence of the carve-out provision. . . delays application of AAA rules until a decision is made as to whether a question does or does not fall within the intended scope of arbitration, in short, until arbitrability is decided.”).

In contrast, the Ninth Circuit and Kentucky find that a delegation provision includes the question of whether a dispute falls within the carve out or not.  See Oracle America, Inc. v. Myriad Group A.G., 724 F.3d 1069, 1076-1077 (9th 2013) (determining that incorporation of the AAA rules unmistakably delegated questions of arbitrability to the arbitrator and rejecting the argument that the carve-out provision negated that delegation); Ally Align Health, Inc. v. Signature Advantage, LLC, 574 S.W.3d 753, 756-758 (Ky. 2019) (“A carve-out provision . . . does not negate the clear and unmistakable mandate of the AAA’s [r]ules that the arbitrability of claims is to be decided by an arbitrator [because to hold] the opposite would conflate the two separate and distinct questions of (1) who decides what claims are arbitrable with (2) what claims are arbitrable.”).

And that brings me back to my admittedly terrible sequel titles.  But I stand by the notion that the Fifth Circuit wants to resurrect the wholly groundless doctrine in a stronger, more potent form.  Under the original wholly groundless doctrine, a court could override a delegation provision if, but only if, a party seeking to compel arbitration had a frivolous interpretation of arbitrability.  Under the standard that the Fifth Circuit now proposes, a court could override a delegation provision anytime it simply disagrees with a party’s interpretation of the scope of the arbitral clause.

So, like all too many sequels, I’m afraid that the original will probably be a lot better and more interesting.  I’m getting out my crystal ball and making a bold predication— I think SCOTUS will overturn the Fifth Circuit once again.  In fact, I wouldn’t be surprised if it’s another unanimous decision.

I have to confess something: I just returned from sunny California where I attended an excellent arbitrator training course put on by the American Arbitration Association and run by Dana Welch and Michael Powell!  If you have an opportunity to take a course from either of them, I highly recommend it.

And, as it so happens, you can!  Check out my prior announcement of the ABA’s Arbitration Institute this March in Phoenix.  Dana Welch will be there along with a number of other exceptional instructors.  I’ll give you a more formal reminder soon, but it’s well worth your time to think about attending if you can.  (The Early Bird registration ends on February 1, so think about it soon.)

Anyway, speaking of confessions and Arbitration 101 . . .   in catching up on my arbitration cases on the plane ride home, I read a recent and cautionary tale by Ninth Circuit about disclosures.

In Monster Energy Co. v. City Beverages LLC, 940 F.3d 1130 (9th Cir. 2019), the court vacated an arbitration award because the arbitrator, a retired California state judge, failed to disclose both his ownership interest in JAMS and the fact that JAMS had administered 97 arbitrations for Monster during the previous five years.

The underlying dispute doesn’t matter much, but, in brief, Monster terminated a distribution agreement with its franchisee.  The franchisee protested under Washington law, and Monster initiated arbitration administered by JAMS.

The important stuff starts with the arbitrator’s disclosure: “Each JAMS neutral, including me, has an economic interest in the overall financial success of JAMS. In addition, because of the nature and size of JAMS, the parties should assume that one or more of the other neutrals who practice with JAMS has participated in an arbitration, mediation or other dispute resolution proceeding with the parties, counsel or insurers in this case and may do so in the future.”

Getting no objections from the parties about partiality after this disclosure, the arbitrator went on to consider the merits and ruled against the franchisee.  Monster sought to confirm the award, but the franchisee cross-petitioned for vacatur.  The franchisee had learned that the arbitrator was a co-owner of JAMS.  (According to the Ninth Circuit, only about 1/3 of all JAMS Neutrals have such an ownership interest.)  It also found out about the much more robust relationship between JAMS and Monster than the arbitrator had revealed.  The district court confirmed the award, but the Ninth Circuit reversed and vacated.

Particularly significant to the outcome was the fact that JAMS had repeat business with Monster.  The court noted that “[c]lear disclosures by arbitrators aid parties in making informed decisions among potential neutrals. These disclosures are particularly important for one-off parties facing ‘repeat players.’”  The decision then cited to a study by Professor Lisa Bingham showing that employees disproportionately failed to recover damages against repeat-player employers compared to non-repeat-player employers. See Lisa B. Bingham, Employment Arbitration: The Repeat Player Effect, 1 Emp. Rts. & Emp. Pol’y J. 189, 209–17 (1997).  The dissenting Judge goes even further in recognizing the importance of the repeat player effect to the outcome of the case: “arbitrators have incentives to make decisions that are viewed favorably by parties who frequently engage in arbitrations. This feature of private arbitration, even if distressing, is an inevitable result of the structure of the industry.”

The moral of the story: Arbitrators should disclose.  Disclose everything.  If you, as an arbitrator, think of something and wonder if you should disclose it, disclose it.  If it crosses your mind, disclose it.  That might seem extreme, but as the court points out,  “[i]t is simplicity itself, and no real burden, for an arbitrator to disclose his or her ownership interest in an arbitration company for which he or she works, as well as the organization’s prior dealings with the parties to the arbitration.”

That reminder about the importance of disclosure aside, I want to close by taking a moment to think more about the so-called repeat player effect, especially based on studies in the employment context. While evident partiality is, in large measure, about the reasonable impression of bias, and it seems pretty clear that repeat player concerns can create such an impression, the existence and magnitude of any repeat player effect in arbitration has not been decisively established.

In fact, there’s been a lot of debate among academics about this effect.  While the notion has some intuitive appeal, the empirical jury is still out (pardon the metaphor in an arbitration blog).  See, e.g., Estreicher, Samuel and Heise, Michael and Sherwyn, David, Evaluating Employment Arbitration: A Call for Better Empirical Research, 70 Rutgers U. L. Rev. 375, 386-87 (2018).  For instance, the repeat-player studies usually make no attempt to control for the size and claims experience of repeat-player employers.  Larger employers have more resources and experience.  Employees are likely to fare worse against these employers as compared to smaller employers with fewer resources and less claims experience regardless of whether the claim is brought in court or before an arbitrator. More importantly, strong cases on the merits are likely to settle and never reach a hearing.  So, win rates for employers with repeated cases that go to award should be high.



Welcome to 2020!

I hope that you all had a safe and rejuvenating holiday season.  A new decade brings us plenty of new opportunities for thrilling arbitration news and developments!

But, up first, more on class arbitrability.  I know.  I know.  So last decade.  But trust me, this is a case you want to keep an eye on, Jock v. Sterling Jewelers Inc., 942 F.3d 617 (2d Cir. 2019).  I’m not a betting fellow, but if I were, I’d put money on the issues raised by the case inspiring cert in the near future.

The Sterling Jewelers matter has found its way to the Second Circuit four times, so this isn’t exactly a “new” case.  (Liz wrote about it, briefly, here and here.) But the most recent ruling puts front and center a very hot circuit split and a novel but related issue about class arbitration, a perennial fascination of SCOTUS.

The underlying dispute involves tens of thousands of female retail sales employees accusing Sterling of paying them less than men doing the same work.  All Sterling employees were required, as a condition of employment, to sign an agreement mandating that they participate in arbitration.

Back in 2010, an arbitrator concluded that 254 claimants could proceed with class arbitration under this agreement. A Southern District of New York court disagreed, but the Second Circuit sided with the arbitrator.  It concluded that the lower court had impermissibly conducted a merits review of the legal analysis on the class arbitrability question rather than deciding if the arbitrator had the power to decide that question in the first instance.  See Jock v. Sterling Jewelers Inc., 646 F.3d 113 (2d Cir. 2011) (“Jock I“).  Basically, the parties had clearly submitted the question of class arbitrability to the arbitrator, so the Second Circuit said it was up to the arbitrator.

(For those keeping track, this is where the hot Circuit split comes into play: recall that the question of whether a delegation clause may assign class arbitrability to an arbitrator has created a huge mess.  See my prior post here.)

Following Jock I, the arbitrator issued a class certification award joining approximately 44,000 women into the arbitration proceeding.  That was a lot more than the original 254 claimants, and it included a bunch of absent class members. Sterling attempted to vacate the award, but the district court refused, believing that Jock I foreclosed the issue.

Turns out that the district court wasn’t very good at reading the Second Circuit’s tea leaves.  Once more, the Second Circuit reversed and this time remanded.  Jock v. Sterling Jewelers Inc., 703 F. App’x 15 (2d Cir. 2017)(summary order) (“Jock II“).  In Jock II, the Second Circuit clarified that its holding in Jock I didn’t address the issue of whether the arbitrator could bind absent class members to class arbitration given that they never consented to the arbitrator’s power in the first place.

On remand, the district court gave it another shot.  It vacated the class arbitration award.  Basically, it restated its original legal conclusion that the arbitration agreement didn’t permit class arbitration.  Then it relied, essentially, on the old saw that arbitration is a creature of contract.  Even if the arbitrator’s erroneous legal conclusion that class arbitration was allowed would bind the original 254 claimants and the company, because they’d committed that question to the arbitrator, it couldn’t bind the tens of thousands of absent class members who had not consented to the arbitration.  Only parties who bargain for the arbitrator’s decision can be bound by it.

Hat Trick! The district court got reversed for the third time!

So, what’d the district court miss this last go ‘round?

Well, first, it’s worth noting that the Second Circuit doubled down on the notion that delegation clauses can empower arbitrators to decide matters of class arbitrability.  According to the court, incorporation of the AAA Supplementary Rules for Class Arbitration, Rule 1(a) (“the arbitrator shall determine as a threshold matter . . . whether the applicable arbitration clause permits the arbitration to proceed on behalf of . . . a class”) means that the question of class arbitrability is to be decided by the arbitrator.

This part of the decision isn’t surprising.  The Second Circuit has, for some time now, stood by its conclusion that questions of class arbitrability can be delegated to the arbitrator.  In the Second Circuit, class arbitrability isn’t special.  It’s just like any other issue of contractual arbitrability.

But the stuff about binding absent class members who didn’t opt into arbitration . . . . now that’s big news.  According to the Second Circuit, although absent class members didn’t affirmatively opt into this particular arbitration proceeding, by signing Sterling’s employment agreement, they consented to the arbitrator’s authority to decide the threshold question of whether the agreement permits class arbitration.  Accordingly, “[b]ecause the absent class members, no less than the parties, . . . ‘bargained for the arbitrator’s construction of their agreement’ with respect to class arbitrability, the arbitrator acted within her authority in purporting to bind the absent class members to class procedures.”

Whoa! This holding gives some serious teeth to class arbitrability determinations by an arbitrator.  It also raises profound due process issues.

The Second Circuit brushes those issues aside, saying that “[c]lass actions that bind absent class members as part of mandatory or opt-out classes are routinely adjudicated by arbitrators and in our courts.”  But this overlooks or, at least, minimizes the persistent due process concerns that SCOTUS has articulated with class arbitrations.  Most recently, in Epic Systems, SCOTUS reiterated the problems associated with arbitrators having to decide whether the named class representatives are sufficiently representative and typical of the class and what kind of notice, opportunity to be heard, and right to opt out absent class members should enjoy. FRCP 23, governing class actions in federal courts, is aimed at providing sufficient procedures to satisfy concerns that absent class members have their rights determined without having an opportunity to be heard.  See, e.g., Hansberry v. Lee, 311 U.S. 32 (1940).

All in all, Jock presents some issues that are almost certain to get considered by SCOTUS, if not on appeal from this dispute, in some very similar situation in the near future.

Seems like I’m picking on the gig economy these days.  I really don’t mean to be.  But a former research assistant of mine brought an important, hot-off-the-presses decision to my attention, O’Hanlon v. Uber Techs., Inc., No. 2:19-cv-00675, 2019 BL 434840 (W.D. Pa. Nov. 12, 2019).

The case presents a couple of important Arbitration 101 reminders, including one about equitable estoppel in the context of arbitration.

The plaintiffs brought a class action against Uber, alleging that it violated Title III of the Americans with Disabilities Act because it failed to provide any wheelchair accessible vehicles through its on-demand ridesharing service in Pittsburgh.  There have been a number of related suits against Uber and Lyft recently, which also makes the case noteworthy.  (Hat tip to my former research assistant, Beau RaRa, and the Minnesota Disability Law Center.)

Uber responded by trying to compel arbitration.  The court refused.  The holding is straight-forward enough: the plaintiffs were all folks who had never downloaded the Uber app and had never booked an Uber ride before.  So, they never consented to the arbitration provision.  Cite Volt Info. Sci., Inc. v. Bd. of Tr. of Leland Stanford Junior Univ., 489 U.S. 468 , 478 , 109 S. Ct. 1248 , 103 L. Ed. 2d 488 (1989) (“the FAA does not require parties to arbitrate when they have not agreed to do so”).  Mic Drop.  Boom.  Arbitration 101.  How you like me now, Uber?

What makes the case more interesting from an arbitration law perspective, however, is Uber’s clever, though failed, equitable estoppel argument.  Uber tried to convince the court that the plaintiffs’ claims necessarily “implicated” Uber’s terms of use.

Let’s dig into that argument, as a refresher on equitable estoppel in arbitration.  (Importantly, not all courts are in agreement about how various flavors of equitable estoppel should work in arbitration, but it’s pretty clear that it’s a slightly different doctrine than applies to regular ol’ contracts.) When a plaintiff brings a claim which relies on contract terms against a defendant, the plaintiff may be equitably estopped from denying the effect of the arbitration clause contained in that agreement.  This sort of equitable estoppel is sometimes referred to as the “direct benefits” theory of estoppel.  It “prevents a nonsignatory from knowingly exploiting an agreement containing the arbitration clause.” Graves v. BP Am., Inc., 568 F.3d 221, 223 (5th Cir. 2009). That is, “a nonsignatory cannot sue under an agreement while at the same time avoiding its arbitration clause.” Id.

Importantly, as a closely related case from the Northern District of California last year notes, the doctrine is, well, equitable.  See Namisnak v. Uber Technologies, Inc., 315 F.Supp.3d 1124 (N.D. Cal. 2018).  As the court said, “’[t]he linchpin for equitable estoppel is equity—fairness,’ and ‘the application of the doctrine is fact-specific.’” (citations omitted).

In O’Hanlon, the court didn’t buy Uber’s argument.  Borrowing from Namisnak, the court concluded that the plaintiffs asserted rights created by the ADA, which are not dependent on or bound up with the terms and conditions of Uber’s service.  Essentially, according to the plaintiffs, Uber was refusing to comply with the ADA not breaching the terms of its contract with riders.

I’ll just briefly conclude by pointing out that Uber also advanced a novel standing argument, predicated on pretty much the same idea.  Basically, Uber said that the plaintiffs necessarily embraced the terms of use for their benefit in pursuing standing.  Although none of the plaintiffs were actually Uber customers, they must have been thinking about becoming customers in order to have any injury.  Thus, Uber tried to say, the plaintiffs needed to stand in the shoes of actual customers who would have been bound to Uber’s terms of use, including the arbitration provision.

The court didn’t buy this either.  It cited to “futile gesture” caselaw on standing and concluded that the plaintiffs “deterrence-based injury is actual and cognizable and their own.”

Happy December!  I hope that everyone has had a restful and well-earned holiday weekend break.

There’s a lot of new and exciting stuff happening in the world of arbitration, and I have some catching up to do.  I want to start, though, in an unorthodox place.

We rarely write about early litigation actions on this blog, but there’s something very interesting happening in California.  A law firm there has taken action to protect its effort to engage in mass individual arbitrations on behalf of a large group of clients.  In two different actions – one in the California Superior Court (Boyd v. DoorDash, Inc., Case No. CPF-19-516930) and one in the federal court for the Northern District of California (Abernathy v. DoorDash, Inc., CASE NO. 3:19-cv-07545-WHA) – several thousand DoorDash couriers are seeking TROs to prevent DoorDash from changing the terms of its arbitration agreement with each of them.

Substantively, the courier’s claims look familiar.  They echo a torrent of similar claims asserted by gig-economy workers recently.  The couriers argue that they have been misclassified as independent contractors when, in fact, they are employees.  Whatever one makes of the merits, DoorDash makes its “Dashers” agree to a broadly worded arbitration agreement that covers such claims and excludes class proceedings.  The original version of this arbitration agreement required arbitration pursuant to the AAA’s Commercial Arbitration Rules.

The Dashers’ law firm decided to embrace the arbitration agreement and initiate, cookie-cutter style, thousands of individual arbitrations against DoorDash.  In the federal case, pursuant to the applicable AAA fee schedule, DoorDash was accordingly billed approximately $11 million for initial up-front administrative fees.

Viewing this situation as a “shakedown” (this is what it called the situation in its brief opposing the TRO in federal court), DoorDash refused to pay.  It argued that the filings were deficient, although the AAA determined that the claimants had satisfied the minimum filing requirements.  Ultimately, because the fees were not paid, the AAA administratively dismissed the Dashers’ claims on November 8, 2019.

On November 9, DoorDash revised its arbitration agreement, changing providers from the AAA to the CPR. (The CPR has much lower administrative filing fees and has instigated a new process for dealing with these sorts of “mass” individual arbitrations.  See https://www.cpradr.org/dispute-resolution-services/employment-related-mass-claims-documents/emp-mass-claims-protocol.)  Any Dashers logging into their app to accept deliveries after November 9 consented to the new agreement, including any of the Dashers who had attempted to initiate arbitration before the AAA but who had their claims administratively dismissed.

The hearings on the TROs have taken place, though the courts have not yet ruled.

This situation raises at least three important issues.  First, how viable is the Dashers’ clever end-run around class action waivers?  DoorDash, not surprisingly, dislikes the tactic.  It asserts that these sorts of “mass” individual arbitrations have “wreaked havoc on the arbitration system.”  But that’s far from clear to me.  Each arbitration remains individual, even if it’s essentially a clone.  In the first ArbitrationNation Bookworm entry, I recommended an article by Andrea Cann Chandrasekher and David Horton,  Arbitration Nation: Data from Four Providers.  In that article, the authors specifically suggest that this sort of mass individual arbitration approach could be ameliorating some of the perceived shortcomings of arbitration in the consumer, employee, and patient contexts.

Second, when a party agrees to arbitrate with a particular institution, can that party effectively avoid arbitration by failing to pay the required administrative fees?  I know, I know.  This is something of an over-simplification, as DoorDash tries to thread a more precise needle here.  It says that the filing requirements weren’t satisfied so the arbitrations weren’t technically initiated.  But that seems like a tough argument to stomach, given that the AAA concluded the minimal filing requirements were satisfied.  Moreover, that seems like the sort of substantive argument that should be heard by the arbitrator.  (I wrote about a similar issue in an Eleventh Circuit decision —   Freeman v. SmartPay Leasing, LLC, 771 Fed.Appx. 926 (11th Cir. 2019) – a few months ago.)

Finally, can a corporate party change the terms of its arbitration agreement after arbitrations have been initiated under the old agreement?  The answer seems pretty clear-cut – no – but this DoorDash situation makes things complicated.  The initial arbitrations were administratively dismissed.  Even if DoorDash was wrong not to pay the filing fee, what’s the right remedy at this stage?  Does DoorDash have to remain bound to its original arbitration agreement?

I’ll keep a close eye on these cases and provide an update as soon as there’s a decision.

I don’t mean to be imprecise, but I think that the Eleventh Circuit may have recently issued the most luddite opinion I’ve seen in a good long while.  See Managed Care Advisory Group, LLC v. CIGNA Healthcare, Inc., 2019 WL 4464301 (11th Cir. Sept. 18, 2019).  According to the court, Section 7 of the FAA, which allows arbitrators to subpoena non-parties and their documents, must be interpreted narrowly.  And when I say narrowly, the Eleventh Circuit isn’t joking around.

Arbitrators can only require summonsed non-parties to appear in the physical presence of the arbitrator.  This means literal physical presence; none of this video conference nonsense.  Additionally, Section 7 prohibits any pre-hearing discovery, which means that documents from non-parties can only be obtained when they physically show up to a hearing.

The underlying dispute isn’t particularly important, but briefly, just for context, it involved arbitration over a settlement agreement.  Essentially, a class of medical providers had sued managed care insurance companies, alleging that the insurers improperly processed and rejected certain physicians’ claims for payment.  A big kerfuffle ensued.  Ultimately, the case settled.  Sometime after the settlement, a group acting on behalf of the class members submitted to an arbitration agreement with one of the insurers.  They were trying to resolve a dispute over a portion of the settlement funds.  Eventually, during the course of an arbitration, the arbitrator summonsed several non-parties to appear for a live hearing and video conference and to bring certain documents along with them.  The non-parties objected to the summonses.

After dealing with some preliminaries – jurisdiction, the nationwide service of arbitral summonses, and the correct court to enforce arbitral summonses – the court gets to the big bang.  It starts by quoting Section 7, which allows an arbitrator to “summon in writing any person to attend before them . . . as a witness and in a proper case to bring with him . . . any book, record, document, or paper which may be deemed material as evidence in the case.”  The text laid out, the Eleventh Circuit turns to a plain-meaning and quaintly originalist reading.

The court observes that “Congress passed Section 7 in 1925, so we must ascertain the meaning of ‘attendance’ and ‘before’ in Section 7’s grant of authority . . . in the same manner provided by law for securing the attendance of witnesses . . . in the courts of the United States” as of 1925.  Since there was none of this new-fangled video conference rubbish in 1925, Section 7 of the FAA doesn’t include technological means of having witnesses “attend” the hearing.  In short, “Section 7 does not authorize district courts to compel witnesses to appear in locations outside the physical presence of the arbitrator. . . .”

Moreover, a simple reading of Section 7, according to the Eleventh Circuit, only allows arbitrators to compel non-parties to come to a hearing with documents.  It doesn’t allow arbitrators to compel non-parties to produce documents before such a hearing.  “Thus, the FAA implicitly withholds the power to compel documents from non-parties without summoning the non-party to testify.”  And, even if the non-party is summoned to testify, the FAA does not allow “pre-hearing” discovery.

It’s worth noting that the Eleventh Circuit aligns itself with the Second, Third, Fourth, and Ninth Circuits.  But the Eleventh Circuit is a bit cheeky.  It insinuates that these other circuits agree with both prongs of its holding: (1) arbitrators can only compel non-parties to attending in-person, physical hearings in the same room as the arbitrator; and (2) the FAA does not permit pre-hearing discovery from non-parties.  In fact, these other circuits seem to only endorse, at least expressly, the second point.  (Then-judge Alito writing for the Third Circuit did say that Section 7 “unambiguously restricts an arbitrator’s subpoena power to situations in which the non-party has been called to appear in the physical presence of the arbitrator and to hand over the documents at that time.”  But the Third Circuit was not faced with non-parties being summoned to a video conference.)

The Eighth Circuit appears to be the outlier with respect to this second point, reasoning that, well, times have changed and efficiency interests suggest that arbitrators should have the authority to require pre-hearing production of evidence from non-parties.

With respect to the Eleventh Circuit’s first holding, I’m not aware of any other decisions to say that arbitrators and non-parties have to be in the same physical space.  I confess, that feels unnecessarily technophobic to me.  The Eleventh Circuit says that the second holding – arbitrators cannot compel pre-hearing discovery from non-parties – compels the first – arbitrators and non-parties need to be in the same room.  Otherwise, how would the arbitrators get any documents that non-parties bring with them?  But Really?  I teach students regularly who are not in the same physical room as me.  And I can easily share, electronically, documents or other information with them in real time.

I don’t mean to be cheeky myself, but this all seems silly.  And radically inefficient.  And to miss the point of much modern discovery, in a world where “documents” are often anything but paper.  Taken to its extreme, the Eleventh Circuit’s decision could mean that arbitrators lack the power to compel non-parties to produce any form of ESI (“electronically stored information”), at least to the extent that the non-parties couldn’t “bring it with them” to the hearing.

This decision is radically anti-arbitration.  To be sure, the notion that non-parties cannot be compelled to produce pre-hearing discovery isn’t entirely new, though it too seems anti-arbitration.  But this appended requirement of physical presence at a hearing . . . I’m flummoxed.

Hat tip to Erika Birg, a partner at Nelson Mullins in Atlanta, for highlighting this important case for me.

For the next installment of the Bookworm, I’m recommending a very recent article by Professor Jean Sternlight: Mandatory Arbitration Stymies Progress Towards Justice in Employment Law: Where To, #MeToo?.

For anyone who isn’t already familiar with her work, Professor Sternlight has been at the forefront of thinking about adhesive arbitration for at least two decades.  Her  articles are at the bedrock of any critical appraisal of consumer, employee, and patient arbitration.  Her 1996 piece, Panacea or Corporate Tool – Debunking the Supreme Court’s Preference for Binding Arbitration, defined, in many ways, the terms of the debate that’s been raging ever since.  And her 2005 article, Creeping Mandatory Arbitration: Is it Just? constitutes, in my mind, one of the most cogent criticisms of so-called “mandatory arbitration” that has been written.

I don’t always agree with Professor Sternlight.  But it’s impossible not to respect the quality of her thinking and the passion of her arguments.

Mandaotry Arbitration Stymies Progress continues her important and thoughtful work.  In her own words, here’s a summary:

If companies can continue to use mandatory arbitration to eradicate access to court, where judges are potentially influenced by social movements, social movements will no longer be able to assist the overall progressive trend of our jurisprudence. . . . [T]he current and powerful #MeToo movement offers a perfect, albeit depressing, case study.  While the #MeToo movement has already exposed many sordid high-profile incidents of alleged harassment, sparked substantial outrage in traditional and social media, and become a talking point in public events and workplaces throughout the country, for the most part this outrage has not yet trickled down to protect ordinary women (and men) in ordinary workplaces. To the contrary, the law of sexual harassment still has a long way to go to catch up with the sentiments being expressed in the #MeToo movement. In the past, one might have expected that the new cultural attitudes surrounding sexual harassment might lead courts to rethink some of their prior restrictive decisions on sexual harassment. However, to the extent that employers are using mandatory arbitration to keep employment disputes out of court, even as powerful a social force as the #MeToo movement may not produce the progressive legal changes one might otherwise have expected. What is true of the #MeToo movement is true of other existing and potential forces for social change as well, such as social movements that might advocate for greater diversity, privacy, or income equality. To the extent companies are permitted to use arbitration to eliminate access to courts, they prevent our law from evolving to become more just.

Four weeks ago, the boundary between public enforcement and private dispute resolution became more blurred.  On September 4, the Justice Department announced that it had agreed to binding arbitration on the key issue in a current merger case—the market definition.

The enforcement action is garden variety.  It challenges Novelis Inc.’s proposed acquisition of Aleris Corporation.  According to the DOJ, the transaction would combine two of only four North American producers of aluminum auto body sheet, which automakers use to produce aluminum parts for automobiles.

But the use of arbitration by the DOJ constitutes a novel use of the Antitrust Division’s authority under the Administrative Dispute Resolution Act of 1996, 5 U.S.C. § 571 et seq.  Although the option has theoretically existed for 23 years, the DOJ has never used it before.

DOJ Antitrust Chief Makan Delrahim suggested, during a speech at George Washington University Law School’s annual Antitrust Salon, that this case “could prove to be a model for future enforcement actions, where appropriate, to bring greater certainty for merging parties and to preserve taxpayer resources while staying true to our enforcement mission.”

He went on to justify the experiment on both efficiency and accuracy grounds.  He noted that antitrust enforcers must be “more attuned to ensuring an efficient process for resolving merger and conduct investigations and, when necessary, litigations.”  He cautioned, however, that “efficiency should not come at the expense of achieving the right result.  Rather, we always should be open to process improvements that can result in economically sound outcomes that are achieved in a more efficient manner.”

If you’re a regular reader of the blog, you know of my abiding belief in the many virtues of arbitration.  Even in the tricky context of adhesive contracts with consumers, employees, and patients, I’m not persuaded that arbitration always poses the grave concerns that some suggest.  But this decision by the DOJ puts even me on high alert.

I’ll refrain from too much commentary, but suffice it to say that arbitration’s virtues extend to resolution of particular disputes.  Arbitration can get the job done at a low cost and in an effective way.  Arbitration, however, does little (and maybe nothing) to promote social policy objectives outside of dispute resolution.  Regulatory enforcement actions, while civil in nature, seem to me to be intended to serve additional public purposes beyond mere resolution of a particular dispute.

This isn’t a simple issue, and I appreciate the DOJ’s desire to streamline a part of antitrust enforcement actions that has proven daunting to generalist judges and lay juries.  Still, the idea that a critical component of enforcement actions could be shuffled off for resolution in private or, at least, outside of the standard public system, gives me pause.

The Third Circuit welcomed us to the fall arbitration season with an important decision for the gig economy, Singh v. Uber Techs. Inc., 2019 WL 4282185 (3d Cir. Sept. 11, 2019).  Relying on the key logic of SCOTUS’s January ruling in New Prime, Inc. v. Oliveira, the Third Circuit concluded that Uber drivers may qualify FAA § 1’s exemption for “any other class of workers engaged in foreign or interstate commerce.”  I say “may qualify” because the Third Circuit technically remanded the case.

This is an important one, so it’s worth thinking through it carefully.  There at least three key takeaways from the case: (1) workers may qualify for the § 1 exemption if they belong to a class of workers moving passengers or goods in interstate commerce; (2) the determination of whether workers fall within such a class hinges on consideration of a non-exclusive list of factors; and (3) lower courts can and should demand discovery necessary to make this factor-based determination.

Exciting stuff!  Let’s dive in!

The facts are simple: a New Jersey Uber driver brought a putative class action in state court.  He alleged that Uber misclassified drivers as independent contractors rather than employees.  That deprived the drivers of over overtime pay and it forced them to incur business expenses that Uber should have paid.  Uber removed the case.  Then it sought to compel arbitration on an individual basis.  The employee resisted on a number of grounds, including that the contract with Uber fell within the exemption of FAA § 1.  The district court, however, decided that the employee did not qualify for the exemption.  It reasoned that the exemption only applies to workers who transport goods, not passengers.  The district court then rejected the employee’s other objections and sent the case to arbitration.

That sets the stage for the first of the three big takeaways: according to the Third Circuit, the FAA § 1 exemption “may extend to a class of transportation workers who transport passengers, so long as they are engaged in interstate commerce or in work so closely related thereto as to be in practical effect part of it.”  This is huge.  Uber argued vigorously that the exemption should be narrowly construed to apply only to transportation workers moving goods.  Some dicta would seem to have supported that proposition.  But the Third Circuit roundly rejected it.

That, in turn, raised the second big takeaway: the lower court needs to evaluate various factors to determine if particular employees belong to a class of employees engaged in interstate commerce.  Notice the phrasing here.  The question isn’t whether the particular workers were engaged in interstate commerce.  It’s whether the particular workers belong to a class of workers who are engaged in interstate commerce.

Both the employee and Uber argued that the question could be resolved based on the existing record.  The employee argued that the court should look at the contract between the parties.  That contract implicitly contemplated a relationship with drivers from all fifty states and thus encompassed interstate travel.  Uber countered that the court should look only to its lived experience – Uber drivers inherently serve a local market, even if they occasionally might cross a state line here or there.

The court rejected both arguments.  The contract between the parties is one source of evidence about whether the workers belong to a class of workers engaged in interstate commerce.  But it’s not dispositive.  Similarly, the local nature of much of the work might be a factor, but it’s hardly the only factor.  Instead, the court instructed the lower court, on remand, to consider “various factors” including but not limited to “the contents of the parties’ agreement(s), information regarding the industry in which the class of workers is engaged, information regarding the work performed by those workers, and various texts—i.e., other laws, dictionaries, and documents—that discuss the parties and the work.”

And that brings us to the third big takeaway, which, in some respects, seems perhaps the most general and significant: the court’s instruction about what procedural framework governs a motion to compel, Fed. R. Civ. P. 12(b)(6) (motion to dismiss) or 56 (summary judgment).  The Third Circuit doubled down on an approach that it laid out in Guidotti v. Legal Helpers Debt Resolution, L.L.C., 716 F.3d 764 (3d Cir. 2013).  That approach uses a motion to dismiss standard for a motion to compel if the existence of a valid agreement to arbitrate between the parties is apparent from the face of the complaint or incorporated documents.  On the other hand, if the complaint and its supporting documents are unclear” as to whether the parties agreed to arbitrate, “or if the plaintiff has responded to a motion to compel arbitration with additional facts sufficient to place the agreement” in dispute, a “restricted inquiry into factual issues [is] necessary . . . .”

In this case, the court concluded that the complaint and supporting documents were unclear about whether the driver belonged to class of workers engaged in interstate commerce.  Accordingly, the court ordered the district court, on remand, to “permit discovery on the question before entertaining further briefing.”

On one hand, this third takeaway threatens, if read for all it’s worth, to authorize the same sort of “smell test” that SCOTUS unanimously rejected earlier this year in Henry Schein, Inc. v. Archer & White Sales, Inc.  Remember, there the Court put to rest the “wholly groundless” doctrine, which some circuits had used to do an end-run around a delegation clause.  Taken at face value, this Guidotti approach could do much the same thing by giving courts the opportunity to second guess the validity of an arbitration agreement.

On the other hand, in this particular case, the Third Circuit probably got things right.  Although the Uber agreement contains a delegation clause, as SCOTUS made clear in New Prime, such a delegation clause only kicks in once a court concludes that an arbitration agreement subject to the FAA exists.  In other words, a court must first determine if the contract falls within the  § 1 exemption.

All that said, the combination of the second and third takeaways from this case make things very messy, at least for a while, for the gig economy.  Until the dust settles, parties may wind up spending a lot of time litigating whether the FAA even applies.