Symposium: For decades, court has built “an edifice of its own creation” in arbitration cases — it’s time to tear it down and rebuild
Deepak Gupta is the founding principal of Gupta Wessler PLLC. He was counsel of record on an amicus brief for the American Association for Justice supporting the employees in Epic Systems Corp. v. Lewis.
The Court has abandoned all pretense of ascertaining congressional intent with respect to the Federal Arbitration Act, building instead, case by case, an edifice of its own creation.
No, those words don’t come from Justice Ruth Bader Ginsburg’s dissent this week in Epic Systems v. Lewis — though they’d fit right in. They were written more than two decades ago, by Justice Sandra Day O’Connor. Since then, the edifice O’Connor described has grown exponentially, casting its shadow over all Americans’ ability to band together to enforce the laws designed to protect them.
In AT&T Mobility v. Concepcion and American Express v. Italian Colors, the Supreme Court gave companies a green light to use arbitration clauses to cut off collective claims by both consumers and small businesses, under both state and federal law — even under antitrust laws designed to police the very market power that enables big companies to insert these clauses in the first place.
Epic Systems sweepingly extends this dangerous trend, blocking workers from banding together to redress the full range of workplace legal violations — including wage theft, sexual harassment and race and gender discrimination. It’s hard to overstate the impact. The main effect is not to channel cases to some private forum but to kill cases entirely, cutting off not only compensation and deterrence but public accountability and the development of the law itself. As Lina Khan and I wrote in the Yale Law & Policy Review, this represents a massive upward transfer of wealth. One judge (quoted in the New York Times’ Pulitzer-nominated series) called it “among the most profound shifts in our legal history.”
The “pretense” of ascertaining congressional intent, as O’Connor put it, is one of two legal fictions upon which Epic Systems rests. In this post, I’ll briefly discuss these fictions and how policymakers can reorient the law to comport with reality.
The fiction of consent
The first fiction that jumps out at you when you read Epic Systems is the notion that the case is about nothing more than the enforcement of bilateral contracts to which the parties freely consented, as when Exxon and Chevron sit down to negotiate the process for resolving a commercial dispute. This unstated fiction permeates Justice Neil Gorsuch’s opinion, which is sprinkled with references to the parties’ freedom “to alter their arbitration procedures to suit their tastes” — as if a factory worker can pick his or her favorite arbitral procedure from the menu.
The fiction shows up in the opinion’s very first sentence, which frames the issue this way: “Should employers and employees be allowed to agree that any disputes between them will be resolved through one-on-one arbitration?”
But, of course, that isn’t at all what this case is about. It’s about whether employers should be allowed to force their employees, as a condition of continued employment, to give up any right to collectively use the legal system to combat wage theft, sexual harassment or other serious violations of law — even if this means the employees will lose any effective means of redress as a result.
This is no small matter. The legitimacy of the Supreme Court’s decision hinges critically on the notion that it is protecting the parties’ freedom to contract; any authority that a court has to send a dispute to arbitration and any authority that an arbitrator has to hear it derive from the parties’ consent.
But, if you’re like most American workers, when your employer emails you an “agreement” and requires you to “agree” if you want to keep your job (as the employers did here), you don’t experience this as freedom. In this scenario, Ginsburg points out, the employees are “faced with a Hobson’s choice: accept arbitration on their employer’s terms or give up their jobs.” Such “take-it-or-leave-it labor contracts,” she observes, harken back to the notorious “yellow dog” contracts used more than a century ago to prevent workers from organizing and asserting their rights.
Describe this issue to most nonlawyers and they’ll see this economic reality, even if they miss the historical parallels. But the Supreme Court is oddly blind to it — never once acknowledging the critical questions of consent and inequality of bargaining power that are at the heart of the case and that have long been central to both contract law and labor law.
The fiction of ascertaining congressional intent
The people who enacted the Federal Arbitration Act and the National Labor Relations Act — the two statutes at issue in Epic Systems — weren’t so blind. Both statutes were specifically written to protect workers from employers’ abuse of their superior bargaining power. Yet the Supreme Court manages to unleash precisely that abuse, and does so in the name of harmonizing the two statutes.
This is the second fiction — the idea that the Supreme Court is on a quest to ascertain congressional intent. Gorsuch’s opinion is filled with lofty language about “respect for Congress” and “respect for the separation of powers.” Judges, he says, should be “expounders of what the law is” rather than “policymakers choosing what the law should be.”
Despite this rhetoric, the opinion ignores the tools we typically use to discern congressional intent — text, history and purpose.
You wouldn’t know it from Gorsuch’s opinion, but the Federal Arbitration Act expressly excludes all employment contracts from its reach, providing that “nothing herein contained shall apply to contracts of employment of seamen, railroad employees, or any other class of workers engaged in foreign or interstate commerce.” But the Supreme Court doesn’t mention this language — presumably because it held, in Circuit City Stores v. Adams, that the statute means the opposite of what it says.
When Congress considered the FAA in 1925, some legislators expressed concern that arbitration might let “the powerful people … come in and take away the rights of the weaker ones.” But the FAA’s architects assured them this wasn’t the case: “It is not intended this shall be an act referring to labor disputes, at all. It is purely an act to give the merchants the right or the privilege of sitting down and agreeing with each other as to what their damages are, if they want to do it. Now, that is all there is in this.”
How did we even get here? For much of the 20th century, arbitration under the FAA worked as Congress had intended: to resolve the garden-variety contractual disputes that arise between businesses. Federal statutory claims were categorically beyond the FAA’s reach, as were all claims brought by workers and all claims in state court. The insertion of arbitration clauses into mass contracts with consumers or workers was unheard of.
It wasn’t until the 1980s and ’90s that the Supreme Court even allowed federal statutory claims into arbitration. And, when it did so, it was always careful to insist on a critical “effective vindication” principle: Arbitration was permissible only so long as it didn’t interfere with the parties’ ability to vindicate their substantive rights effectively.
Remarkably, that limiting principle also makes no appearance in the Supreme Court’s opinion in Epic Systems. That principle — which was supposed to preserve the legitimacy of arbitrating statutory claims in the first place — now appears to have vanished entirely, without a trace.
Moving from fiction back to fact
In fairness to Gorsuch and his colleagues, Epic Systems wasn’t written on a blank slate. It is yet another chapter in a chain novel the Supreme Court has been improvising for decades — now resembling something more akin to common-law constitutional adjudication than statutory interpretation.
But, despite its apparent ability to trump federal statutes, the FAA isn’t a constitutional provision. It’s a statute. And so it’s within Congress’ power to limit arbitration where parties have vastly unequal power and where statutory rights are at stake.
Ginsburg called for just such congressional intervention when she read her dissent from the bench. (In addition, as Daniel Hemel recently observed, there are some limited tools that can be employed at the state and local level, like disclosure requirements and the use of market power to police government contractors.)
A congressional fix needn’t happen all at once. Congress can limit arbitration in particular areas or delegate its power to do so. As we explained in our brief for the American Association for Justice, over a dozen federal agencies have taken steps in this direction, under administrations of both parties, to protect the rights of military servicemembers, farmers, investors, airline passengers, nursing-home patients and others. The trend picked up steam in the final years of the Obama administration but has been reversed by the Trump administration.
The high-water mark was a 2017 rule by the Consumer Financial Protection Bureau, following the most comprehensive study on arbitration’s effects to date. Unfortunately, the rule was repealed last October under the Congressional Review Act. But the Senate vote was close — a bare majority was needed, and it passed 51-50. The bureau retains the statutory authority to revisit this issue and Congress retains the power to delegate parallel authority to other agencies in the future.