Conservative justices embrace delegation to arbitration panels while questioning administrative process.
1984 was a dramatic year for literature and law. George Orwell’s classic novel came of age, with its four Ministries of Peace, Love, Plenty, and Truth that brilliantly described their opposites. Less observed that year—except by administrative lawyers—was the U.S. Supreme Court’s decision in Chevron v. Natural Resources Defense Council, the case that was to recalibrate judicial review of agency decisions. Also decided that same year was Southland Corp. v. Keating, a case preempting state arbitration laws that even administrative lawyers may have ignored.
It is tempting to tie all three events together—since Orwell’s ministries are, after all, agencies. But I will focus on the two decisions by the Supreme Court. Taken together, they tell an unappreciated tale of doctrinal contradiction—a kind of Orwellian “doublethink.”
The question for today is how did arbitration, an alternative decision regime about which the Court knows little, become so favored that a majority of justices have been willing to embrace it almost without question, while administration, an established regime about which the Court knows a lot, has been questioned to the point of being disfavored?
Chevron is the well-known tale, as the most cited administrative law case and the subject of endless discussion that has felled many trees. I intend to leave most of that aside and focus on Chevron’s overarching purpose: to give agencies judicially delegated power to decide executive matters of policy either initially or finally—the famous “two-step” process.
Chevron supported the administrative state by recognizing that agencies derive their power from the more democratic executive branch, subject to the procedural protections contained in the Administrative Procedure Act (APA). That consensus, accepted by the Court’s liberals and conservatives in the era of a conservative Reagan presidency, is increasingly breaking down. Conservative justices now see flaws in administration not highlighted since the early New Deal period. For example, in his dissent in City of Arlington v. Federal Communications Commission, Chief Justice John Roberts described agencies outside the “traditional executive departments” as the “headless fourth branch of government.”
Indeed, scholars such as Philip Hamburger now openly postulate that the entire administrative state is illegitimate, an argument not heard since Roscoe Pound made it in the 1930s. This reactionary turn recasts Chevron as an impediment to the Court’s jurisdiction over the administrative state.
Furthermore, Chevron’s influence is shrinking along with related cases like Auer v. Robbins, which asks judges to defer to agency rules as well as policy. At the same time, the Court is challenging the APA itself in new ways such as how administrative law judges (ALJs) are appointed and controlled.
In contrast with Chevron, Southland was a stealth administrative law case decided in the same term. And it continues to live a charmed life. Chief Justice Warren E. Burger’s majority opinion converted the Federal Arbitration Act (FAA) into a regulatory statute by using it to set national arbitration policy through the preemption of state law. Justice Sandra Day O’Connor’s fiery dissent, joined by Justice William Rehnquist, argued that the FAA was a procedural statute only and not meant to foreclose state legal regimes. “Although arbitration is a worthy alternative to litigation,” she wrote, “today’s exercise in judicial revisionism goes too far.” Commentators have called the majority opinion “extraordinarily disingenuous.”
Southland’s impact on the Court’s jurisdiction has been enormous. By preempting California law, it effectively allows mandatory arbitration to defeat state remedies such as employee and consumer-based class actions. But even though the resistance to Southland began among conservative justices, the Court’s current conservatives have by all appearances undergone a doctrinal conversion. Today, there is as much a Southland doctrine as a Chevron doctrine, although discussion about it is not felling many trees. Let us call this essay a falling sapling.
The Southland and Chevron doctrines have one thing in common: a desire to reduce the judicial workload by resolving cases in alternative venues. By endorsing arbitration as an alternative regime to litigation, Southland not only legitimated the arbitral process but made arbitration an alternative to administration as well.
This aspect of the FAA is less well understood. Imre Szalai, who has done deep research into the FAA’s legislative history, calls the statute an “evolutionary step in the rise of the administrative state where experts are delegated the authority to handle complex problems.” By equating arbitration and administration, Szalai’s insight connects us back to the New Deal period when procedural choices were hotly contested. But that was before the APA, which, according to Justice Robert H. Jackson in Wong Yang Sung v. McGrath, enacted “a formula upon which opposing social and political forces have come to rest.”
The mystery is this: Why has the Court seemingly opted procedurally for the FAA while undermining the APA? The FAA is a bare bones statute. Its key provision simply declares that arbitration agreements are “valid, irrevocable, and enforceable” generally. The FAA does not tell parties how to do arbitration—it just says do it. It is not concerned about the quality of the resulting arbitration nor with displacing the alternative regimes of state law that regulate arbitration agreements.
The FAA makes no attempt to mimic the careful procedural regime set up by the APA. This makes sense since arbitration, unlike administration under the APA, is intended to be consensual, and whatever process two equal parties agree to should suffice, whether it be a coin flip or a more formal structure created by private groups such as the American Arbitration Association. With arbitration, the parties get to choose their deciders, which is something the administrative process—or the judicial, for that matter—does not permit.
One can see why Congress in 1925 wanted the FAA to be accepted by a then-reluctant judiciary. This time period was before the Federal Rules of Civil Procedure, and the courts were overwhelmed with a burgeoning workload brought on by Prohibition cases. But the FAA was not a sophisticated statute. It never anticipated the problem that arises when the parties are not equal and arbitration is forced on the weaker one through mandatory arbitration clauses.
In consumer and employment situations, states such as California have tried to redress the balance through opt-outs and class actions under the rubric of unconscionability. But once the Supreme Court favored arbitration in Southland, lower courts have turned against these arguments. What remains is an alternative decision regime endorsed by the Supreme Court that procedurally comes nowhere near what the administrative process offers in terms of fairness.
Favoring arbitration because of its effect on judicial workload has long been appealing to overcrowded courts. Unlike with administrative actions, judicial dockets are permanently reduced once putative cases are sent to arbitrators or, as happens more now, simply disappear. But this self-interested desire on the part of federal courts to reduce dockets is no longer supported by the facts. As Judith Resnick has shown, the federal courts’ fears of overwhelming filings in the 1990s were exaggerated—the estimates were two-thirds higher than the actualities.
Today, there is no caseload crisis. Moreover, even valid workload arguments should not apply when states see the need for judicial action. Preemption under the FAA eliminates state court resources that stand ready to be deployed, based on state statutes that carefully question the consensual basis for arbitration actions. The Supreme Court’s decision to let the FAA “occupy the field” and preempt state law has drastic consequences for states that want to monitor the fairness and efficacy of arbitration agreements and the arbitrators who decide under them. The Court has effectively washed its hands of concerns about both procedural and substantive rights arising in this alternative decision regime.
At the same time, the Court continues to disrupt the carefully constructed procedural regime that the APA created in the years following the New Deal. In a series of cases highlighted by Lucia v. SEC, the Court’s concern with executive authority under Article II has jeopardized the APA’s careful structure for the independence of ALJs. After Lucia established ALJs as “inferior officers,” the President transferred jurisdiction to appoint them from the Office of Personnel Management to the agencies directly. This shift in appointment authority has efficiency advantages but may well have negative consequences for ALJ independence.
The judicial interest in micro-managing the administrative structure continues with Seila Law v. CFPB, a case challenging single-headed independent agencies, currently before the Court. The Court’s intense focus on administrative structure is totally at odds with the dismissive nature of its approach to the FAA’s traumatic effect on the arbitration field.
No case better illustrates the Court’s contrasting approaches to the arbitral and administrative regimes than Epic Systems Corp. v. Lewis. At stake was an employee’s right to bring collective action for lost overtime wages under the Fair Labor Standards Act. The amounts were so small that individual actions in arbitration—mandated by the employer’s agreement—would never have proven a practical remedy. Petitioners sought to use the National Labor Relations Act’s (NLRA) “concerted activities” clause to set aside the restrictions on class actions in the arbitration agreement. The Court, divided 5-4 with heated rhetoric from both sides, held that the FAA protects employers from the administrative regime set up by the NLRA.
Justice Neil Gorsuch’s majority opinion refused to allow the elaborate administrative system established under the labor laws to weaken the power of the FAA’s application. Even though the National Labor Relations Board (NLRB) had ruled that the NLRA presented no fundamental conflict with the FAA, Justice Gorsuch rejected the Board’s reasoning. He found no Chevron deference, since the Board had no special expertise with the FAA, and he dismissed the NLRB’s conclusion by noting that the NLRB and the Solicitor General filed competing briefs on the issue. Once free of deference to agency interpretation, the majority opinion granted the FAA equal status with the NLRA and used the FAA to defeat the long established administrative structure surrounding labor relations.
Justice Ruth Bader Ginsburg, in dissent, attacked the “egregiously wrong” majority opinion on several fronts. Her opinion, joined by Justices Stephen Breyer, Elena Kagan, and Sonia Sotomayor, discusses the 75-year history of the NLRA and the labor movement. It focuses on the movement of the Court from cases upholding “yellow dog” contracts that prevented employees from joining unions to its New Deal awakening and willingness to uphold legislation protecting employees’ rights. She accused the majority of returning to the Lochner era by denying plaintiffs a class action remedy under the NLRA.
Justice Gorsuch dismissed the Lochner charge by pointing out that statutory interpretation of the FAA is not constitutional construction under the Due Process Clause. He is correct theoretically, of course, since Congress could amend the FAA to fix the case. In the current political environment, however, any bill on this issue that emanates from the U.S. House of Representatives will die quickly in the U.S. Senate. The reality is that, in the current world of legislative gridlock, the difference between statutory and constitutional interpretation has lost much of its practical effect.
Epic Systems stands for the proposition that the interests of the FAA and arbitration clauses are stronger than the administrative structure undergirding federal labor law. By choosing no structure over an established one, the Court interpreted the FAA to deny substantive remedies either in state courts or administrative forums. As a result, no forum is available to hear the case. That outcome arises even though the NLRB has the power to consider class relief and thus can satisfy basic labor law rights even while leaving the preemption of state law standing.
Why would the Court prefer arbitration over access to administrative solutions? It is hard not to see the Four Horsemen of Conservatism riding again, joined by the Chief Justice as the Headless Horseman of the Fourth Branch to make a fateful five. Antipathy to administration has to be one answer as to why the “rule” of arbitration has bested the rule of administrative law—if not the rule of law itself.
Still, it is hard to see how this all evolved. Until Southland, preemption has never been a conservative cause. But once the step was taken in Southland, there seems to have been no turning back.
The unsettled status of Chevron demonstrates that basic assumptions about the allocation of administrative and judicial power are being rethought. Is the FAA really an accidental accomplice in the reassessment of the APA and the administrative state it helped legitimate? Or is it more a part of the Court’s long held bias in favor of big business interests and perhaps the larger project of rolling back the New Deal?
Adam Cohen’s recent book, Supreme Inequality, charts a detailed course for the Court in this direction that is deeply troubling, especially for those who believe in neutral principles rather than judicial politics. Is “Equal Justice Under Law” still such a neutral principle?
This essay expresses the author’s views only and does not necessarily reflect those of any organization.
