In 2022, NFL coach Brian Flores filed a lawsuit in federal court against the league and several of its teams alleging race-based employment discrimination. But Flores’ employment contract contained a mandatory arbitration agreement, which required that he arbitrate, rather than litigate, any grievances against the NFL. This basic arrangement is familiar to tens of millions of workers nationwide. More than half of private-sector non-union employees labor under forced arbitration agreements. When an employee sues their employer, a judge must decide whether to enforce the agreement and push the claim into a private arbitration or allow the plaintiff to proceed in court. Bound by decades of precedent, judges almost always opt to enforce these agreements.

But not in Flores’ case. That’s because his contract had an unusual term: it selected as arbitrator the NFL’s commissioner. In other words, the contract said the league’s chief executive, Roger Goodell, would decide whether his league was liable for discrimination. The Second Circuit refused to enforce the agreement, finding it so “inherently biased” against Flores that it didn’t resemble “traditional arbitral practice.” The NFL asked the Supreme Court to weigh in, but the justices denied cert. Flores can stay in court.

The Second Circuit was surely right to see bias here: Goodell wasn’t capable of impartially deciding Flores’ case. But was the court right to think Flores’ arbitration agreement was so different from others? It’s true that most employment arbitration provisions don’t appoint the boss as arbitrator, and instead provide for selection of an “independent” adjudicator. Still, there’s a reason employers impose these agreements on employees anyway: forced arbitration systematically favors employers. Even when the boss doesn’t formally decide an employee’s case, employees cannot escape inherent bias in forced arbitration.

Forced employment arbitration favors employers.

Arbitration agreements are usually presented as take-it-or-leave-it conditions of employment in a contract drafted by the employer; most non-union employees have little ability to negotiate for something else. The employer’s contract sets the terms of the arbitration, including the mechanism by which the arbitrator will be selected.

Given the employer’s control of the forum, it’s perhaps no surprise that arbitration favors employers. Research shows employees win significantly less often and receive smaller awards in arbitration than in court; one study finds that employees are “nearly forty percent more likely to win and receive average awards nearly twice as large” when employment discrimination cases are litigated as compared with those that are arbitrated.

The evidence reveals a particular bias in favor of repeat-player employers who arbitrate regularly against one-shot employees. Arbitrators may have financial incentives to favor employers if they seek to solicit future business from repeat players. The New York Times spoke with dozens of arbitrators who said they “felt beholden to companies. Beneath every decision, the arbitrators said, was the threat of losing business.”

And it isn’t just that employees are losing at higher rates in arbitration—evidence suggests the vast majority of aggrieved employees subject to arbitration agreements don’t bring claims at all. One study compares filings of employment claims in court versus in arbitration; it finds that if employees subject to arbitration agreements “were as willing and able to arbitrate their claims” as other employees “are willing and able to litigate,” we’d expect that hundreds of thousands more claims would be filed in arbitration each year than are now. The study hypothesizes that low success rates and small recoveries discourage employees from filing claims and attorneys from taking employees’ cases.

Courts enable this inherently biased system.

Forced employment arbitration wasn’t always so widespread. In the early 1990s, just 2% of nonunion companies imposed mandatory arbitration agreements on their employees. But the Supreme Court has since transformed federal law to encourage these agreements, largely through interpretation of a 1925 law called the Federal Arbitration Act (“FAA”). The Court has said the FAA manifested “a liberal federal policy favoring arbitration agreements” and read the Act as a command to enforce valid contracts to arbitrate. But for much of the 20^th century, many considered employment contracts beyond the Act’s reach. As one scholar notes, employers and employees “assumed that public policy concerns would prevent courts from compelling employees to resolve employment discrimination claims through binding arbitration.”

The Supreme Court rejected that assumption in a pair of cases, Gilmer and Circuit City, in which it (1) upheld the validity of agreements to arbitrate statutory employment claims and (2) held the FAA applicable to most employment contracts. In the decades since, the Supreme Court has “sharply limited courts’ ability to police the fairness of arbitration agreements,” including by holding that the FAA preempts state law that might empower courts to scrutinize these agreements too closely. Lower courts may lean on narrow doctrines like the “effective-vindication rule” to void arbitration agreements in unusual cases (the Second Circuit applied this rule in Flores), but the default is enforcement.

None of the Court’s precedents support challenges to arbitration agreements that rest on claims of systemic pro-employer bias. In Styczynski v. MarketSource, a Pennsylvania plaintiff brought a discrimination suit in federal court against her former employer. In opposing the employer’s efforts to compel arbitration, the plaintiff lodged a “global objection to arbitration”: that arbitration is an “inherently unfair and biased forum” that favors employers. In a striking opinion, the judge endorsed the plaintiff’s policy concerns about the “adequacy and fairness of arbitration as the primary mechanism for enforcing federal laws.” But he explained that the Supreme Court’s case law forecloses invalidating arbitration agreements based on the “general proposition that arbitral panels are inherently biased.” Thus “bound by the rulings of higher courts,” the judge enforced the agreement.

Congress should act.

The Styczynski court wrote that “[o]nly [higher] courts, or Congress, can address the policy issues raised by motions such as this.” But this Supreme Court appears unlikely to revise its fundamental approach to employment arbitration anytime soon. Despite several recent decisions excluding narrow categories of transportation workers from the FAA’s reach, the Court’s pro-arbitration framework for assessing these agreements remains firmly intact.

Congress has shown more of an appetite for reform. In 2022, it enacted the “Ending Forced Arbitration of Sexual Assault and Sexual Harassment Act” (“EFAA”), which specified that survivors may pursue sexual assault or harassment claims in court even if bound by a pre-dispute arbitration agreement. A congressional report on the law finds that “forced arbitration deprives millions of Americans of their day in court to enforce state and federal rights.” And the House has twice passed versions of the Forced Arbitration Injustice Repeal (“FAIR”) Act, which would prohibit all forced arbitration in employment.

The Second Circuit was right to let Coach Flores press his case in court. But inherent pro-employer bias pervades the entire private arbitration system—not just proceedings where the boss plays arbitrator. Congress can and should act to ensure access to court is the rule for workers, not the exception.