Do I believe in arbitration? I do. But not in arbitration between the lion and the lamb, in which the lamb is in the morning found inside of the lion.
Samuel Gompers1
The most famous judge in America isn’t a judge: she’s an arbitrator. Judy Sheindlin—aka Judge Judy—presides over America’s top syndicated TV program.2 Judge Judy promises “real people” and “real cases” and delivers just that, because to appear on the show, contestant-litigants must agree to let Sheindlin conduct binding arbitration of their disputes. Sheindlin’s syndicated parade of dysfunction has aired since 1996, meaning that Judge Judy has been in her seat longer than Justices Kavanaugh, Gorsuch, Kagan, Sotomayor, Alito, and Roberts have been in theirs (and nearly as long as Breyer and Ginsburg). Besides extraordinary tenure, Sheindlin has resolved far more cases (at least 6,000) than the Court has over the same period and attracts a considerably larger audience than even the most popular Justice (the Notorious RBG).3 In shaping America’s impression of civil litigation, Judge Judy’s only real peer is the pioneering Judge Wapner of The People’s Court, the John Marshall of the boob tube.
It would be fun to give Judge Judy the full Barthes/Derrida treatment, “interrogating” mythologies, analyzing signifiers and signified, and deconstructing all around. There’s the psychodrama of Sheindlin’s ex-husband succeeding Wapner on The People’s Court. There’s also Sheindlin’s own legal history, including various lawsuits ranging from grimly prosaic allegations of bias to oddballs such as a $500,000 suit over Sheindlin’s alleged theft of tableware. And then, of course, there’s the meta-issue of the show itself requiring that any disputes about its workings to be resolved by arbitration (understandable given the cutlery suit). Judge Judy provides material enough for dissertation-level fireworks about the show-as-portent-of-legal/cultural-collapse, sure to guarantee tenure (at, let’s be honest, Brown).4 But if Judge Judy would entrance social studies departments, the legal elite prefers to overlook the gross spectacle entirely.
Nevertheless, Judge Judy demands something between overinterpretation and willful blindness, because the show offers a fundamentally truthful (if exaggerated) picture of the arbitration system, the giant parallel world of private justice. Arbitration is nothing more than a contractual arrangement to resolve cases by means faster, less formal, and less rigorous than court justice. Judge Judy does exactly that. The show’s seemingly made-for-TV oddities very much feature in “real arbitration” (which, to repeat, Judge Judy legally is): low stakes, a glaring lack of procedural standards, disregard for written opinions, disconnection from precedent, and a narrow-minded focus on the case at hand without reference to notions of public policy or legal development. Of course, the show differs from mainstream arbitration in important ways, though surprisingly, usually for the better. Judge Judy resolves its cases swiftly, publicly, and mostly at its own expense; the contests are roughly equal; the parties knowingly consent to arbitration; and its mistress is a former judge of a real court. “Proper” arbitration cannot always say the same.
Judge Judy’s greatest cultural significance is in embodying the migration of jurisdiction from public courts to private arenas. The shift has accelerated considerably since the 1980s, making actual mass arbitration synchronous with arbitration-television. Though the transition to privatized justice is well known to the professoriate, it commands little public notice. This is notwithstanding the fact that most Americans are party to an arbitration clause, often dozens of them, embedded in form contracts for everything from telephones to food delivery. What these clauses require, if invoked, is that the parties forgo normal civil court proceedings in favor of private resolution before hired “judges.” We are all potential contestants on the real-world equivalent of Judge Judy.
And rest assured, virtually everyone has signed arbitration agreements. If you shop on Amazon (as most households do), you are party to an arbitration agreement.5 If you have a bank account, a credit card, maintain student loans, are employed by a large corporation, or use medical services, you are likely party to further arbitration agreements.6 Essentially all cell-phone users are covered by arbitration clauses.7 Not all financial firms use these clauses, but those that do tend to be large, so the clauses affect a disproportionate amount of economic activity. In a Consumer Financial Protection Bureau (CFPB) sample, arbitration clauses covered 44.4 percent of checking deposits and 53.0 percent of outstanding credit card loans, and 85.7 percent of the student loan contracts studied required arbitration.8 Some 56.2 percent of nonunion private sector employees are subject to clauses, and larger employers were particularly likely (65.1 percent) to use these clauses.9 Doctors, hospitals, and nursing homes have also discovered the pleasures of arbitration agreements. Not many consumers know that they are party to arbitration clauses, and fewer still truly understand what rights have been waived—perhaps less than 10 percent.10
These agreements, usually embedded deep in form contracts or online “conditions of use” essentially foreclose recourse to courts should you wish to sue a wayward counterparty. That corporations very much want you to relinquish court access can be inferred from the fact that General Mills seemingly tried to construe the mere act of opening a Cheerios box as “consent” to arbitration.11 Though likely tenable as a matter of law (at least when coupled with a coupon), the Cheerios gambit collapsed due to freakishly bad publicity. But the Cheerios debacle notwithstanding, corporations continue pushing arbitration onto their customers and employees, upending the system of civil litigation.
This unseemly rush into arbitration, where every “I Accept” click or unscrewing of a toothpaste tube vaporizes access to public justice, raises some obvious questions. Why are corporations so eager to arbitrate—is arbitration that good, or are courts that bad? Why would courts acquiesce to diminishment of their social roles? What might be lost when justice migrates from public to private fora? Judges (driven by the conservative wing of the Supreme Court) and corporations have their answers ready, all premised on the speed, efficiency, and choice offered by arbitration, which they portray as an unalloyed good. The academic consensus runs the other way: the displacement of real courts represents a catastrophe, especially for employees and consumers.12 The academics generally have the better of the argument. Mass arbitration degrades the law. It also tilts outcomes, because if consumers and employees aren’t guaranteed to lose in arbitration (though some groups make that claim), they certainly find it very difficult to really win.
Arbitration arose to serve specific purposes and remains a fair and useful mechanism for the kinds of disputes that occasioned its invention. To fully appreciate where modern mass arbitration went wrong, it’s helpful to know why anyone ever wanted arbitration in the first place.
Before there were judges, there were arbitrators, people who helped members of their communities resolve disputes.13 In many parts of ancient Europe, arbitrators allowed disputants to be heard and have their injuries addressed, preserving community harmony. Their methods were necessarily crude, and sometimes coercive, but so long as the community was preserved, the outcome was deemed good enough. Even as modern procedural machinery developed—the whole complex of lawyers, judges, courthouses, forms, and rules—the communitarian impulse behind arbitration remained. In many traditions, Islamic legalism being a prime example, arbitration remains esteemed for its ability to facilitate the sort of conciliation and comity that litigation’s war-by-paperwork makes difficult.14
In other traditions, including Anglo-American law, which emphasized individualism, personal vindication, and procedural objectivity, communitarian justifications for arbitration were always weaker. Western courts arose partly because informal arbitration could not always serve these values, especially as disputes extended beyond a single community. As courts institutionalized, they naturally viewed arbitration as a relic: ad hoc, semiliterate barbarity rendered moot by progress. Judges also disliked arbitration because it threatened their monopoly on justice, and the political and social establishments that sponsored judges tended to agree. It helped that the late seventeenth century, when European courts really took off, was technophilic and understood courts as the latest word in doing justice: Why walk when you can ride; why arbitrate when you can litigate?
However, courts could not solve every case, and arbitration flourished in special circumstances. Merchant shippers, for example, had limited time ashore to deal with courts; their disputes involved arcane matters (as anyone with the misfortune to know a weekend sailor is aware); and their community was sufficiently self-policing that it could resolve cases without resort to the state. These factors, coupled with the transnational aspects of many maritime disputes, rendered courts unusually ill-suited—and sometimes jurisdictionally incapable—of providing help. Even the crustiest English lawyers appreciated that arbitration was a “civilized” and “normal” process in maritime use, and not the second-class “abhorren[ce]” that arbitration was in other contexts.15
Maritime arbitration persists, as does labor arbitration, which thrived after the Civil War. Again, circumstances all but demanded arbitration. Postbellum courts recoiled at forcing people to work, and did not savor the proto-socialist prospect of forcing capitalists to employ anyone. Most importantly, capital and labor often both wanted arbitration, and had the means to establish a fair system. The essential point is that early arbitration succeeded because it was specialized, and provided not just a cheaper product, but a better one.
By the early twentieth century, a final set of special circumstances arrived: the disarray of federal courts in an era where federal (i.e., interstate) commerce had become crucial. The federal courts sank into chaos, crippled by archaic rules; indeed, district courts lacked a unified code of civil procedure until 1938. For a time, courts felt second-class. Larger merchants and their lawyers pushed Congress to relax the judicial monopoly so that companies could avoid court hassles and be assured that arbitration clauses would be respected in the many states in which business was conducted.16 In 1925, Congress passed a law (now known as the Federal Arbitration Act, or FAA) to help.17 Just four pages long, no one expected that the FAA would reconfigure the entire legal system. Rather, Congress intended to make arbitration provisions binding in the sorts of classic, multi-jurisdictional commercial disputes long subject to arbitration, with the FAA’s express reference to “maritime” disputes highlighting continuity with established practice.18
The FAA was a significant development, but one firmly rooted in conventional arbitral principles. Nothing in it seemed to augur a major break. The Act had been solicited by commercial interests that wanted a guarantee that they could reliably arbitrate among themselves; they did not ask for a mechanism to force arbitration on those outside their communities. Nor did the Act seem to propose anything so radical as modify the laws of contract generally, which were set by state courts and legislatures.19 Indeed, it’s doubtful the law’s original drafters foresaw that the FAA would—or even could—reach into purely intrastate matters. The general view at the time was that if Peoria slaughterhouses sold beef to a Chicago supermarket, any resulting dispute lay beyond the FAA. It was years after the FAA was passed that the Court developed the idea that essentially all commerce was “interstate commerce,” allowing federal policy to override state law when Congress desired.20 Moreover, the FAA hardly seemed to express any desire to intrude on the substance of commercial litigation.* 21 The courthouse doors remained open, and for some time, it was doubtful that any arbitration clause not wanted by both parties would shut those doors.
Indeed, courts could never be fully excised from deciding civil suits, a fact the FAA confirmed. The Act explicitly granted courts supervisory roles over arbitration and confirmed courts’ longstanding powers to overturn biased, unhinged, or otherwise defective arbitral awards. Crucially, the FAA explicitly declined to modify existing contract law, leaving courts free to void arbitration clauses on grounds like unconscionability.22 Those who read the FAA’s text, and not subsequent case law, would not expect that strong parties could use the Act to force weaker parties (consumers, employees, etc.) into unfair arbitrations. Yet, that is precisely what has happened over the past thirty-five years. The blame can be leveled squarely at the Court, which rewrote the FAA in a series of momentous and ill-considered decisions.23 These cases had little basis in the FAA’s text, betrayed basic principles of statutory interpretation, and made daring assumptions about arbitration that proved largely untrue.
For decades, the Court generally left the FAA alone, before reversing course in 1984’s Southland. In that case, the Court—without evidence—concluded that Congress declared a “national policy favoring arbitration” (dubious) and that the FAA “withdrew the power of the states to require a judicial forum for the resolution of claims which the contracting parties agreed to resolve by arbitration” (wrong).24 Intriguingly, the majority opinion was politically mixed, and the strongest dissents came from conservatives O’Connor and Rehnquist, who criticized the majority’s legal logic and worried that Southland would concentrate federal power at the expense of state sovereignty.
After Southland, the Court continued souping up the FAA until the power to compel civil arbitration was essentially absolute. Over the next thirty years, the Court overrode state laws, forced essentially noncontractual claims (like antidiscrimination suits based on statutes) into arbitration, made class-action arbitrations impracticable, and undermined every traditional defense to arbitration contracts. These developments may seem recondite, but they had enormous practical consequences for consumers and employees, who now find themselves stuck in arbitrations they don’t want and to which they did not knowingly agree. The consequences for intra-corporate contests were more limited, as modern arbitration was designed for corporate disputes. In high-stakes matters, corporations also frequently reserve the right to litigate in normal courts or provide for arbitration procedures that, save for being conducted privately, substantially replicate conventional court practice.25 As a result, the Court’s arbitration jurisprudence meant that corporations could force weak opponents onto Judge Judy, while bigger players could treat themselves to Cadillac justice.
Because the Court’s arbitration cases make little legal sense, they can only be justified by the policies they claimed to serve: freedom of contract, fairness, and efficiency. Forcing parties into arbitration, the Court mused, simply reflected the prior bargain made by the parties, would not be an unfair substitute for court proceedings, and would speed resolutions. None of the Court’s policy arguments have withstood contact with reality, with grievous consequences for consumers and the judicial system the Court nominally leads.
Starting in reverse order, arbitration is not efficient for many parties. A process is only efficient if it generates a lot of product relatively cheaply. The correct “product” should be justice, which many arbitrations can’t produce, but then again, in many contexts, arbitration clauses don’t even produce many arbitrations. For example, between 2009 and 2014, AT&T’s wireless base ran between 85 and 120 million users.26 Yet, over that same period, AT&T’s wireless customers, bound by arbitration clauses, filed a mere 134 arbitrations against the company.27 The same sort of calculations hold for credit card companies, banks, and the like: very few arbitrations.28 “Efficiency” seems to be a term defined by the stronger party, translating to “quashes almost all legal action at almost no cost.” If that’s the efficiency the Court had in mind, it could just as easily void a criminal defendant’s right to jury trial. (Actually, the Court’s done just that, as we’ll see.)
The counterargument is that consumers bring near-zero arbitrations against telecoms, card companies, and banks because they have no disputes with these companies. This defies common experience and consumer satisfaction surveys.29 It also can’t be reconciled with prosecutions brought by government. In 2017, government settled two suits against telecoms for “bill cramming” and unauthorized charges, one for $90 million against T-Mobile and another for $88 million against AT&T (which, recall, had endured just 134 piddling arbitrations filed against it in a five-year period). Rather gallingly, government-as-consumer sued Sprint, recovering $15.5 million because Sprint had overcharged for court-ordered wiretaps; not only had telecoms fleeced private customers, they had overcharged taxpayers to (legally) spy on them.30 Government has also regularly sued financial firms, extracting gigantic settlements from institutions like Wells Fargo and Countrywide. One reason government brings suits against abusive businesses while consumers do not is that the government isn’t bound by arbitration clauses and has the resources to prosecute mass wrongdoing in a way lone consumers cannot.
The Court’s response reduces to let-them-eat-arbitration, even when arbitration can offer no practical relief. This is especially a problem in matters in which each individual wrong is small, but the collective harm large—i.e., class actions. While it’s true that there’s no legal barrier to commencing a solo arbitration against a cell phone company for $378 of fake charges, there’s also no way to make a $378 arbitration economically viable. First, consumers would have to frame a complaint based on their cell contracts, and reading those contracts requires a lawyer’s assistance at $500 an hour, as cell contracts are functionally unreadable by most of the population.* 31 Cell contracts are, one might suspect, even designed to deter comprehension and there is only a single documented case of a consumer reading one all the way through and understanding it. That person was Judith Resnik, a Yale Law professor, who loathes consumer arbitration in a way that makes Nabokov’s hatred of Freud seem tepid. In a lather, Resnik once attempted to prove her point by not only reading a cellular contract, but attempting to negotiate one afresh. (Vendors declined.) But even if consumers read a contract, and could point to a potential breach of some provision within, that’s not enough: consumers must then prove instances of overcharging, rummaging up bills and data stored within corporate archives, and perhaps some evidence that the cell carrier had a practice of overcharging. Doing so would take a skyscraper of Resniks—in other words, a Manhattan law firm. Consumers therefore face a non-choice: pay $500 per hour to litigate a $378 claim or find a lawyer who assigns such a low value to his time that the customary one-third contingency fee of $125.99 seems worth it.
This is a problem the law has seen before and solved. When individual stakes are low, but many people have suffered the same injury, law allows for the class-action suit. These suits aggregate plaintiffs and damages, making it economically viable for lawyers to participate. The Federal Rules of Civil Procedure expressly permit class actions, which is partly how tobacco and asbestos companies were brought to heel.32 In theory, arbitration can accommodate class actions, though industry had a well-publicized disaster when it conducted a class arbitration: after the Pacific Gas & Electric Company contaminated Hinkley, California’s groundwater, residents sued and the cases went to arbitration. After the first batch of plaintiffs won $120 million, PG&E settled the first-wave cases for $333 million, and Steven Soderbergh made the story into Erin Brockovich.33 Tobacco, asbestos, and groundwater suits involved claims external to contracts, but industry learned its lessons, and reinforced its arbitral armor: contracts banned class-action arbitrations, effectively eliminating court cases from large swathes of economic life.34 Some observers thought this was overreaching, but in 2011, the Court upheld class-action waivers in arbitration contracts in a case against AT&T.35 This must have been an easy case, for at least Chief Justice Roberts. Roberts had, after all, already briefed the class-waiver issue in his prior career as a private attorney, asking the U.S. Supreme Court to reconsider an inconvenient anti-arbitration ruling by a California court.36 The Court declined. After Roberts joined the Court, it heartily embraced a more expansive view of arbitration. All Chief Justice Roberts had to do was make a little law, precisely the sort of thing that he testified judges don’t do.
By shutting down class-action arbitrations, the Court functionally voided rights embodied in state and federal law, and in the Federal Rules of Civil Procedure. The net effect is that when disputes involve small amounts, corporations may safely extract extra dollars from consumers, facing judgment only in the comparatively infrequent cases in which they are sued directly by government watchdogs or convicted in the court of public opinion by media exposés. It’s perhaps worse than that, as the Court suggested in 2012 that not only contract claims, but potentially tort claims, can be shoved into arbitration, which might derail future Erin Brockoviches. The pertinent case, involving allegedly fatal neglect in nursing homes, produced a brief opinion limited to pooh-poohing a West Virginia statute that forbade compelled arbitration of personal injury and wrongful death claims.37 A regime that guts class actions and consigns abused (or dead) seniors to lopsided arbitrations may not, it’s reasonable to say, be producing social goods with optimum efficiency.* But those sorts of goods and that sort of efficiency were not what the Court focused on when it began its torrid affair with arbitration.
What did worry the Court in the 1990s, as it busily expanded the arbitral frontier, was judicial efficiency. Chief Justice Rehnquist fretted that federal courts would soon be overrun by litigation.38 Rehnquist accordingly asked Congress to build new courthouses and appoint new judges. His core tactic, however, was not to expand capacity, but to reduce throughput: he wanted some new judges, but many fewer cases. In 1991, the Chief urged Congress to reconsider its rapid federalization of criminal law, radically curtail inmates’ ability to file federal habeas petitions, and greatly limit “diversity jurisdiction” cases (interstate disputes not premised on substantive federal law).39 Some of these ideas were worthy, others callous, and all ideological hobbyhorses for Rehnquist, who preferred legal issues to be devolved to the states and stay there. They also reflected a serious, if ultimately misplaced, concern about rising judicial workloads.
Congress granted Rehnquist his new courthouses and largely ignored his other pleas, so Rehnquist engaged in self-help. He could do fairly little to unilaterally reduce the judiciary’s criminal docket; the Constitution made that difficult. The civil docket was another matter, were the Court willing to set aside some legal principles in the name of the greater (or anyway, its own) good. The Court therefore made it much easier for parties to force each other into arbitration, and thus away from courts. It did so without any legal basis, as Rehnquist (circa 1981 to 1984) had himself pointed out.
As a practical, rather than ideological, matter, the gambit proved unnecessary: what Rehnquist didn’t appreciate in 1991 was that federal court filings were in a bubble. The S&L crisis and criminal drug epidemic, which had driven so much federal litigation, would shortly peak. Litigation would soon return to growing roughly in line with population and GDP. The tangible result of Rehnquist’s reforms was a series of overbuilt and frequently desolate new courthouses, wastes of taxpayer money made uglier by the architecture, done in a style charitably described as exurban Hyatt. The intangible result was a doctrinal shift toward arbitration. That, it turns out, was uglier still.
What made the intangible result so ugly was arbitration itself, because there were never any guarantees that nonjudicial fora would be just or effective. At its best, arbitration complements courts. But arbitration never substitutes for courts—because it can’t. Critically, arbitrators cannot enforce anything; only state actors have coercive power. Should a party withhold documents or refuse to obey arbitral decisions, parties must return to court.*
The lack of state power is the only truly universal feature of arbitration. Beyond that, arbitrations can be as close to, or far from, normal court procedure as an arbitration clause requires. Arbitration clauses can exclude attorneys, curtail discovery, limit motion practice, require or forbid written reasons for judgment, constrain appeal, shift fees and costs, require proceedings to occur at the South Pole; whatever the (stronger) parties prefer. Only when the procedures selected are embarrassingly cursory or abusive are courts likely to intervene, and then only reluctantly. In one case, arbitrators failed to hold a preliminary hearing, set a discovery schedule, nor required parties to produce documentary evidence prior to the main event. This triggered a complaint to a federal district court, which refused to intervene, stating that these procedural defects did not constitute a viable claim.40
All American law suffers from underdeveloped structural safeguards, but the problem is especially acute in arbitration. Though, as we’ve seen, there are no mandatory requirements for being a federal judge, the confirmation process and attending publicity make for a pretty talented bench. And once on the bench, judges operate within standardized rules of procedure, conduct open hearings, and face reversal on appeal. No such checks apply to arbitrators, so the need to screen for talent and temperament is arguably far greater than it is for judges. Yet, as the Bureau of Labor Statistics notes, “[t]here is no national license for arbitrators,” “[f]ew candidates receive a degree specific to the field of arbitration,” and “a bachelor’s degree is often sufficient.”41 Many states do not require certification or training for mediation/arbitration work; those that do often content themselves with training programs running twenty to sixty hours. Even some of the better voluntary certification courses, like those offered by USC (“one of the top 10 ADR programs in the world” per an industry rag), do not require candidates to have a law degree and advertise the brevity of their coursework—completable in “as few as two semesters,” as one of USC’s programs enthuses.42
Arbitration is almost entirely self-regulating, a characteristic denied to utilities, emergency rooms, and most other providers of social necessities. Arbitration firms, which range from tiny to giant, have equally variable standards. Standards at smaller firms can be extremely loose. Larger, more prestigious firms do pride themselves on having talented arbitrators (including former state court trial judges), but even the largest, most reputable firms, like the American Arbitration Association (AAA), allow non-lawyers with “senior-level” experience and “appropriate” degrees to become arbitrators.43 As in most markets, some very good firms exist, including JAMS (f/k/a Judicial Arbitration and Mediation Services), a corporation founded by a judge, whose standing rules and procedures tend to follow normal court practices and which declines to preside over the sleazier sorts of arbitration, such as those that require adjudication in inconvenient places, discourage counsel, or eschew written decisions.44 Notably, when corporations arbitrate among themselves, they prefer JAMS and its peers.
The singular advantage arbitrators have over judges is that many have significant subject matter knowledge; indeed, this was an original justification for maritime arbitration. For example, construction arbitrators know more about building codes, cement drying times, and architectural practice than the average judge. The degree of specialization can be astonishing: nested in AAA’s seven major specialty areas exist many subspecialties, including practices as hyper-specific as Olympic athletes and sports doping (in fact, the U.S. Anti-Doping Agency has designated AAA as the primary venue for Olympic doping disputes).45 For intercompany litigation, this can produce more knowledgeable decisions, especially when an arbitration panel includes experts in both subject matter and legal procedure.
However, the recycling of industry insiders into arbitrators raises the possibility of bias. Arbitrators who come from industry may unconsciously favor their old employers; a doctor-arbitrator may be far more skeptical about malpractice claims than, say, a tax lawyer. The evidence for this proposition is mixed, but the consequences may be the same regardless, because the fundamental problem is perceptual: a justice system perceived to be rigged suffers many of the same legitimacy deficits as one that is actually rigged. Moreover, it’s easy to show that arbitrators always have at least some structural conflict of interest that judges do not, because arbitrators only get paid if a dispute is arbitrable. Somewhat oddly, “arbitrability” is an evaluation courts often allow arbitrators to make.46 Courts view this bias as minor, since it’s a bias toward arbitration as an institution, not “classic bias” like overtly favoring a particular side. That judicial logic is not persuasive, however, if arbitration systematically disfavors one class of litigants (and we have and will see that it does). If so, structural bias becomes the kind of bias courts normally disallow.
Because arbitration is confidential, rates of classic bias—corruption, favoritism, prejudice, and so on—are not fully knowable. We do have glimpses, however, when arbitrations enter the public record via subsequent public litigation, as when courts hear petitions to vacate arbitral judgments when arbitrators exhibit “evident partiality.” As the ABA notes, these cases are “difficult to prove and rarely succeed.”47 They are hard to prove because there’s no freestanding rule that arbitrators keep a complete record of the proceedings—or indeed, any record. As for success, the procedural barriers to overturning arbitration are so high that few challenges are brought and, seemingly, even fewer succeed.48 But arbitrators do have conflicting interests, even if we don’t know to what extent, and can’t do much about it.
In real courts, litigants may move to disqualify a judge at almost any time.49 In arbitration, litigants have only two opportunities to challenge bias, and the mechanisms are nonoptimal. The first chance is at arbitration’s beginning. This is a weak protection, as there are no hard rules on what must be disclosed, conflicts are self-diagnosed, and “trivial” conflicts of interest don’t necessarily disqualify.50 Worse, this regime sterilizes bias challenges, since merely disclosing bias (even actual conflicts of interest) can foreclose remedial action, even if parties didn’t read or understand the disclosures. After disclosure, bias cannot be challenged until arbitration concludes—even if, during the course of arbitration, it seems likely that a court will find bias.51
Clearly, in cases that must be relitigated due to bias, arbitration is less efficient than court proceedings—but excepting these outliers, is arbitration faster and cheaper than court? Generally, yes, for the same reasons that Greyhound buses are faster and cheaper than the QE II: fewer amenities. In commercial arbitrations involving larger sums, the median time between filing a demand and the rendering of award runs about 11.6 months, versus 24.2 months just to get to trial in district court.52 Arbitral pace is even speedier compared to the busier district courts (e.g., New York’s SDNY) and always-congested state superior courts.53 Boosters of arbitration ice their cake by noting that arbitrations slice a further two years off the appeals process, though appeals are rare because there are so few grounds to appeal arbitration, so the pro-arbitration argument here is akin to a dietician commending amputation as an efficient means of weight loss.54 As for legal savings, the same logic holds: less law practiced means lower legal bills overall. The dollar savings, while hard to quantify, are likely nontrivial. Arbitration can indeed save time and money, at the cost of process and thoroughness. For knowledgeable parties, especially those facing the prospect of multiyear litigation, this exchange may be attractive. If legal savings are large enough, arbitration can be rational, even if a party believes it has a much better chance in court. These are not arbitrations that should provoke much worry.
Arbitrations that do raise concerns include those that involve the numerically largest group of potential litigants: consumers and employees. For the many small-dollar cases that arbitration clauses make uneconomic, there is simply no litigation. Gains in speed and cost are ∞ percent (or mathematically undefined, to be precise), though this doesn’t represent a true improvement. Were bloody-minded holders of small claims to arbitrate anyway, the process is genuinely fast. But small claims courts are also fast. These courts, which handle low-dollar cases (usually $10,000 and under, depending on state), dispense with baroque procedures, and by statute must conclude preliminaries (if any) quickly, before conducting a brisk trial (often thirty minutes, about as long as a Judge Judy episode); most cases are resolved in three or fewer months.55 Even corporations, generally averse to full-scale litigation, seem to prefer small-claims courts, and it’s standard for lenders to exclude small claims from arbitration waivers, as courts can more expeditiously enforce debt collection than arbitrators.
Arbitration’s friends extend their argument by citing the economic gains—invariably gigantic—that arbitration offers. That contention withers under scrutiny. For the smaller claims that comprise most potential arbitrations, small claims courts already provide expedient fora, and there’s not much fat to trim. The customary fallback, especially in consumer cases, is that arbitration clauses permit standardized contracts and therefore reduce legal fees, with the savings shared by both parties. But that might not be true. When a court settlement forced some credit card companies to temporarily abandon arbitration clauses, life went on without consigning card issuers to oblivion or consumers to noticeable extra costs.
Where consumer arbitration clauses can achieve genuine efficiencies is in mooting frivolous class actions, proceedings that increase costs/lower profits for everyone but the class lawyers. However, corporate defendants should not be “repairing” class-action law by preprinted form, as they lack appropriate incentives or a Constitutional mandate. The proper locus for reform is Congress, which has reformed certain types of class actions without judicial–corporate freelancing.56
Whether or not arbitration is procedurally fair or economically efficient (not that some Justices really care), the Court argues that arbitration preserves freedom of contract, a cardinal virtue in market economies. This is something of a red herring, as the Court has allowed interference with contractual freedom when it believes that doing so serves public policy. Whatever else decisions upholding bans on prostitution, gambling, and drug dealing are, they are not laissez-faire. Intellectual inconsistency aside, consumers are not freely contracting with many of the companies that require arbitration. For a start, form contracts are often not contracts at all, in the traditional legal sense of being agreements to which both sides freely consent. Many contracts provide that their terms and conditions may be changed at any time, so that the cell contract signed in the store may not be the operative contract when the wrong occurs and it’s time to litigate.
Clearly, no “freedom of contract” exists when one party can shift contractual terms on the fly; equally, no freedom exists when consumers have no practical choice but to sign a contract. For example, telecom is a necessity as much as water and food. The Justices, as we saw in Chapter 6, may not appreciate the centrality of cell phones and Internet to modern life, but these goods are essential all the same. Consumers have few choices when it comes to telecom providers (partially due to limp antitrust enforcement by other parts of the legal system). Because essentially all major telecoms require arbitration, consumers must give up court access to secure a basic necessity—and if telecom doesn’t qualify as a necessity, consider that the increasingly oligopolistic health-care industry also uses arbitration when it can.57 It’s no surprise that concentrated industries like telecom and financial services simultaneously have terrible consumer reputations and great enthusiasm for arbitration.
A slightly weaker version of this holds for employment by major corporations, which have hopped on the arbitration bandwagon. If Americans can choose among employers, few can choose to be unemployed. Increasingly, workers have no choice but to sign contracts that include arbitration clauses designed to reduce employers’ potential liability for wrongful conduct. This is not pure corporate villainy: by being so permissive about these liability-minimizing clauses, the Court may have put CEOs in a position where the profit-maximizing principles of corporate law encourage use of arbitration clauses. In 1992, just 2 percent of nonunion private sector employees were subject to mandatory arbitration, but as the Court expanded arbitration possibilities, more employees were saddled with arbitration clauses—56.2 percent by 2017.58 The figure will certainly rise, especially after May 2018. That month, Justice Gorsuch dealt employees another blow, construing the FAA to trump the National Labor Relations Act (designed to protect workers, not that Gorsuch cares), allowing employers to squelch collective litigation.59
Judicial enforcement of one-sided contracts that foreclose just outcomes—in this case, arbitration clauses—offends some of the oldest principles in Western law. In the fourth century BC, Plato suggested that magistrates might not want to enforce unjust contracts.60 (Rather aptly, Plato assumed magistrates would be responsible for voiding such contracts after the failure of an arbitration.) Roman law similarly forbade certain grossly unfair contracts. English lawyers, immersed in the classics, adopted the same ideas no later than the sixteenth century, including the “doctrine of unconscionability,” and exported those notions to colonial America. After the Revolution, American courts continued to void contracts on equitable grounds. The Court referenced the idea in nineteenth century, and it became part of statutes everywhere when the states adopted the Uniform Commercial Code.61 In 1965, Williams v. Walker-Thomas, still a staple of first-year Contracts classes, made the doctrine famous and helps many students feel better about their sometimes-grubby profession.62 (It’s the “furniture usury” case, for former 1Ls.) As legal pedigrees go, this is about as venerable as it gets.
Unconscionability is a doctrine with limits, not some tool for hippie judges to void contracts with Big Business. Contracts are not unconscionable merely because they are form contracts or one party has superior bargaining power. A court can void contracts only when the contract has been procured via means either unethical, unknowing, or unfree, or when the result of enforcement would be grossly abusive.63 Thus, arbitration clauses negotiated between large corporations are not unconscionable. Neither are clauses between consumers and companies where consumers receive some reasonable compensation for giving up their right to court.64 But the worst arbitration clauses are one-way provisions extracted without a consumer’s knowledge, free consent, or adequate compensation. They are, practically speaking, licenses to pilfer a few hundred dollars a year from consumers, raising serious questions of unconscionability; some state courts have held as much. The Court, however, says such contracts are fine, though a fair reading of the FAA and centuries of sensible precedent say the Court is wrong.65
The Court’s enforcement of questionable arbitration clauses opens divides between what the law says and how the law works, undermining legal legitimacy. Recall that Fuller’s jurisprudential principles require law to be intelligible and predictable; the Court’s disregard for the FAA’s plain text and venerable precedent disserves those goals. The Court further undermines legal legitimacy by trampling on states’ Constitutional rights to determine their own contract law (traditionally a big deal for federalism-minded conservatives when they’re not pushing their arbitration agenda) and by shunting important policy goals into arbitration’s black hole. Previously, claims enforcing statutory rights, including the right to be free from discrimination and harassment, were thought non-arbitrable. Legislatures adopted those employment laws to advance the policy of fair employment. Parties generally cannot contract out of legislative mandates: an employee cannot be required to agree to sexual discrimination any more than a person can contract to sell himself into slavery. If, and only if, arbitration provides a truly adequate substitute for courts—and we know it often cannot—can arbitration of statutory claims be considered just. Otherwise, it’s just a court-sanctioned back door to evade the law.
As arbitrations are confidential, many of these statutory violations occur without public scrutiny. The problem here is that public enforcers—agencies like the FTC, SEC, CFPB, etc.—have long relied on private claims to give them notice of when industries misbehave. Law enforcement becomes more time consuming and expensive, with agencies partly shut off from private discoveries; policy goals suffer; and law’s legitimacy—here, that important statutes will be enforced—withers.
Closeted arbitrations also thwart legal development, because even when arbitrators provide written justifications for judgments, those ersatz opinions are rarely as structured and rigorous as those produced by judges/clerks. They are also not public, and never precedential; they just do their work and disappear. Once a case slips inside arbitration’s event horizon, no information emerges. We cannot fully know how or on what bases the parallel justice system resolves cases until all arbitration firms mutually agree to release data (which they have not done, and will not do). The consequences of private justice for public policy have not attracted enough attention. Our common-law system is premised on the evolution that occurs one public precedent at a time, building and clarifying the law openly. Presently, there’s so much litigation that arbitration poses no great threat to precedent, but as whole areas of the law vanish like species of butterfly in clear-cut rainforests, the effect will become large.
In contrast to consumer disputes, arbitration in its classic contexts has more benefits than costs. Arguably, the benefits are growing as international trade expands. Resolution of arcane, trans-border disputes between large corporations needs arbitral outlets because national courts lack the structural capacity to handle these disputes quickly and without the appearance of bias. NAFTA’s arbitration provisions have their flaws, but forcing Canada and Bombardier to litigate against Boeing and Washington in American courts (or vice versa) may lead to more friction than would international arbitration. Rapid inflation of legal costs also makes arbitration attractive for domestic tussles that involve more than small claims, but less than bet-the-company trials for which legal costs are a tiny fraction of the sums at risk.
For everything else, arbitration risks the integrity of the legal system, usually without sufficient compensation. The Court premised its arbitration cases on the idea that out-of-court procedures would be faster than normal litigation. But the judiciary should have used the tools already at its disposal before ginning up tenuous doctrine. Courts already have procedural rules that allow judges to expedite discovery, push cases along quickly (the “rocket docket”), and to herd litigants into pretrial mediations and settlement conferences before either side becomes too entrenched in its litigation position.66 Speeding up litigation doesn’t require napalming entire classes of legal claims; it just requires judges to use the tools they already have, though more energetically.
Nor is arbitration some sideshow in the greater legal discussion. As John Marshall noted in Marbury, “the very essence of civil liberty certainly consists in the rights of every individual to claim the protection of the laws, whenever he receives an injury.”67 One purpose of an open, rule-driven adjudication is to assure citizens that they will be heard fairly. Anything less compromises the rule of law, with savings far exceeded by the costs. Indeed, the rule of law is sufficiently important that the capitalism-minded World Bank compiles an annual index.68 In a $20 trillion economy, the benefits of a stable and transparent legal system are worth hundreds of billions, at least. Yet, the entire federal judiciary costs just $7 billion, not even enough to buy 55 percent of the Gerald R. Ford aircraft carrier or to fund federal health-care programs for fifty hours.69 Adding judges to expedite litigation would be a rounding error in the budget. So whatever metaphor one chooses—the arsenal of democracy or health of the body politic—proper justice is hardly an unaffordable luxury, even adding in whatever extra billions court litigation demands in legal fees and delays. Perhaps Rehnquist believed he struck a good deal by pawning courts’ monopoly on justice, but the savings haven’t materialized. Courts, their many limitations notwithstanding, serve functions arbitration cannot. Judge Judy should be entertainment, not litigation’s new normal. But it is, and undoing the slide into arbitration will require, among other things, addressing legal costs, which are such an important rhetorical device for arbitration fans.