Do You Want COVID-19 with That?:Public Nuisance and Worker Protections at McDonald’s

Ieshia Townsend was scared to return home after her job at a South Side McDonald’s, she said at a rally for frontline workers in downtown Chicago: she could infect her children with coronavirus. Other McDonald’s workers at the rally may have gotten peace of mind from a decision in Cook County court one month earlier. In Massey v. McDonald’s Corp. (Ill. Cir. Ct. 2020), Judge Eve Reilly partially granted a preliminary injunction in a class action against three McDonald’s franchises on the West and South Sides, finding that an increased risk of coronavirus transmission could constitute a public nuisance. Though the employers had taken some measures to mitigate the spread of COVID-19, their social-distancing and mask-wearing policies were deficient. Judge Reilly required the employers to comprehensively train their employees on social distancing and to enforce mask wearing.

As William Landes and Judge Richard Posner observed in an influential 1984 article: “Regulation and tort law are alternative methods (though often used in combination) for preventing accidents.” The COVID-19 pandemic has brought particular urgency to the question of whether the liability system can substitute for regulation in the promotion of public health. In workplace safety cases, tort liability is supplemented—and sometimes precluded—not only by federal, state, and local safety regulators, but also by liability imposed in state-level workers’ compensation proceedings, as discussed infra in Part II of this Essay. Yet so far, the pandemic has led to little regulation, and far fewer workers’ compensation claims than expected. In contrast, there has been significant litigation during the pandemic.

Indeed, workers in Oakland, California, brought a similar suit to Massey. In Oakland, both the employer misconduct and the prospective relief were rather more severe. At the “doggy diaper” McDonald’s, workers alleged that they were given dog diapers and coffee filters to use instead of masks. The store shut down in May when the employees refused to work after a COVID-19 outbreak. In July and again in August, Judge Richard Seabolt issued preliminary injunctions in Hernandez v. VES McDonald’s (Cal. Super. Ct. 2020) ordering McDonald’s to allow sick workers to stay home and to provide them with adequate protective equipment, wellness checks, testing, and handwashing breaks.1 (Judge Seabolt’s written decision is unavailable, so this explanation comes solely from news reports and court filings.) The case settled in August 2021; the settlement agreement created a joint “management-worker committee” that will meet monthly to discuss safety measures at the restaurant.

Massey and Hernandez show that workers can win COVID-19 protections via public nuisance litigation. Yet the cases’ biggest impact may be in the streets and the statehouse, not the courtroom. Though the liability system can sometimes substitute for regulation, here, causation issues and low damages make public nuisance cases and workers’ compensation claims unlikely to spur broader compliance. Rather than a replacement for regulation, the McDonald’s litigation is best seen as one of a suite of strategies that has emerged from the past few years of fast-food worker activism in the “Fight for $15” movement. Along with direct action and lobbying elected officials, workers are litigating to challenge their employers, build momentum for political change, and prod regulators that have been asleep at the wheel.

In this Essay, I argue that the liability system’s role in directly promoting workplace safety during the pandemic has been—and will likely continue to be—limited. Nevertheless, workplace safety litigation’s indirect social effects may lead to safer workplaces in the future. This Essay proceeds in four parts. First, I examine the legal elements of a public nuisance claim and the particular challenges plaintiff-employees face in the COVID-19 context. Next, I turn to another bar that might stop similarly-situated plaintiffs from bringing additional suits—judicial deference to safety regulators and workers’ compensation systems, the administrative substitutes for tort law. Then, I use law-and-economics theory to show that the possibility of liability remedies for workplace conditions cannot replace the Occupational Safety and Health Administration (OSHA), the federal agency in charge of workplace safety. Finally, I situate the McDonald’s cases against the backdrop of worker organizing efforts to understand the social context that animates the public nuisance litigation and in which its effects will resonate. Even if the liability system as currently constituted provides insufficient incentives for adequate workplace safety, strategic litigation can catalyze social responses that could revitalize the regulatory state, thereby ensuring that workers ultimately receive the protections they deserve.

I. Public Nuisance and COVID-19 Worker Protections

Though the Massey and Hernandez plaintiffs succeeded in obtaining preliminary relief, there are barriers to the use of public nuisance law to address insufficient COVID-19 worker protections. To show a public nuisance, a plaintiff must demonstrate that there is a public right, the defendant’s conduct has infringed on that right, and the plaintiff has been injured by the nuisance. The standard is essentially the same in Illinois and California. Plaintiffs can likely establish that there is a public right to reduced coronavirus risk at McDonald’s. In so doing, they characterize their workplace conditions as an issue that affects everyone. But the other prongs will vary significantly across cases and maybe across judges. Causation—showing that a COVID-19 infection was caused by the defendant employer—will be particularly challenging because tracing the source of a particular infection is difficult.

A. A Public Right to COVID-19 Precautions

To win a public nuisance case, plaintiffs need to show that they are vindicating a communal right, not their parochial employment interests. To be a public nuisance, a business’s noncompliance with COVID-19 precautions must “affect an interest common to the general public” or be “harmful to the public health.”2 Despite weak direct support from caselaw, higher risk of coronavirus transmission probably rises to this level. The spread of COVID-19 is a public health concern, and harms to public health are classic public nuisances. The Restatement (Second) of Torts states that “the threat of communication of smallpox to a single person may be enough to constitute a public nuisance because of the possibility of an epidemic.”3 Simply due to the nature of viral transmission, people far away from a coronavirus hotspot are still threatened by it. For example, a health-care conference in Boston in March 2020 led to coronavirus cases in at least six states—an early “superspreading” event.

Even though COVID-19 spread fits the doctrinal test for harm to the public, caselaw in California and Illinois is not precisely on point. In Massey, Judge Reilly cited Heigert v. Reidel (Ill. App. Ct. 1990) in support of the proposition that “public health includes monitoring the spread of infectious diseases.” In Heigert, a nurse caught tuberculosis from a patient, allegedly due to her supervisors’ negligence. But Heigert was an individual negligence action. The court emphasized the particular risks the plaintiff faced as a nurse, rather than the common harm of disease transmission. In a pair of 1930s cases, the California Supreme Court found that the state can exterminate tuberculosis-ridden cattle and destroy trees infected with “peach mosaic,” a viral disease, because preventing the spread of a contagious disease is within the police power to abate a public nuisance.4 In their focus on direct state action, these cases are somewhat far removed from a business’s failure to mitigate disease spread. But later California rulings allow private actors to be enjoined from engaging in conduct that causes a public nuisance.

Harms to public health are typical public-nuisance claims. So it seems likely, despite somewhat weak direct support, that there is a public right to COVID-19 protections for the purposes of a public-nuisance injunction. Plaintiffs are more likely to struggle with other aspects of the public nuisance claim—the relationship of the defendants’ conduct to the plaintiffs’ injury—as well as external doctrines that might preclude relief.

B. The Effect of Defendants’ Conduct on Plaintiffs’ Injury

Having shown that there is a right to coronavirus precautions, plaintiffs next need to establish that a defendant’s conduct violated that right and caused an injury to the plaintiff. These aspects of the public nuisance claim are quite case- and judge-specific. The bulk of Judge Reilly’s opinion is devoted to a fact-intensive analysis of the particular employer’s practices to determine if they infringe on rights to public health. For the injury, Judge Reilly explained that in a preliminary injunction, it is unnecessary to show that defendants have actually caused a COVID-19 case—just that their conduct has “ma[de] a positive case ‘highly probable.’” The Hernandez plaintiffs experienced an actual, not merely probable, injury—an outbreak at the store.

The hardest part will be causation: showing that an injury (or probable injury) was (or is likely to be) caused by the employer. It is hard to say where, when, or why a particular infection occurred. If an employee gets COVID-19, transmission may not have occurred at the store, or may have occurred in spite of the employer’s precautionary measures. Even if there is an outbreak at a store, if there is high community transmission as well, it may be just as likely that an employee caught coronavirus outside work. Success on such claims will depend on judicial (or jury) evaluation of these likelihoods, which vary.

Indeed, in a similar case, a federal judge found that the plaintiffs could not satisfy causation, although that specific finding was regarding the likelihood of harm for the purposes of a preliminary injunction, rather than the causation element of a public nuisance claim per se. In Rural Community Workers’ Alliance v. Smithfield Foods (W.D. Mo. 2020), workers at a meat-processing plant alleged that the plant did not take sufficient precautions to mitigate COVID-19 spread. The plaintiffs sought a declaratory judgement and a preliminary injunction requiring the plant to implement a number of safety measures. The judge ruled that the risk of harm was “too speculative” to justify an injunction.5 The court explained that the fact that “employees risk exposure if they continue to work” does not mean that “the spread of COVID-19 at the [p]lant is inevitable or that Smithfield will be unable to contain it if it occurs.” In Smithfield, like Massey and Hernandez, the court confronted a workplace where some precautions were being taken but the spread of COVID-19 was possible, though transmission was also likely in the communities employees lived in. Whether that risk was high enough to justify a judicial response seemed to depend on the particular judge.

II. Bars to Tort Claims and Administrative Substitution for Tort Law

In addition to facing difficulty proving causation, plaintiffs in public nuisance cases might be barred by other doctrines that courts use to avoid infringing on other governmental actors. This section outlines these doctrinal challenges, as well as introducing the administrative substitutes for tort law to which courts may defer in COVID-19 workplace litigation. Under the primary-jurisdiction doctrine, courts defer to administrative agencies with relevant expertise, like OSHA, the federal workplace-safety regulator, and analogous state and local agencies. And under the exclusive remedy rule, courts will dismiss actions for damages for workplace injuries; state legislatures established workers’ compensation systems to provide the only avenue for such claims.

A. OSHA and the Primary-Jurisdiction Doctrine

Primary-jurisdiction doctrine is a discretionary barrier that public nuisance plaintiffs have faced in similar cases to Massey and Hernandez. As both a matter of legal doctrine and judicial practice, it is less likely to apply in state courts than federal courts. Under primary jurisdiction, courts can dismiss or stay cases and refer them to the appropriate regulatory agency even if the court would otherwise have jurisdiction. There is no “fixed formula for applying the doctrine,” but judges should look to “promot[e] proper relationships between the courts and administrative agencies” and achieve “uniformity” in regulatory administration.

When it comes to workplace safety, the relevant federal agency is OSHA. Established by the Occupational Safety and Health (OSH) Act of 1970, the agency is part of the Department of Labor and is responsible for setting standards to “assure[ ] the greatest protection of the safety or health of the affected employees.” In addition to setting standards, the Secretary of Labor can obtain an injunction in district court to restrain practices at a particular workplace if there is an “imminent danger” of “death or serious physical harm.” Employees can file a writ of mandamus if “the Secretary arbitrarily or capriciously fails to seek [such] relief.”

Two federal courts dismissed public nuisance claims during the pandemic on primary-jurisdiction grounds. Aside from finding a low likelihood of harm, as discussed above, the Smithfield court applied primary jurisdiction to defer to the “special competence” of OSHA and the Department of Agriculture (which had authority at the meat-processing plant along with OSHA pursuant to an executive order). The court explained that OSHA was “better positioned” than the court to evaluate appropriate workplace regulation, and regulation by OSHA would lead to more “uniform national enforcement” of safety standards. A public-nuisance claim against Amazon was dismissed for the same reason in Palmer v. Amazon.com, Inc. (E.D.N.Y. 2020).The Palmer court explained that “[r]egulating in the age of COVID-19 is a dynamic and fact-intensive matter fraught with medical and scientific uncertainty”—better suited for “OSHA’s expertise and discretion” than for judicial resolution.

Yet despite its expertise, OSHA has been somnolescent during the pandemic. The agency has done few workplace inspections and issued no new safety regulations. In fact, under the Trump Administration, OSHA “issued a series of guidelines and directives that have weakened protections for front-line workers” (emphasis added). And OSHA’s penalties under existing regulations have been too little, too late. The agency issued its first coronavirus-related citations on September 10, 2020, for only $13,494, after four workers died of COVID-19 at a Smithfield meat-processing plant (a different plant than the one at issue in Smithfield).

Thus, in COVID-19 public nuisance litigation, federal courts referred claims to agencies that seemed unlikely to take them up. The Smithfield court suggested that the plaintiffs petition OSHA to obtain an emergency injunction pursuant to the statutory scheme, and file a writ of mandamus if OSHA refused. Yet inspections, much less emergency injunctions, were not forthcoming from OSHA during the Trump Administration (or under any other administration),6 and a Pennsylvania district court rejected the one such writ of mandamus that was filed during the pandemic. The court explained that the writ is only available if an OSHA inspector has first made a “finding of imminent danger.” That seems unlikely, for an agency that was not doing much enforcement at the time. OSHA has seen some change in the new administration, but it is slow going. After blowing past a March 15 deadline set by President Biden, OSHA finally issued emergency temporary safety standards—for healthcare employers only—on June 21, 2021. (This was a separate rule from OSHA’s controversial vaccination standard that was recently struck down by the Supreme Court).

Because deference is a matter of judicial discretion, the Smithfield and Palmer courts were within the law, but those courts are outliers in federal judicial practice. Judges rarely use primary jurisdiction. Professor Richard Pierce has explained that the doctrine is “no longer viable in today’s conditions.” Federal courts are reluctant to punt disputes to agencies that might never resolve them, so they have largely replaced formal invocations of primary-jurisdiction doctrine with requests to agencies to submit their positions in amicus briefs. The concern Professor Pierce notes—that agencies will simply not resolve a dispute submitted to them—is certainly present in the COVID-19 cases. Criticizing Palmer, Professor Michael Duff explained that judicial expertise relative to agencies is beside the point; rather, “the question is whether a court trying to do something could do a better job than an agency ideologically committed to doing nothing.” And the courts could have requested amicus briefs from OSHA to obtain the benefit of agency expertise. Yet likely, the Smithfield and Palmer courts wanted to avoid engaging in a politically salient battle between labor and employers. Perhaps this suggests that Professor Pierce’s interment of the doctrine is premature.

But if primary jurisdiction is rare in federal court, it is perhaps even rarer in state court. In Illinois, if a court finds that the doctrine applies, a case “should never be dismissed from the court but may only be stayed.” The Illinois Supreme Court has been very reluctant to allow the use of primary jurisdiction, overruling dismissals where courts and state agencies both had concurrent jurisdiction. In other words, unlike in federal court, it would not be enough to permit application of the doctrine for an agency to have jurisdiction over COVID-19 safety precautions, if an Illinois court has jurisdiction over the dispute as well. In addition, the Court has explained that primary jurisdiction is a “matter of [judicial] self-restraint,” not an independent provision of jurisdiction.

Like Illinois, California limits the scope of the doctrine. California courts have explained that application of primary jurisdiction “lies within the court’s discretion.” They may use the doctrine “where the issues raised in the trial court action [are] pending before an administrative agency” or where there is a particular need for “specialized agency fact-finding and expertise.”7 Thus, where an issue is not actually pending before an agency, it seems likely that California courts will decide it. In fact, according to the transcript of a hearing in Hernandez, Judge Seabolt granted the injunction even though “[c]ounty health authorities came in and began to supervise” the restaurant, because he “worr[ied] . . . about the bandwidth of the . . . health authorities given the pandemic.”8 Though the transcript does not reflect a particular concern with primary jurisdiction, Judge Seabolt granted the injunction despite the role of local (and federal) agencies in regulating workplace safety because, in his judgment, agency resources were not up to the task.

In sum, both doctrine and judicial practice in Illinois and California give judges a great deal of discretion in their application of the primary-jurisdiction doctrine. This means that these state courts are even less likely to apply the doctrine than federal courts—which are also reluctant to use it. For these reasons, despite difference in its application across jurisdictions and considerable scope for judicial discretion, on the whole it seems unlikely to bar tort suits in state court against employers.

B. The Exclusive Remedy of Workers’ Compensation

The McDonald’s plaintiffs’ litigation strategy was clearly shaped by the broader legal framework of workers’ compensation. States adopted workers’ compensation schemes in the early 20th century, in response to an “explosion” of tort litigation regarding workplace injuries. Industrial conditions were extremely dangerous, and outcomes in tort litigation were wildly variable and unpredictable. Workers, employers, and insurers alike wanted a solution. In a “Grand Bargain,” workers got predictable damages for workplaces injuries (and, later, occupational diseases9 ) on a strict liability basis, in return for tort immunity for employers. Tort immunity is therefore known as the “exclusive remedy” rule: plaintiffs cannot recover for workplace injuries that are compensable under a state’s workers’ compensation law.10

Other commentators have noted that the exclusive remedy rule might bar COVID-19 public nuisance claims. Because the McDonald’s plaintiffs were seeking redress from their employers for injuries suffered at work, they ought to be channeled into worker’s compensation and barred from court. But in neither Massey nor Smithfield was workers’ compensation even discussed. And it did not bar preliminary relief in Massey or Hernandez. In Palmer, the court found that the plaintiffs’ claims were barred by the exclusive remedy rule, but as an alternative to the court’s already adequate primary-jurisdiction grounds for dismissal. Likely, the Massey and Hernandez plaintiffs included family members of workers in order to ensure that they could maintain their claim even if employees of the franchises were dismissed from the case. In addition, they asked only for injunctive relief, not damages—and it is not clear whether the exclusive remedy bar would apply to public-nuisance suits for injunctive relief.11

Professor Duff recently summarized the arguments regarding whether the exclusive remedy rule bars a public-nuisance suit for injunctive relief. A suit for damages would definitely be barred, because courts have extended employers’ immunity to civil actions for nearly all work-related injury claims, regardless of the theory of recovery. For an injunction only, it is a close question. Only plaintiffs who have “the right to recover damages” can obtain an injunction to abate a public nuisance. As noted, workers’ compensation systems preclude employees from recovering monetary damages. But the right to damages might not be coextensive with the right to receive monetary damages per se. Rather, it is more like standing: the “right to damages” distinguishes plaintiffs who have been especially harmed by a public nuisance—and thus can sue to abate it—from “minimally impacted members of the general public.” On the latter view, the exclusive remedy rule would not be implicated.

So far, no cases in state courts have squarely addressed the question. The Palmer court, however, ruled that New York’s workers’ compensation scheme would preclude a public-nuisance injunction because the statute says that “the scheme serves ‘in place of any other liability whatsoever’”; that broad phrase must include injunctive relief. Yet it is not clear that injunctive relief is a “liability” under the law in the first place. And California, for its part, seems likely to allow a suit in this context. Under California’s “dual capacity” doctrine, employees can recover against their employers for injuries suffered “based upon the existence of a duty arising independently from the employment relationship.” Perhaps because the harm arises from the duties McDonald’s operators owe to the general public pursuant to public-nuisance doctrine, rather the employment relationship, the Hernandez plaintiffs would escape the bar.

Even if the exclusive-remedy rule applies to suits for injunctive relief only, plaintiffs have a few other arguments they could use to avoid the bar. First, they might argue that the suits fall within an exception for intentional torts. Workers’ compensation statutes allow a tort action if employers intentionally harmed their employees.12 Admittedly, employers usually skimp on protection because it is cheaper, not because they want to harm their employees. Yet in some states, it only must be the case that conduct was “substantially certain” to cause injury or death to satisfy intentionality.13 Neither California nor Illinois is in that group of states, but it is reasonable to imagine that those states might join the jurisdictions that have eroded the intentionality requirement. Plaintiffs could tee up the question and argue that infection with COVID-19 is substantially certain given low levels of mitigation measures at a workplace and high disease spread in the community.

Second, plaintiffs could argue that infection with COVID-19 is outside the coverage formula of worker’s compensation statutes. Workers’ compensation only covers occupational diseases. Plaintiffs would argue that because COVID-19 “might as readily be contracted in everyday life or in other occupations” as at McDonald’s,14 it is not an occupational disease; rather, it is an “ordinary disease of life.”15  An occupational disease must arise from a “hazard which distinguishes [the occupation] in character from the general run of occupations.”16 Infection with COVID-19 might be a particular hazard of, say, working as an ambulance driver during the pandemic, but it is not specific to working at McDonald’s in the same way that a respiratory disease inhaling cooking smoke would be. Thus, COVID-19 would be excluded from workers’ compensation coverage and the attendant tort immunity.

But this argument is almost certainly a non-starter, ironically, precisely because of legislative activity during the pandemic intended to protect workers. California passed a statute (codifying an earlier executive order) that created a presumption that first responders, healthcare workers, and workers at a workplace that experienced a COVID-19 outbreak were infected at work, for the purposes of workers’ compensation. It also defined an “injury” to expressly include infection with COVID-19, but only if the worker was diagnosed within two weeks of a work day between March 19, 2020, and July 5, 2020. Illinois enacted a similar presumption, but only for first responders and healthcare workers. These presumptions shift the burden to the defendant to refute causation, making employer liability more likely. But to the point here, the legislative enactments indicate that workers’ compensation is intended to encompass COVID-19, at least under some conditions.

It is somewhat mysterious why the Massey and Hernandez courts did not consider the exclusive remedy rule. The employers certainly must have argued that workers’ compensation systems, not courts, are the appropriate venue for these claims. But plaintiffs have reasonable arguments that they should be able to proceed in court, most plausibly, that an action for a public nuisance injunction is outside workers’ compensation’s ambit. Perhaps the courts decided that in the context of motions for preliminary injunctions, the plaintiffs had a high enough likelihood of success—even considering arguments over workers’ compensation’s exclusivity—that injunctions were appropriate.

III. Why Liability Is a Poor Replacement for COVID-19 Worker Protection Regulations

Despite the primary-jurisdiction doctrine and the exclusive remedy rule, workers in Chicago and Oakland succeeded in getting preliminary injunctions for public nuisances. But more generally, in the absence of regulatory action, does the prospect of liability provide worker protections commensurate to the public health challenge of COVID-19? A birds-eye view suggests not. The United States has had one of the worst coronavirus outbreaks in the world,17 driven by infection at service jobs. Outbreaks among essential workers have been proffered as an explanation for higher rates of the disease in Black and Latino communities.

Theoretical findings within the law-and-economics paradigm can explain why neither tort liability nor its workplace substitute, workers’ compensation, provides adequate controls on employers’ use of COVID-19 precautions. Professor Steven Shavell developed a formal model to compare the effects of ex-ante regulation to tort liability in incentivizing potential tortfeasors to take optimal care. Under the model, regulation is more likely than liability to lead to optimal care as the probability of suit declines, the harm is less variable across defendants, or defendants do not internalize the full cost of harm via assessed damages.18

Professor Shavell’s factors tend toward regulation over liability, whether liability is imposed in public-nuisance actions or workers’ compensation proceedings. To the first factor, there have only been a handful of public nuisance suits regarding COVID-19 workplace safety. The workers most affected by dangerous conditions at work have few resources to hire lawyers. Success could bring more lawsuits, but cases asking for prospective relief will have low, if any, money damages, meaning that they are unlikely to incentivize the private bar.

On the workers’ compensation side, these systems tend to be underused. Small employers and entire classes of workers like domestic workers and farm workers are simply excluded from some state systems. Many workers never actually file for benefits, perhaps deterred by employers, fears of stigma or retaliation at work, lack of legal sophistication or English proficiency, or worries that use of the system would alert authorities to their immigration status. Even “people who sustain injuries at work as clearly compensable as amputations (amputations!) do not file claims.” Filing for occupational disease claims is even rarer than injuries. Though it is hard to draw conclusions from the available data (unfiled claims, of course, are not recorded anywhere), in California, there have been 250,922 COVID-19–related workers’ compensation claims overall during the pandemic19 —representing under 2.8% of the 9.1 million cases in the state so far. Yet a median of 36.7% of COVID-19–related claims were denied, compared to 12.95% of unrelated claims. These denials are probably due to the difficulties discussed above in showing that a disease transmission occurred at work, rather than somewhere else. Even in states like California that have adopted burden-shifting presumptions, workers’ compensation insurers have denied “a significant percentage” of claims related to COVID-19.

The second factor is whether harm is variable across potential defendants. The less variable, the better regulation looks relative to liability. Certainly, some worksites will see more outbreaks, simply due to the stochasticity of disease transmission. But over time, worksites that use the same mitigation measures are likely to generate similar numbers of new infections, given the same baseline infection rate in the community.

The third factor is how much of the cost of harm defendants will internalize via tort damages. If they internalize the full cost of harm, liability is more likely than regulation to lead to optimal care. Because many McDonald’s franchisees are small business owners with precarious finances, their assets may be less than the full costs of harm. In addition, transmission at McDonald’s could lead to transmission chains in the community—exponential disease spread causing exponential harm not reflected in damages. Damages in workers’ compensation systems are limited to medical costs for the affected worker. In public nuisance, the Hernandez and Massey plaintiffs asked only for injunctive relief, not damages, and because of the exclusive-remedy rule, all potential public-nuisance plaintiffs would probably pursue such a strategy.20 In that case, the cost incurred by losing defendants would just be the cost of precautions, not harm—and precautions are probably much less expensive than the costs of a disease outbreak. As one notable commentator observed, “an Ounce of Prevention is worth a Pound of Cure.” Thus, these factors tend towards ex-ante regulation over tort liability to control behavior.

Professor Saul Levmore has called cases “recurring misses” when defendants increase the chance of harm but remain under the preponderance-of-the-evidence causation standard: “The identifying feature of these cases is . . . that there is a wrongful party who is more than 0 percent but possibly never more than 50 percent likely to have caused an injury.” As we have seen, causation in COVID-19 worker-protection cases has this exact structure. Employers increase the chances of infection by failing to take precautions, but that failure may not be the cause of any particular infection. The straightforward solution to recurring misses is to use “regulatory or administrative sanctions to overcome the deterrence gaps in the [liability] system.”

Professor Levmore suggested that within the existing tort system, when faced with recurring misses, judges strategically use proximate cause and burden shifting to hold defendants partially liable in some cases to set proper levels of deterrence. Judge Reilly’s finding that infection at the Chicago McDonald’s locations was “highly probable” may have had more to do with her judgment of appropriate employer deterrence than causation. In practice, primary-jurisdiction doctrine may be another lever judges use to avoid defendant liability when they believe there is already adequate deterrence. But judicial discretion in determining which plaintiffs get relief could exacerbate the already disparate racial effects of COVID-19. Even if it is possible for judges to do such fine-grained deterrence-setting through liability, regulation may be superior as a normative matter.

IV. Public Nuisance and Worker Protections in Context

Under the Trump Administration, OSHA abdicated its responsibility to assure workplace safety during the pandemic.21 With the door to traditional workplace regulation shut, the desire for greater mitigation measures found a predictable outlet in tort litigation. Workers may achieve partial success in court, but this door may be closing as well. A number of states have responded to coronavirus by passing liability shields for businesses. Senate Republicans attempted to do the same on a national scale, though prospects for such legislation are dimmer with a Democratic President.

Yet it would be a mistake to evaluate the McDonald’s public nuisance cases solely through the lens of courtroom success or regulatory efficacy. Both cases emerged from broader worker organizing strategies. In Chicago, frontline fast-food workers marched in the Loop as part of a nationwide labor action, drawing connections between precarious conditions at work and the Black Lives Matter movement. Before Judge Seabolt granted relief to protect workers at the Oakland McDonald’s, they went on strike to protect themselves. Professor Kate Andrias has chronicled fast-food worker activism in the Fight for \$15 movement, explaining how the movement has expanded beyond traditional workplace organizing to enlist the state in tripartite bargaining between workers, industry, and the public, via direct action and political lobbying. A New York campaign began with walk-offs at fast-food restaurants in 2012 and resulted, in 2015, in the New York Wage Board—comprised of one representative each from labor, management, and the public—raising the minimum wage for fast-food restaurants to $15 an hour. By bringing in the state as a bargaining party, “the campaign is transforming the post-New Deal conception of labor disputes as private affairs” into one that sees labor conditions as directly linked to public concerns.

The McDonald’s cases are both a product and constituent of this activism. Fast food workers’ aggressive litigation strategy reflects an employment relationship charged by the militancy of Fight For $15. And by channeling their complaints through the legal framework of public nuisance, the plaintiffs reinforce Fight for $15’s framing of labor issues, treating the workplace practices of private employers as a set of social conditions that affects the public at large. In its articulation of workplace protections as a public right vindicated by workers, the litigation strategy is linked to “bargaining for the common good,” a negotiation strategy recently used by teachers in Chicago in which unions incorporate community demands into collective bargaining. The expressive function of pursuing public nuisance as a vehicle for claims is that workers are addressing not only their parochial interests, but the public interest. Restaurant safety affects customers at McDonald’s, their families, and the larger communities that the restaurants are situated in.

Massey and Hernandez might directly lead to broader social reform of the workplace outside of the liability system in which the cases arose.22 Litigation dramatizes issues and drives media coverage.23 In the pay equity campaigns of the 1980s, for instance, litigation received more than five to ten times the media coverage as legislative, electoral, direct-action, or negotiation strategies.24 Far more than regulation or use of workers’ compensation systems, public nuisance litigation is likely to lead to increased public consciousness of workplace safety. (That said, lack of regulation has certainly engendered significant media coverage as well.) As workplace safety becomes more salient to voters, legislators and regulators might respond to public pressure with more stringent requirements for businesses or greater benefits for workers—for example, the state legislatures that have adopted burden-shifting presumptions in workers’ compensation for COVID-19.25 Members of the public might take action at their own workplaces, as well.26

Besides the wider public, participation in tort litigation has an effect on workers, binding them together to pursue common goals. Professor Benjamin Sachs has argued that statutory employment rights can become a “locus” for the formation of “collective identity” and worker organizing. For example, a group of New York garment workers took action after learning of their federal right to overtime pay, resulting in an ultimately successful political and legal challenge to their employer’s violation of that right—and to the Department of Labor’s under-enforcement of federal law. Similarly, judicial recognition of public nuisance rights can help workers conceptualize an unsafe workplace as a legal wrong which they can collectively redress, whether or not relief ultimately comes through the liability system itself.

Thus, workers’ participation in litigation could feed into their participation in the Fight for $15 movement. Analyzing campaigns in the 1970s and 1980s, Professor Michael McCann explained that workers’ participation in pay equity litigation was “a catalyst to other forms of political activism.”27 Litigation campaigns “raise expectations” about what workers can achieve, and provide a rights discourse that empowers workers to “identify and criticize [ ] hierarchical relations.”28 Fast-food workers who are activated by litigation may see workplace reform as more attainable than they did before, leading them to get involved in other efforts. The Chicago and Oakland McDonald’s workers will likely continue to lobby federal and state workplace-safety officials after litigation runs its course.

Once engaged, McDonald’s workers might broaden their horizons beyond workplace safety. They may take a more combative tack toward other aspects of their working conditions after their involvement in the lawsuits. Perhaps this would lead to the formation of unions. Fast food remains largely non-union because the franchise structure has created legal difficulties for union organizers, despite Fight for $15’s efforts to unionize the sector. Often, litigation draws new members into unions, helping advertise the union to potential new members.29

If Massey and Hernandez do lead to unionization, that could have a more enduring effect on workplace safety than the injunctive orders in the cases. Unions can collectively bargain for increased workplace-safety measures.30 In addition, unionization would increase the efficacy of both regulation and the liability system. Unions play a “critical role[ ]” in “amplifying the practical effect of law within workplaces,” not only through private litigation to enforce laws, but through “communicating and interpreting statutory rights within the workplace.”31 Workers in unions file more workers’ compensation claims.32 And unionized workplaces are more likely to see safety complaints and inspections for safety violations, ensuring employer compliance with regulations—at least when regulators conscientiously enforce the law after complaints. Unionization, then, might improve the ability of the liability system to respond to the next pandemic.

In fact, organized labor was instrumental to the creation of OSHA in the first place. Unions aggressively lobbied Congress for national health and safety legislation. Labor actions were also key to the passage of the legislation. A “wave of combative strikes over dangerous workplace conditions shook heavy industry” in the 1960s and early 1970s. In a 2011 speech to Public Citizen, an OSHA administrator connected a strike by Memphis garbage collectors—sparked both by the death of Dr. Martin Luther King, Jr. and two coworkers who were “crushed to death in a trash compactor”—to Congress’s passage of the OSH Act three years later. Yet today, OSHA has limited resources, and most workplaces are not even inspected, much less remediated. Commentators have suggested that Congress increase funding for OSHA, allow OSHA to impose larger fines, or create a private right of action to enforce the OSH Act. Congress will not act on its own. Worker organizing, perhaps catalyzed in part by these and similar suits, will likely be necessary to revitalize the muscular regulatory state that can protect the health of everyone.

Conclusion

Wins in court are not the only way that the liability system improves health and safety. In Massey and Hernandez,McDonald’s workers challenged lax standards at the restaurants where they worked, arguing that the threat of COVID-19 spread created a public nuisance. Both of these cases were successes for the plaintiffs: Judge Reilly and Judge Seabolt issued preliminary injunctions requiring the restaurants to boost safety measures. But there are barriers to public-nuisance litigation that aims to increase worker protections. Especially in federal court, courts might defer in favor of safety regulators or state workers’ compensation systems, two schemes that have developed to complement tort law. And even if workers prevail, success in these cases probably will not translate directly into broader compliance with safety measures among employers not subject to suit. Yet action in the liability system is connected with broader social change. For private litigation to meaningfully improve workers’ health, we need the political will to create a higher floor of regulated safety standards and lower barriers to tort and workers’ compensation systems for potential plaintiffs. Aside from winning real gains, the Massey and Hernandez plaintiffs raised public consciousness, and perhaps their own, as well. The full realization of their goals will happen outside the courtroom.

• 1See Notice of Filing of Proposed Preliminary Injunction Order, Ex. A, Proposed Preliminary Injunction Order, at 2–4, Hernandez, No. RG20064825 (Cal. Super. Ct. Aug 25, 2020).
• 2See also People ex rel. Gallo v. Acuna, 929 P.2d 596, at 604 (Cal. 1997) (explaining that a public nuisance is an “interference with collective social interests” that is “substantial and unreasonable” (emphasis omitted)).
• 3Restatement (Second) of Torts § 821B cmt. g (Am. Law. Inst. 1979).
• 4See also People ex rel. Lockyer v. Sun Pac. Farming Co., 92 Cal. Rptr. 2d 115, 119 (Cal. Ct. App. 2000) (affirming a ruling that the “maintenance of citrus trees infected with the citrus tristeza virus (CTV) constitutes a public nuisance”).
• 5In the alternative, the court ruled that the primary-jurisdiction doctrine applied, see infra Part II.A, and that the plaintiffs were unlikely to succeed on the legal elements of their public nuisance claim because of the mitigation measures Smithfield had already implemented.
• 6The Department of Labor stated in legal proceedings that it has sought such an injunction only three times.
• 7But see Farmers Ins. Exch. v. Superior Court, 826 P.2d 730, 747 (Cal. 1992) (Mosk, J., dissenting) (“California has never recognized the doctrine of primary jurisdiction.”).
• 8Notice of Filing of Proposed Preliminary Injunction Order, Ex. B, Reporter’s Transcript of Proceedings (Aug 13. 2020), at 12, Hernandez, No. RG20064825 (Cal. Super. Ct. Aug 25, 2020).
• 94 Larson’s Workers’ Compensation Law § 52.01.
• 109 Larson’s Workers’ Compensation Law § 100.01.
• 11See Robert Iafolla, McDonald’s Case Tests Nuisance Theory for Job Virus Safety, Bloomberg Law (June 4, 2020) (reporting Professor Michael Duff’s opinion that plaintiffs “pursuing injunctions under a nuisance theory” might be able to circumvent the exclusive-remedy bar, “depending on the state”).
• 129 Larson’s Workers’ Compensation Law § 103.03.
• 139 Larson’s Workers’ Compensation Law § 103.04[1].
• 144 Larson’s Workers’ Compensation Law § 52.03[3].
• 15Id.
• 164 Larson’s Workers’ Compensation Law § 52.03[4] (quoting Underwood v. Nat’l Motor Castings Division, 45 N.W.2d 286, 287–88 (Mich. 1951)).
• 17As of April 2022, the United States had experienced over 79 million reported coronavirus cases (the most of any country in the world by over a factor of two) and had the twentieth highest death rate.
• 18This is actually a slight extension of Professor Shavell’s model. His version of the third term is just the defendant’s level of assets. But the term is intended to model how much of the cost of harm defendants internalize, and Professor Shavell assumes that the only limitation is the defendant’s assets; otherwise, damages equal harms, stating: “[I]f a party causes harm h and is sued, he will be liable for h, but will pay that amount only if h ≤ y [his level of assets].” Here, there are other reasons why tort damages will not reach the full cost of harm.
• 19COVID-19–related claims have comprised a median of 12.7% of monthly claims during the pandemic, in context of a 4% decline in claims. It stands to reason that workers’ compensation claims would decline overall—people working from home are probably less likely to be injured.
• 20See supra Part II.B.
• 21See supra Part II.A.
• 22This discussion builds on Michael McCann’s analysis of the role of litigation in the movement for pay equity in the 1970s and 1980s. See generally Michael McCann, Rights at Work: Pay Equity Reform and the Politics of Legal Mobilization (1994). McCann analyzes workplace litigation through the lens of “legal mobilization,” an analytic framework that focuses attention away from top-down judicial decision-making, and instead on how law “structures social relations” and shapes the way legal claimants understand themselves and pursue their goals. See George I. Lovell, Michael McCann, and Kirstine Taylor, Covering Legal Mobilization: A Bottom-Up Analysis of Wards Cove v. Antonio, 41 Law & Soc. Inquiry 61, 63 (2016).
• 23See e.g., McCann, supra note 23, at 58–61.
• 24Id. at 59.
• 25See supra Part II.B.
• 26See McCann, supra note 23, at 62 (explaining how pay-equity litigation was “carefully coordinated with publicity campaigns to dramatize the wage discrimination issue . . . to activate potential advocates for the cause”).
• 27Id. at 81.
• 28Id. at 64–65.
• 29See also Trevor Colling, What Space for Unions on the Floor of Rights? Trade Unions and the Enforcement of Statutory Individual Employment Rights, 35 Indus. L.J. 140, 158 (2006) (demonstrating, via case studies of two British unions, that litigation strategies help recruit new union members from unorganized sectors).
• 30See also McCann, supra note 23, at 141–42 (explaining that union activists consider collective bargaining to be “the most effective avenue for [pay equity] reform,” but litigation is a “useful leveraging tactic” in negotiation).
• 31Trevor Colling, Trade Union Roles in Making Employment Rights Effective, in Making Employment Rights Effective 183, 194 (Linda Dickins ed., 2012).
• 32See also id. at 195.