The Volokh Conspiracy

Mostly law professors | Sometimes contrarian | Often libertarian | Always independent


A Backgrounder on the Proposed OSHA Vaccination Mandate and Likely Legal Challenges (Updated)

Why legal challenges to the new rule are more likely to focus on the details than on broad challenges to OSHA's authority.


President Biden's announcement that the Occupational Safety and Health Administration (OSHA) will require large employers to mandate the vaccination or testing of their employees has already prompted a raft of commentary and controversy. The standard, once issued, is sure to invite legal challenges, the most serious of which will likely focus on the yet-to-be-disclosed particulars of the rule. This post is intended to provide some background on OSHA's authority and identify the legal issues the rule is likely to create. (As will become clear, I see some of the issues differently from some of my co-bloggers.)

A quick note: My employer requires vaccination (in addition to periodic testing) and I support that. I wish more people were vaccinated and that more employers required it of their employees. I also believe that the control of infectious disease is one of the more important functions of government. This post is not about those questions, however, but about OSHA's authority to adopt the standard the White House has announced.

The White House announcement describes the new policy as follows:

The Department of Labor's Occupational Safety and Health Administration (OSHA) is developing a rule that will require all employers with 100 or more employees to ensure their workforce is fully vaccinated or require any workers who remain unvaccinated to produce a negative test result on at least a weekly basis before coming to work. OSHA will issue an Emergency Temporary Standard (ETS) to implement this requirement. This requirement will impact over 80 million workers in private sector businesses with 100+ employees.

The same proposed standard will also require employers "to provide paid time off for the time it takes for workers to get vaccinated or to recover if they are under the weather post-vaccination."

We do not have the details of the OSHA rule—and the details will matter.  Nonetheless, it is possible to identify the legal issues that this new OSHA standard will (and will not) raise.

For starters, it is important to recognize that this proposed standard represents a far more traditional use of agency authority than did the CDC's eviction moratorium. While this is unquestionably an aggressive use of OSHA authority, it does not represent a dramatic departure from the sorts of things OSHA has done over the past fifty years, and does not raise quite the same specter of an agency without unlimited authority that the CDC moratoria did. This doesn't mean the standard should or will survive legal challenge, but it does indicate the issues are somewhat different.

Under the Occupational Safety and Health Act of 1970, OSHA has broad authority to set workplace safety standards ("occupational safety and health standards"), including standards aimed at reducing the threat of disease. Section 3(8) of the OSHAct provides:

The term "occupational safety and health standard" means a standard which requires conditions, or the adoption or use of one or more practices, means, methods, operations, or processes, reasonably necessary or appropriate to provide safe or healthful employment and places of employment.

A few things are important here. First, the authority here is quite broad, reaching just about  anything that could pose a health or safety risk in the workplace. OSHA standards are not just about exposures to chemicals, but cover all manner of potential workplace threats. Second, any standards set under the statute must be "reasonably necessary or appropriate." Third, as interpreted by the Supreme Court, this language (in conjunction with the provisions of Section 6) requires that OSHA conclude that the danger to be regulated poses a "significant risk" to employees.

The specifics of how and when OSHA may adopt standards under the OSHAct are provided in Section 6. The authority here is also broad for the adoption of permanent standards, which must go through the normal rulemaking process. That's a long and arduous process, which is why OSHA is relying upon special authority to adopt an "emergency temporary standard" (ETS) without going through notice-and-comment first, as it did earlier this year to set a new COVID standard for health care workers. (Of note, labor groups have sued OSHA over this ETS arguing that it is not stringent enough.)

Under Section 6(c), OSHA may bypass notice-and-comment to adopt an ETS upon making two findings:

(A) that employees are exposed to grave danger from exposure to substances or agents determined to be toxic or physically harmful or from new hazards, and

(B) that such emergency standard is necessary to protect employees from such danger.

This authority has been rarely used, and has been carefully scrutinized by the courts (as discussed in this CRS report).

Both Section 3(8) and 6(c) present potential challenges to the Biden Administration's plans, but not necessarily the challenges you have read about.

First, contrary to my co-blogger Ilya, I do not believe that it matters that the ETS is focused on a virus instead of a chemical or toxin. Section 6(c) allows OSHA to set an ETS when a grave danger is posed by "agents," among other things, and OSHA has long regulated "biological agents," including viruses. OSHA has standards and guidances that apply to exposures to various biological agents in laboratories and years ago adopted bloodborne pathogen standards that focus on controlling workplace exposure to Hepatitis B (including by requiring employers to make vaccination available at no cost to the employee). And even if a contagious disease were not considered a biological "agent," a newly emergent disease or disease variant could well be a "new danger."

Where OSHA will have to be careful in drafting and defending the ETS is in substantiating that COVID-19 (or newly emergent variants of it) pose a "grave danger" in the workplace, and that the ETS is "necessary" to limit that danger. Note that each of these requirements is more stringent than what is required for a generic standard under Section 3(8). There must be a "grave danger," not merely a significant risk, and the standard must be "necessary," as opposed to merely "reasonably necessary or appropriate." In other words, OSHA must clear a higher hurdle than is required for a regular workplace standard of the act.

One challenge for OSHA may be in demonstrating that the continuing spread of COVID-19 poses a grave threat to employees in covered workplaces. For starters, OSHA will likely have to focus on unvaccinated workers, because it would be hard to argue that COVID poses a "grave danger" to vaccinated employees. I expect it will also argue that the presence of unvaccinated employees is the source of that grave danger. (Note, however, that the risk to workers comes from their own behavior or from other workers is not a problem, as that's often the case with risks controlled by OSHA rules.)

While the White House is focused on the broader spread of COVID, and overall vaccination rates, these cannot be OSHA's focus. Nor can OSHA adopt an overbroad rule, roping in all large employers when the relevant risks are concentrated in a subset of firms. That there are some workplaces (say, a meat-packing plant) where there is a serious risk of spread, particularly if employees are unvaccinated, may justify an ETS that applies to those workplaces. It cannot justify an ETS applied to all workplaces generally. This is why I raised an eyebrow at the White House's focus on the number of employees in a firm (as opposed to the number of employees at a given workplace). If the eventual ETS is similarly broad, it could have a problem.

Not only must the ETS focus on a grave danger to employees at the workplace, it must also be "necessary" to reduce that risk. Here, too, I could see OSHA having some problems. As described by the White House, the ETS will require employers to mandate vaccination or conduct testing. But what if employees work remotely? What if they are not coming into contact with other, potentially unvaccinated, employees? Can it really be said that a workplace vaccinate-or-test requirement is "necessary" to control the risk of workplace spread to people who are not in the workplace? If the Administration wants to survive court challenge, it will likely have to consider this sort of alternative. And what about those who claim natural immunity from prior infection? OSHA will also have to address why applying this ETS to such employees is "necessary" to protect employees. I am not sure OSHA will have to address broader arguments against compulsory vaccination, however, as the plan is to include alternative means of compliance (e.g. periodic testing).

These sorts of questions (and more) are often addressed by agencies during the notice-and-comment process. Yet because OSHA is planning to adopt an ETS, there will be no such comment period until after the standard is in place. This means OSHA's lawyers have to flyspeck the ETS before it is issued, though they will have the opportunity to tweak the rule before it become permanent. Under Section 6(c), OSHA has six months to consider comments and develop a permanent standard to replace an ETS (but it is unclear whether the six month period may be extended, which could matter because six months would be lightning fast for a controversial notice-and-comment rulemaking).

Once the ETS is issued, it will certainly be challenged (as Josh has discussed). Even though some employers will welcome the standard so they can push their employees to get vaccinated while blaming the government, other employers (and employees) will raise concerns. (State governments, however, will not have standing to challenge the ETS as employers because the OSHAct's definition of covered employers does not cover "any State or political subdivision of a State.")

Some challenges will focus on the questions I've just outlined, but the particulars will have to wait until we see the actual rule. No doubt some others will challenge OSHA's authority to adopt such a rule at all, perhaps claiming that the OSHAct violates the nondelegation doctrine.

The OSHAct's broad language certainly raises nondelegation concerns. This concerned the Supreme Court in International Union v. API (the "Benzene case") and is the subject of this paper by Cass Sunstein. And were the OSHAct being challenged on a clean slate, it would not surprised me if several justices were willing to strike the statute down on such grounds, but a clean slate we do not have.

The Supreme Court considered and rejected the nondelegation argument against the OSHAct in the Benzene case. There the Court (re)interpreted the statute's language so as to constrain OSHA's discretion and thereby eliminate the nondelegation concern. This is important because, under principles of stare decisis, this prior interpretation of the OSHAct holds. So insofar as the Court interpreted the OSHAct to provide less open-ended discretion than the statutory text might suggest, statutory stare decisis would counsel the current Court to follow suit. Thus it would be a heavier lift to reject OSHA's standard-setting authority on nondelegation grounds.

UPDATE: Some have asked how this mandate differs from the CDC eviction moratorium, which was (properly, in my view) rejected by the Supreme Court. So here's a quick update on why I think the two measures present quite different legal questions.

First, and most importantly, the OSHAct provides clear authority to reduce workplace risks, where as the CDC statute gives no indication that it could be used to regulate landlord-tenant relationships. Among other things, this made the eviction moratorium particularly vulnerable to "major questions" objections, and gave force to the challenge that "if the CDC can do this, what can't it do?"

Second, OSHA has long-standing regulatory authority over workplace conditions, whereas the CDC was wading into new territory in seeking to regulate evictions. So blocking the CDC did not blunt the CDC's longstanding authority, whereas categorically rejecting OSHA's authority to address COVID in the workplace could undermine other OSHA measures.

Third, a properly drafted ETS will focus on COVID risks in the workplace, specifically the risks posed by and to unvaccinated workers who are in contact with other unvaccinated workers, and will be tailored to address those risks.  The eviction moratorium's relationship to the interstate spread of COVID, on the other hand, was quite attenuated, even in round two.

Fourth, under existing precedent, the federal government has broad authority to regulate the terms and conditions of employment as necessary and proper to the regulation of commerce among the states. Thus any challenge the OSHAct on Commerce Clause grounds would necessarily implicate a broader swath of federal authority. This was not necessarily the case with the eviction moratorium, as the Supreme Court has repeatedly said that the commerce power should be construed so as not to displace traditional state functions, and there is a colorable argument that landlord-tenant law (like land-use regulation generally) is a traditional state function with which the federal government cannot interfere (at least not without a clear statement from Congress to that effect).

The bottom line here is that a properly drafted OSHA ETS covering COVID exposure in the workplace should be significantly less vulnerable to legal challenge than the CDC's eviction moratorium.