Parts of the country have begun the process of returning to work, in places where COVID-19 infection rates have flattened or shown a decline. But the risk of becoming infected with COVID-19 remains, and some employers may be faced with parts of their workforces refusing to return to work or to perform certain assignments, citing the health risk. What are employers’ options with respect to such employees? There are both legal and practical considerations.

From an OSHA Perspective

Employees may claim that working in an environment where COVID-19 is a potential hazard remains a health risk so dangerous that they have a right to refuse to work. The Occupational Safety and Health (OSH) Act protects employees from retaliation in certain circumstances when they refuse to perform work in “imminent danger” situations. According to Occupational Safety and Health Administration (OSHA) guidance, an employee may refuse an assignment that involves “a risk of death or serious physical harm” if all of the following conditions apply: (1) the employee “asked the employer to eliminate the danger, and the employer failed to do so”; (2) the employee “refused to work in ‘good faith’” (a genuine belief that “an imminent danger exists”); (3) “[a] reasonable person would agree that there is real danger of death or serious injury”; and (4) “[t]here isn’t enough time, due to the urgency of the hazard, to get it corrected through regular enforcement channels, such as requesting an OSHA inspection.”

It is not clear whether COVID-19 rises to the level of an “imminent danger.” As Secretary of Labor Eugene Scalia recently noted, “Coronavirus is a hazard in the workplace. But it is not unique to the workplace or (with the exception of certain industries, like health care) caused by work tasks themselves. This by no means lessens the need for employers to address the virus. But it means that the virus cannot be viewed in the same way as other workplace hazards.”

No one can guarantee health for the public—much less employees—in the middle of a global pandemic. No court or administrative panel has ruled that COVID-19 is an “imminent danger.” In another context, a federal district court declined to grant an injunction to pork plant workers asserting that their plant was inadequately protecting its workforce from COVID-19. The apprehension of contracting COVID-19 was not an actual harm, the court ruled, and “in this time, no essential-business employer can completely eliminate the risk that COVID-19 will spread to its employees through the workplace.”

Even if COVID-19 is deemed an “imminent danger,” it is clear that employers do not have to pay employees if they refuse to work. In a 1980 case, Whirlpool Corp. v. Marshall, the Supreme Court of the United States held that the OSH Act did not require employers to pay employees who refused to work for safety-related concerns. The Court noted that “Congress very clearly meant to reject a law unconditionally imposing upon employers an obligation to continue to pay their employees their regular paychecks when they absented themselves from work for reasons of safety.”

From an NLRA Perspective

Employees’ Right to Refuse to Return to Work

Whether an employee has a protected right to refuse to work, or refuse to return to work, because of safety concerns related to COVID-19 will turn, in most cases, on whether the conduct constitutes protected concerted activity under the National Labor Relations Act (NLRA). The National Labor Relations Board (NLRB) has recognized that an employee engages in protected concerted activity when he or she takes action “with or on behalf of other employees” concerning the terms and conditions of their employment. This typically involves group activity (i.e., two or more employees), a single employee seeking to incite group action, or an individual employee “bring[ing] truly group complaints to the attention of management.” By contrast, actions that an individual takes without the actual or inferred support of other employees, and actions that relate only to an individual’s personal interests, are generally not concerted and thus not protected. Additionally, even concerted employee actions may not be protected where they relate to an issue over which the employer lacks control.

This right is available to almost all nonsupervisory or nonmanagerial employees who work for a private-sector employer. The concept of protected concerted activity applies to employees who are represented by a union, but also employees who are not represented by a union. For example, the NLRB has held that a concerted work stoppage by employees who are not represented by a union constitutes protected concerted activity where it is based on a group concern relating to workplace health and safety issues. To be protected, employees need only have an “honest belief” that working under certain conditions would not be safe or healthy. Such activity has been found to be protected even where the employer has deemed the disputed conditions to be safe, and even where the employees’ concerns may not be objectively “reasonable.”

Especially during the COVID-19 pandemic, the mere fact that employees are engaged in collective action relating to workplace safety concerns can be sufficient for their activity to be found protected, so long as the employees do not engage in conduct that could remove them from the protections of the NLRA—i.e., as the Supreme Court held in a 1962 case, conduct that is “unlawful, violent, in breach of contract, or otherwise indefensible.” The fact that the employer believes sufficient measures have been taken to address safety concerns is not generally sufficient to remove employees’ conduct from the protections of the NLRA. Additionally, the fact that the workplace may be deemed safe by government agencies (or that employees could seek assistance from government agencies) is similarly not sufficient to render employees’ conduct unprotected. The right to engage in protected concerted activity over safety and health issues under the NLRA is “distinct from and … not subordinate to the provisions of the [OSH Act].”

Notably, employees are not required to identify or demand that the employer address safety issues before they can strike or take other job actions in order for their concerted conduct relating to workplace safety issues to be found protected.

Refusals to Work May Be Sanctioned Notwithstanding a “No Strike” Provision

Is it important for employers to understand that unionized employees potentially may also refuse to work due to the employees’ fear of becoming infected with COVID-19, even if the applicable collective bargaining agreement contains a “no strike” provision.

Pursuant to Section 502 of the NLRA and Supreme Court precedent, a refusal to work over safety concerns is protected if the assignment is “abnormally dangerous.” Unionized employees must have a “good faith belief” supported by “ascertainable” and “objective evidence” that there is an “abnormally dangerous” working condition in order to be absolved of their contractual obligation not to strike. Refusal to work in this context is generally protected, even if there is a no-strike clause in the relevant collective bargaining agreement.

In Gateway Coal Co. v. United Mine Workers of America, 414 U.S. 368 (1974), the Supreme Court held that an “abnormally dangerous” working condition is one that presents “some identifiable, presently existing threat to the employees’ safety” when judged by ascertainable, objective evidence. In TNS, Inc., 329 NLRB 602 (1999), the NLRB found conditions to be abnormal in cases where “risks that are ordinarily present have been intensified.”

To show “good faith belief,” employees must do more than simply claim a condition is abnormally dangerous. According to the NLRB in TNS, Inc., “A purely subjective impression of danger will not suffice; nor will a speculative doubt about safety in general.” The belief must be objective and quantifiable. The NLRB requires demonstration by a preponderance of the evidence that “the employees believed in good-faith that their working conditions were abnormally dangerous; that their belief was a contributing cause of the work stoppage; that the employees’ belief is supported by ascertainable, objective evidence; and that the perceived danger posed an immediate threat of harm to employee health or safety.”

The burden of demonstrating the application of Section 502, and that an abnormally dangerous working condition excuses an otherwise clear violation of a no-strike provision, rests with the party asserting that such a condition exists. Merely stating that the condition exists, or arguably having a general fear of a potential condition, would be insufficient to carry this burden. Ascertainable, objective evidence must be presented before a union or employee may seek the safe harbor of Section 502.

The clear language of the statute compels the union or employees to come forward with evidence establishing the conditions at issue are abnormally dangerous and not just that hazards exist at a particular work site. Such evidence might include conditions deviating from the norm or from a reasonable level of risk; safety equipment operating improperly or not at all; significant deviation from industry safety standards; OSH Act violations; or an employer’s failure to provide sufficient safety instructions. In the current climate, “ascertainable” and “objective” may include violations of or a failure to conform to federal or state directives relating to the COVID-19 pandemic. The argument could be made that an employer’s compliance with COVID-19-related directives, federal guidelines, and/or recommendations from the U.S. Centers for Disease Control and Prevention (CDC) and World Health Organization (WHO) undermines a union’s claim that the work environment is abnormally dangerous.

Different Standards Applied

While there is a substantial degree of overlap between Section 7 (protected concerted activity) and Section 502 (work stoppages due to abnormally dangerous conditions) of the NLRA, there are also notable differences. First, a work stoppage protected under Section 7 requires the element of “concertedness,” where two or more people are acting together or one person is acting on the behalf of others. Section 502 additionally, and expressly, covers stoppages by a single employee without the requirement of “concerted” activity. Second, under Section 502, there appears to be a higher bar for establishing the employees’ perception of possible harm—not only must there be good faith, but the harm must also be objectively demonstrable. The case law under Section 7 sets a much lower bar. In those situations, the employees’ belief does not need to be objectively reasonable. Lastly, and perhaps most significantly, the case law under Section 7 suggests that the source of the employees’ concern must relate to a condition over which the employer has control in order for the stoppage to be protected. Section 502 appears to be devoid of any such requirement. As long as the predicate condition exists, it appears immaterial as to whether it is within the employer’s control.

For healthcare employers there are at least two additional issues to consider. First, any analysis under either Section 7 or Section 502 involves a determination as to whether the working conditions at issue are “abnormally dangerous” or, at the very least, outside the norm. This raises particularly difficult fact questions in the healthcare setting where, for at least some employees, exposure to all kinds of pathogens is a virtual constant working condition that has not markedly changed as a result of COVID-19. Second, unlike employees in other industries, healthcare workers (at least when directed to do so by a labor organization) are prohibited by Section 8(g) from engaging in a strike or concerted refusal to work without providing their employer with 10 days’ advance notice. This requirement is in complete tension with the provisions of Section 502 that clearly contemplate a spontaneous refusal with no notice.

There does not appear to be any direct case law on point. On the one hand, it could be argued that the notice requirement does not apply to actions protected by Section 502, since such a requirement could largely repeal that provision. On the other hand, it could be argued that because the 8(g) notice requirement was enacted after the passage of Section 502 that the notice requirement in the healthcare setting applies even to stoppages under that section. The most prudent course of action for a healthcare employer may be to treat employees under such circumstances as immune from discharge because the opposite course could expose employers to substantial liability. For employees in this position, the fear of illness or death will almost certainly trump the fear of adverse employment consequences.

Employers’ Responses to Protected Refusals to Work

Taking any disciplinary action (including the issuance of attendance points and progressive discipline under its attendance policy) in response to protected concerted activity would likely be found unlawful. While an employer may not discharge striking employees, it may temporarily or permanently replace employees who engage in a protected work stoppage in order to continue its operations. One caveat: if the basis of the strike includes allegations that the employer violated the NLRA in some way (also referred to as an “unfair labor practice strike”), striking employees may only be temporarily replaced.

While an employer does not have to pay employees who refuse to work (strike), it cannot withhold accrued benefits from employees based on their participation in the strike. In addition, under the Affordable Care Act, it is generally recommended that benefits continue during a limited strike in order to avoid any potential penalties. To the extent employees are still working and benefits are still available, striking employees should continue to receive healthcare coverage. However, employers can require that employees pay their portion of the premiums (including providing them with information about how to make such payments). An employer does not have to allow employees to use vacation or paid time off (PTO) for days spent on strike; however, employees should be able to utilize vacation or PTO if they are absent for other reasons. If an employer chooses to allow employees to use vacation or PTO to deplete their leave banks while on strike, it can do so.

If employees are engaged in a protected “safety strike” under Section 502, they cannot be permanently replaced and must be returned to work once the strike is over. The employer can, however, use temporary replacements, including supervisors, employees from other locations, or temporary employees, to complete work while the employees are on strike.


Which training method is of interest to you?


Which training method is of interest to you?

Skip to content