Bona fide occupational requirements

The ethical and legal justification for vaccination mandates commonly begins with the argument that, for some jobs, vaccination constitutes a bona fide occupational requirement3; that is, something every individual performing a job must do or have done (with reasonable exceptions) because it is considered necessary for the effective and safe performance of that job (ie, safe for all those who may be affected within or by an occupational environment, including employees and the public).4 5

Now, if one is unconvinced that vaccination can ever be a bona fide occupational requirement, they are unlikely to think it is ethically acceptable to not hire or fire people on the basis of their vaccination status. Hence, to examine the question at hand, we must assume vaccination can at least in some instances be plausibly considered a bona fide occupational requirement. And because the objection to vaccination mandates grounded in the harms experienced by existing employees who are unwilling to be vaccinated does not necessarily entail the claim that vaccination is unnecessary for the effective and safe performance of certain jobs, we can assume some making this objection will agree vaccination can represent a bona fide occupational requirement in at least some cases.

A bona fide occupational requirement is a requirement for new employees as it is for veteran employees. It is connected to the functions of the position, not the length of time an employee has been employed. Nevertheless, a unique case exists when a bona fide occupational requirement is established only after an employee has been hired. In such cases, one’s intuition may be that it is reasonable and fair to expect prospective employees to meet the bona fide occupational requirement but unreasonable and unfair to expect employees whose employment predates the establishment of the requirement to do so. But this does not stand up to scrutiny.

To see why, it is helpful to consider other sorts of occupational requirements beyond vaccination. Consider the case of vision standards for drivers of commercial vehicles, first introduced in the USA in 1939.6 The purpose of such standards was to identify a level of vision that must be met in order that commercial drivers will not present a safety risk to themselves or the public. These standards did not exist when some drivers of commercial vehicles were initially hired. Should such drivers have been exempted from the standards, allowing them to continue driving even if unable to meet new vision standards? No, because doing so would suggest it is unacceptable for prospective, but not existing, commercial drivers to present a safety risk to themselves or the public, and there are no reasonable grounds to think this is the case. Moreover, it undermines the aim of the standards, that is, to reduce the safety risk to drivers and the public.

Consider another example, that of criminal record checks for teachers, which first became mandatory in the Canadian province of Ontario in 2001.7 Similar to the example of visual standards for commercial drivers, it would be incoherent to agree such standards are necessary for health and safety but exempt teachers whose employment predates 2001 from meeting them. Doing so would suggest the risks associated with existing teachers having a criminal record are acceptable but the risks of prospective teachers having a criminal record are not.

By the same token, if one believes vaccination requirements are necessary for the effective and safe performance of a job, like vision standards for commercial driving or criminal record checks for teachers, so much so that it is justifiable to introduce them for prospective employees, then one should consider it justifiable to introduce them for existing employees as well. Existing employees should not be ‘grandfathered’ (ie, exempted) out of a policy since the health and safety justification for the policy is no different for them. That existing employees may be fired, for example, for failing to meet the requirement does not affect the necessity or rationale of the requirement. If conditions for the effective and safe performance of a job change—for example, emerging evidence suggests asbestos is harmful to human health, new and improved safety helmets are invented or a pandemic of a novel virus occurs—it is reasonable for new health and safety conditions for employment to be established in response (indeed, employers may have duties to introduce such conditions in response), and the justification for meeting those conditions should be the same for both existing employees and prospective employees. Hence, if the conditions for the effective and safe performance of a job change, it is not unfair nor unreasonable to expect existing employees to meet those conditions (with reasonable exceptions), despite the fact they were not included in their original employment agreements. The upshot is that the locus of scrutiny should continue to be on whether vaccination constitutes a bona fide occupational requirement. If it is, then it is difficult to see why this requirement should not apply equally to prospective and existing employees.

Asymmetry in consequences

Now, one may agree the health and safety justification for vaccination requirements is the same for all relevant existing and prospective employees but still insist such groups ought to be treated differently due to the asymmetry in consequences members of the two groups will experience should they fail to meet those requirements. In other words, one may agree there are no reasonable grounds to think it is acceptable for one group but not the other to present a safety risk to fellow employees or the public (eg, by not adhering to vision standards, criminal record check requirements or vaccination requirements), but nonetheless argue the consequences resulting from non-compliance with relevant employment conditions outweigh those risks in the case of existing employees but not prospective employees.

This argument is plausible if analysed in terms of costs and benefits, which may correspond to one way policymakers seek to justify vaccination mandates, for example, by showing their benefits outweigh their costs. Though, this involves a tricky empirical question that turns on whether the harms (however defined) experienced by existing employees who are unwilling to be vaccinated (eg, being fired) ‘outweigh’ the harms (however defined) that could be expected to occur in the absence of a vaccination requirement (eg, infections, illness, hospitalisations, deaths and related sequelae).

But proceeding in these terms ignores another way in which vaccination mandates may be justified that does not simply rest on a favourable cost-benefit ratio. Health and safety employment standards may be justified when they are seen as necessary for the effective and safe performance of a job, not only when the harms averted by the safety standard outweigh the harms that may be experienced by employees unwilling to meet them. One straightforward reason that may be supplied for why this is the case is because proceeding otherwise would be to ignore any legal and ethical duties employers and/or employees have to protect the health and safety of other employees and the public. Another reason is that some harms, like preventable nosocomial infections in high-risk settings, may be considered ethically unacceptable and thus worthy of taking steps to prevent even if those steps involve significant costs. In such cases, failing to meet what are considered to be health and safety employment standards could be considered disqualifying irrespective of the harms visited on those unwilling to meet them.

Again, an example may be illustrative. Consider Jurisdiction A, which has a criminal record check requirement for teachers, and Jurisdiction B, which does not. Imagine a school year where no children are harmed in any way at the hands of teachers in either jurisdiction, but where 10 teachers in Jurisdiction A are fired as a result of failing to comply with its criminal record check requirement. Putting aside challenges related to counterfactuals (ie, it is difficult to know whether the requirement in Jurisdiction A was in fact responsible for the lack of harm to children experienced in that jurisdiction), an argument resting strictly on a favourable cost-benefit ratio may suggest the harms associated with 10 teachers losing their jobs outweigh the harms averted by the criminal record check requirement. Jurisdiction B does not have such a requirement and their outcomes related to child welfare were the same as Jurisdiction A, and so this represents a more favourable cost-benefit ratio. But this ignores the obligations society has to protect the welfare of children (discharged in this case via mandated criminal record checks for those interacting closely and frequently with children). Failing to meet such an employment standard could be considered disqualifying irrespective of the harms visited on those unwilling to meet it.

Hence, that the consequences may be particularly severe for existing employees should they fail to meet newly introduced employment conditions, or that such harms are seen to outweigh possible benefits, does not tell us whether those consequences are necessarily unjustified. Firing a 30-year veteran of a job—someone who has a mortgage and several kids to feed—is something we should try to avoid if we can, and so the pertinent question is whether it is truly necessary to fire that person based on their vaccination status; whether vaccination is truly necessary for the effective and safe performance of her job. Or, put in another way: whether vaccination should be considered a bona fide occupational requirement. If one were to agree vaccination constitutes a bona fide occupational requirement, it would therefore be inconsistent to argue vaccination mandates are unjustified due to the harms experienced by existing employees who are unwilling to be vaccinated. Indeed, the harms experienced by existing employees unwilling to be vaccinated are acceptable precisely in cases where vaccination is considered a bona fide occupational requirement. Taking a condition to be a bona fide occupational requirement is, in effect, taking it to be settled that the consequences for employees who choose to not meet the employment condition should not outweigh the consequences expected to occur in the absence of that condition or because of some ethical and legal duties employers and/or employees have to protect the health and safety of other employees and the public. In fact, the process for establishing a bona fide occupational requirement does not generally even consider the consequences for employees who choose to not meet them.5 Conversely, in cases where vaccination is not considered a bona fide occupational requirement, the ratio of costs to benefits is likely to have greater moral salience in evaluating the prospect of a vaccination requirement.

Lest one thinks the conclusion to this particular argument begs the question, it is helpful to emphasise the distinct insights it generates. First, it tells us that the harms existing employees may experience as a result of being unwilling to get vaccinated cannot on their own be dispositive that it is wrong to require them to be vaccinated. We should therefore be sceptical of arguments that simply count the harms experienced by existing employees who are unwilling to be vaccinated as a sufficient reason to object to vaccination mandates. Second, it tells us that it would be inconsistent to agree an employment condition counts as a bona fide occupational requirement but argue the consequences for employees who are unwilling to meet that requirement are so severe that we should not expect existing employees to meet it.

The nature of employment conditions

A final way in which the introduction of new employment conditions is morally distinct for existing employees relative to prospective employees is that existing employees are in a much more difficult position to not agree to such conditions. If Company A has a vaccination requirement and Companies B through Z do not, prospective employees who are not vaccinated and are unwilling to get vaccinated can apply to work at Companies B through Z. But existing employees of Company A are in a more difficult situation, assuming the vaccination requirement was introduced during their tenure. For existing employees, their choice is between employment and unemployment (in the case where the consequence of failing to comply with the vaccination mandate is being fired). It is a choice between maintaining their income—which they likely expected to maintain in their financial planning—and losing their income. While they ultimately have a choice in the matter, the consequences of that choice for existing employees may make them feel like they have little choice at all.

But this is not terribly instructive since it is true of many employment conditions. For instance, if Company A requires employees to wear safety helmets and Companies B through Z do not, prospective employees can take this into account when choosing where to apply for work. If they are unwilling to wear safety helmets, they can apply to work at Companies B through Z. By contrast, if Company A has a compelling reason to change its policy on personal protective equipment and begin requiring employees to don safety helmets, existing employees of Company A may feel they have little choice but to comply with this change, for their choice is between employment and unemployment; between maintaining their income and losing it (assuming, again, that the consequence of failing to comply with the change is being fired). But this asymmetry in consequences between prospective and existing employees does not suggest that only prospective employees should be required to wear safety helmets. Nor does the severity of consequences (ie, being fired) for existing employees unwilling to meet those new conditions suggest the shift to wearing safety helmets can never be justified (with the possible exception of cases where employment contracts are explicit about such conditions and changes to them, which is less likely to be the case for possible future vaccination requirements for novel emerging pathogens).