Mandatory Vaccination Policies in the Non-Union Context

 At the outset, let me clarify and reiterate that I support vaccination and that, throughout the COVID-19 vaccine era, I have believed that employers have good reason to take measures to protect the workplace from COVID-19 - including through policies that require vaccination.

However, I've always been skeptical of whether they have a contractual entitlement to impose such a requirement as a term and condition of employment.

I noted in a recent entry that there's a question about the scope of an employer's policy-making powers in a non-union environment. In unionized environments, the rule is that an employer has the right to make any rule that is 'reasonable'. In non-unionized contexts, there was never a lot of consideration of this question - until the COVID vaccine cases started getting heard, and these cases seem to apply the 'reasonableness' approach without a lot of scrutiny as to its appropriateness.

In this post, I will explain the bases and operations of the 'reasonableness approach', versus a 'category approach', to understanding the scope of managerial prerogative in non-union environments; and I will argue that the reasonableness approach undermines certainty and fairness, conferring overly expansive rights upon non-union employers.

The Reasonableness Approach

This approach, as noted, is popular in unionized environments. Basically, unionized employers have a right to implement any policy that is reasonable and not otherwise inconsistent with terms of the collective agreement. This derives from an old Ontario decision frequently referred to as "KVP".

Within the labour relations context, this makes a fair bit of sense, because of the features of that regime: Firstly, they have relatively comprehensive express collective agreements, and don't apply various 'implied terms' said to arise at common law for individual contracts of employment. Secondly, employers are dealing with sophisticated entities of meaningful bargaining power (unions). Thirdly, collective agreements always include dispute resolution mechanisms that can effectively restrain employer overreach, creating an accessible process for determining which policies are 'reasonable', with due regard to the employer's reasons for imposing the policy and the impact on employees. So if the employer develops an oppressive and overreaching policy, the Union has a right to challenge that policy and get an Award striking it down and prohibiting the employer from relying upon it.

A typical 'management rights' clause in a collective agreement is a contractual residual clause: Anything not spoken to, management can dictate.

This approach dovetails with a labour relations rule of "obey now, grieve later": If you think an employer is breaching a collective agreement, the expectation isn't that you'll ignore the direction, but that you'll obey it in the interim and then pursue remedies under the collective agreement.

In a non-union context, things are very different. Most employment contracts are predominantly written to protect the employer's interests; they don't generally constrain employer action at all, and treating managerial prerogative as residual in the same way as a management rights clause in a collective agreement would give extraordinarily expansive powers to management.

You'll never know if a policy was 'reasonable' until you get through a wrongful dismissal action after being fired for breaching it - and you probably won't even get the employer's rationale for the policy until after the dismissal. There are typically no mechanisms to challenge the enforceability of such a policy during the life of the employment relationship. So the "obey now, grieve later" rule translates from unionized environments to non-unionized environments as simply "obey" - conferring the whole benefit of any uncertainty (and there's lots of it in this approach) on the employer.

However, common law doctrines - fully displaced by collective bargaining regimes - imply various terms into individual contracts of employment. All in all, these are vastly different regimes.

The Category Approach

The underlying theory of this approach is that the implied terms creating managerial prerogative are premised on the context of the employment relationship - that, in light of the specific nature of the employer and the role of the employee, parties would reasonably anticipate employers having the right to exercise authority over certain areas, and this assessment is to be made as at the time the contract was entered into - and not with an ex post facto perspective on whether a given policy or its application may be reasonable at the time it is invoked or applied.

Basically, I'm suggesting a relatively fixed framework of employer authority where the contractual rights of the parties don't organically expand or contract based on circumstances external to the relationship itself (subject to doctrines like frustration, etc.).

Let's start with the basics of what employers can do:

Within certain limits (defined by contract and constructive dismissal law), employers unquestionably have the power to dictate what work will be done, by whom, when, where, and how. Policies, rules, and directions governing the performance of work are straightforwardly within employer authority.

As well, rules relating to conduct in the workplace itself are straightforwardly within the authority of an employer. Respect in the workplace policies, safety policies, dress codes, etc. Again, within certain limits.

Also, people reasonably expect that an employer is able to require an employee to show up to work in a state fit to do the work in a way that accords with those reasonable policies. This creates a very limited inherent right to govern conduct outside the workplace. Again, this will vary by context, but we think it's pretty reasonable for an employer to set a policy prohibiting employees in safety-sensitive roles from showing up to work with alcohol or impairing drugs in their systems; requiring them to attend work sufficiently alert to not create a safety hazard; etc.

The "Don't come to work tired" implied rule is manifestly different depending on whether you're an accountant or a heavy equipment operator. As an accountant, the reasonable expectation is that you perform your work to a given standard, and it's disciplinable to fail to do so. Any responsibility to 'get enough sleep' is entirely derivative, and if you don't make any mistakes, your employer has no reason to object. However, if you're operating a crane on a construction site, then reasonable safety rules mean that showing up tired to work can be disciplinable misconduct even if you don't make any mistakes.

There are larger and more complex questions of what kinds of conduct, outside of working hours and outside the workplace, employers have a reasonable interest in regulating. These kinds of questions are fairly exceptional in the first place: Only certain types of employees, under certain types of circumstances, engage legitimate employer considerations in the first place. And I'd say that the proper analysis is a front-end one - one of reasonable expectations of the parties as to the rights of the employer - as opposed to an ex post facto one where we're looking at the whole picture afterwards to ask if an employer acted 'reasonably'.

I would argue that employers are always entitled to protect their brand, and entitled to regulate communications and actions associated with that brand. So if my employer's name is on my Twitter profile, my employer is probably entitled to set expectations for how I conduct myself on Twitter on that basis. More broadly, I'd say they're entitled to tell me not to hold myself out as representing them in public, and to otherwise set rules for communications that do represent the employer. But for most employees, this is a right to control use of the employer's name and trademarks, and not otherwise about limiting use of social media.

Suppose an assembly line worker for an auto manufacturer doesn't have anything about his employment on his social media profile, but he says something controversial and gets doxxed - so third parties connect his statements to the employer. I would argue that, regardless of how offensive or controversial those statements may be, under those circumstances the employer's legitimate interests are never engaged. They're free to distance themselves, say that the remarks don't represent them, say that they vehemently disagree with those remarks, etc., but I don't think they're entitled to broadly set rules for public communications by employees who simply cannot reasonably be taken as representing the organization.

For higher responsibility workers, the calculus changes. If an individual does represent the organization - to the general public, to customers, to business partners - then that opens the door to employers prohibiting communications and conduct inconsistent with that part of the role. Even then, it's still not a 'bright line' analysis. There's a difference between a salesperson who is the sole point of contact between an organization and various major customers, versus a customer service rep who answers phones to field miscellaneous customer queries.

The premise isn't that employers have some extensive freestanding right to regulate what employees do when they're not being paid; the premise is that the nature of the employment relationship is such that everyone can be taken to understand - at the front end - that the employer has a legitimate interest in regulating certain conduct outside the workplace.

It's similar, I think, to the 'key employee' test - a test that I have excoriated in its true context (as a test for who owes fiduciary obligations), but that makes more sense here: Employees at an executive level can't necessarily separate their own personal public image from that of the organization; and employees who are the 'face of the company' also can't step outside of that role - at least for people who know them as the face of the company.

So if I'm the company's official spokesperson, I probably can't maintain a personal non-anonymous social media presence at all without people looking at it as representative of the company. If I'm the main point of contact for a major client, then public conduct likely to alienate that client likely engages the employer's reasonable interests.

Put into a policy lens, my take is that appropriate policy language has to target conduct based on those factors: Instead of setting a policy that governs everyone the same way, set a policy that specifically targets communications that (a) will be seen as representing the company and/or that (b) will reflect poorly upon the company or damage its relationships with stakeholders.

So, boiling these categories down to their essence, I'd suggest that there are basically four categories of implied management authority: Near-absolute authority over the performance of work; near-absolute authority over personal conduct in and pertaining to the workplace; near-absolute authority over communications using the name and/or intellectual property of the employer; and limited authority over outside conduct for specific classes of employees.

Contractual Limits on Managerial Prerogative

One thing we do know is that recognized express or implied contractual terms fetter managerial powers. Even outside of my suggestion that the prerogative is finite in the first place, it can be restrained by contract (so if my employment contract specifies my hours of work as, for example, 9-5, without reserving employer authority to change them, then I would argue that the employer has ceded any implied prerogative to change those hours).

The problem is that the employment contract is rife with implied contractual terms, and most of them aren't well defined or studied. A great deal of the average individual contract of employment is captured by 'implied' terms (differing from unionized employments) and it's difficult to coherently set limits on managerial authority as being within the confines of those implied terms because even employment law experts couldn't come to a consensus on what a comprehensive list of those implied terms would look like. (I wrote a paper just a few years ago about the most discussed implied term, because the Alberta Court of Appeal had come to a different fundamental view from the rest of the country as to what that implied term actually said.)

Among other things, we know that, absent an express term to the contrary, there's an implied term of reasonable notice. Under most circumstances, there's an implied rule against unpaid layoffs and suspensions.

Just within those categories, we can divine a couple of broad propositions - that employees have a contractual right to work (or at least to be paid for) their usual hours without interruption. On that basis, I rule out any employer right to impose an unpaid leave of absence, absent express contractual terms conferring such a right. For any reason at all, short of just cause for dismissal.

There's a good argument to be made that paid leaves can be imposed in certain cases. But not unpaid leaves. This gets important later on.

If you'd asked me this pre-COVID, I'd have said the same thing: An employee can choose to take unpaid time off because they're sick, but if an employer requires employees to stay home while sick (and there have always been reasonable bases to do so), the time has to be paid.

Jurisprudential History of Managerial Prerogative

In 1997, the New Brunswick Court of Appeal in MacNaughton articulated a test for just cause for insubordination as including breach of "lawful and reasonable" policies or orders. This articulation has been picked up a few times, including in the Echlin and Certosimo text on Summary Dismissal, and a handful of times in the case law.

So, full disclosure, people who say that the 'reasonableness' approach applies in non-union contexts have a basis. In fact, the case might be a little more persuasive than most NBCA decisions because this decision was authored by Michel Bastarache, who was appointed to the SCC a few months later.

However, if we dig down a little bit on that articulation, we realize that MacNaughton wasn't a case about delineating the scope of an employer's authority: On its facts, Mr. MacNaughton was dismissed for giving a demanding customer the benefit of two discounts (instead of just one), in breach of Sears' pricing policies. It's uncontroversial that Sears has the right to set and enforce pricing policies, but the case itself turned on the severity of the misconduct - and on those facts, Justice Bastarache concluded that the mitigating factors were sufficient to overturn a finding of just cause. (Even if there was analysis about what makes a policy enforceable, it would be obiter.)

By contrast, there's an earlier articulation of the test for insubordination in Alberta, by Justice Veit in Amos, that, in addition to requiring a work order to be "lawful and reasonable", also requires that the order be "authorized, that is, that it came within the scope of the worker's duties." (In my view, this has to be right, in a general sense: An employer's control over an employee derives from contract, and it would be unduly broad to express this implied term in a way that allows employers to expansively regulate employees' conduct outside the scope of anything connected to the job or workplace, on their own time when they're not being paid. Most hourly-paid employees expect that their employer's authority over them ends when they clock out, and this is - for the most part - a reasonable expectation.)

While some of the (very limited) case law on the subject treats these two articulations as being roughly parallel, I would suggest that the requirement that an order be 'authorized' in the sense used by Justice Veit puts her analysis squarely into what I've described as a 'category' approach here, that even a 'lawful and reasonable' directive cannot be enforced if it falls outside of the contractual authority of the employer.

But the real problem is that where the "lawful and reasonable" standard is articulated, it's often dealt with in a fairly cursory fashion because it's simply not what the case turns on. For example, in the 2005 ABQB case of Holwen, the Court quoted Justice Veit's list of factors in its entirety from Amos; however, in 2007, in Dawson, Justice Langston referred to Holwen, and inclusively quoted 8 of the 10 factors from Justice Veit's list, one of the two omissions being "authorized", and stating as a broader proposition that "Willful disobedience to a lawful and reasonable policy or order may be sufficient [to constitute just cause]".

Since then, Dawson has been cited a handful of times as authority for a 'lawful and reasonable' proposition. Still, pre-COVID, in the non-union context, it was nearly unheard of for a case to turn on the scope of managerial prerogative at all.

I'd suggest, then, that the case law is mixed, and quite confused, on whether managerial prerogative extends to anything reasonable that's not contractually excluded, or if managerial prerogative has its own implicit limitations in scope.

COVID Cases

We've started to develop a body of case law surrounding an employer's rule-making authority for vaccines, masks, and leaves of absence in the COVID era.

The leading case appears to be a BC decision, Parmar: In that decision, the Court looked indepth at the arbitral case law applying KVP, and noted that those treatments were useful but not binding, and went on to conclude that a policy that required mandatory vaccination, failing which an unpaid leave of absence would be applied, was reasonable and enforceable.

I have several concerns with this decision: First and foremost, the court explicitly moves away from treating the 'unpaid leave' as disciplinary, instead treating the leave as simply an application of the policy. That, alone, makes the decision - to my mind - categorically wrong. The court was inattentive to the very significant case law in various jurisdictions that have found an implied contractual term against unpaid layoffs, leaves, and suspensions - the exception being Carscallen where Justice Echlin contemplated (in persuasive obiter) that an unpaid disciplinary suspension might be justifiable where just cause for dismissal exists.

Finding, as the BCSC did, that an employer has a right to create policies that dictate circumstances for unpaid leaves of absence is deeply problematic, from an analytical perspective.

If a mandatory vaccination policy were otherwise valid, however, you could get into the analysis from the other direction by treating it as disciplinary, derived from insubordination against a valid policy, and arguing that the insubordination amounts to just cause. (Not the line of analysis used in Parmar, though.)

This brings me to my second objection: That while the court recognized that arbitral determinations on whether vaccine policies were reasonable weren't binding, it failed to consider whether the test applied in unionized settings was at all appropriate in a non-unionized context: In other words, the court assumed, without examining, that the underlying question here was "Is the policy reasonable?" The Court asserted, without any analysis or citation of any line of authority, that the employer's "right to implement policies was only subject to the implied qualification that any such policy would be reasonable and lawful."

If we're really asking - as I am in this article - whether a non-unionized employer has a right to impose absolutely any requirement so long as it's 'reasonable', Parmar is an absolutely unpersuasive response to that question.

In Benke, Justice Feasby considered a similar fact-pattern derived from a 'mask' policy - an employee refused to mask, and was placed on an unpaid leave. The conclusion and analysis was analogous to Parmar, though I'm a bit less critical of this one because of its differences: Firstly, a mask policy is far more closely analogous to a dress code; it's a much more direct assertion of authority over the workplace itself. Secondly, while the policy in Parmar was found not to be legally mandatory, mask policies at issue in Benke were required by a patchwork of municipal bylaws and Provincial legislation - which would create a frustration event if Benke's interpretation of the employment contract were accepted.

And finally, Justice Feasby's analysis echoes the Carscallen framework (though without citing it): The employee's refusal to comply with the mask policy amounted to a "repudiation" of the contract, and the employer did not accept that repudiation, applying the unpaid leave instead.

Bill Johnson, Vice-Chair of the ALRB, recently rejected a constructive dismissal claim in a context that parallels Parmar. In doing so, he quoted, and bolded, the statement in Parmar about the only constraints on policy-making power being that it be reasonable and lawful. This is an incredibly expansive view of employer power.

Analysis

I don't question the application of the KVP rule to the unionized context. Analytically, that's a well-established approach.

But my perspective is that the same principles don't obviously hold up when interpreting the individual contract of employment, and that the case law applying a KVP-type test to non-union employers' vaccine policies ignores important principles, such as Justice Veit's requirement that policies be 'authorized'.

And while, at the end of the day, I'm not all that sympathetic to these anti-vaxxers losing their claims, I'm nonetheless concerned that this is a "bad facts make bad law" kind of situation, where the analysis we're adopting in order to achieve this result confers nearly-limitless power upon employers who already benefit from immense power over their employees.

If employment policies can assert authority over the individual, outside the workplace and working hours (and in some cases they can), and 'reasonableness' is the only limit, with no mechanism for challenging reasonableness except for dismissal remedies after it's enforced, then it's easy to imagine employers exploiting their power and uncertainty: Can a television media company prohibit employees from watching shows produced by their competitors? Prohibit employees from purchasing products advertised on competing networks? Can a law firm that represents large industrial and institutional clients tell its support staff that they're not allowed to donate to, or express public support for, politicians who oppose the interests of those clients? (In Provinces - including Alberta and Ontario - where 'political belief' isn't a prohibited ground of discrimination, this policy would be 'lawful', so the question becomes whether it's 'reasonable' - which, while I might have my opinions, isn't a clearly one-sided question.)

As I recently discussed, there are revolving issues surrounding BYOD workplaces - where employers allow (or require) employees to use their own personal electronic devices to carry out work functions. This is a cost saver for the employer, but it also impacts their ability to control the security of their electronic documents, and you sometimes run into problems with employers wanting to exercise control over the employee's device and personal files at the end of the employment relationship. If it's my electronic device, then I'm allowed to use it for other personal purposes which ultimately intermingle my personal electronic files with the employer's. Or am I?

With a vague 'lawful and reasonable' approach, and not a lot of case law governing it, there's at least an argument that an employer is automatically entitled to tell employees (a) that they have to have their own personal smartphone which they will use for work purposes; (b) that they not use that device for specified purposes; and (c) that the employer is entitled to control, access, or even delete any files - including personal files - on the personal device.

These types of policies are uncontroversial for an employer to implement in respect of its own devices, because it owns the devices and is entitled to assert control over them. If I download a personal document onto my employer's computer, I can't be too shocked if the employer ends up getting access to that document. There can be privacy concerns raised in this scenario, but the reality is that it's not my device.

But when I'm using my own personal device for work, I have an absolute in rem proprietary right to control that personal device and my own electronic information on that device, and the 'lawful and reasonable' test would suggest that employment contracts implicitly relinquish the employee's property rights - allowing the employer to inspect, seize, and reformat my device so long as they can say it's 'reasonable'.

Does Vaccination Generally Fall Into An Established Category?

One might argue that vaccination in the workplace is an extension of the 'control over the workplace' principle. But that's sufficiently different from "Don't drink before your shift" that I don't think the analogy is helpful. It's certainly not the case that "This is a rule of the condition you must be in at the workplace" creates an absolute entitlement to require somebody to do something outside the context of work. (For an absurd example, imagine if I insisted that my employees tattoo my brand to their foreheads. I'm entitled to insist on my employees wearing branded headbands at work, but insisting they tattoo it to themselves would clearly be out of line.)

If we're framing this as a matter of 'reasonable expectations of the parties', then I don't think that, when I started my current job in 2018, anyone would say that the parties reasonably anticipated that my (non-healthcare) employer might dictate what vaccines I should receive. This might be different in some industries, however.

I've been consistent on this position for years now: I believe that there are/were many circumstances where employers ought to remove anti-vaxxers from the workplace, in support of their obligations to maintain a safe and healthy workplace, but I don't think that this automatically creates a contractual entitlement to dismiss employees 'for cause' for refusing to vaccinate.

*****

Dennis Buchanan is a lawyer practicing labour and employment law and civil litigation in Edmonton, Alberta.

This post does not contain legal advice, but only general legal information.  It does not create a solicitor-client relationship with any readers.  If you have a legal issue or potential issue, please consult a lawyer.