Voiding Employment Contract Terms Because of Hypothetical Statutory Conflicts

By default, most employees dismissed on a not-for-cause basis are entitled to 'reasonable notice', by virtue of an 'implied' term of the contract. There's a common law doctrine for figuring out how much that is, but in many cases it can be substantial. So many employers try to limit those employee entitlements by expressly contracting to something different (and usually significantly less) instead.

That's allowed. It works. If done properly. (Some management counsel will gripe about how shifting goalposts make it nigh-impossible...which has a grain of truth, but there are ways of dealing with that, too.)

However, employment standards legislation sets out guaranteed minimum termination entitlements, and contracting to less than that is not allowed. Across the country, all the employment standards statutes I've reviewed have similar provisions voiding such contractual clauses. (That's also why you can't agree to a wage lower than minimum wage; you can't sign a contract waiving statutory overtime, etc.)

The guaranteed statutory minimum termination entitlements are generally 'formulas' keyed to length of service. In Alberta, for instance, entitlements start at 1 week after 3 months of service, and cap out at 8 weeks after 10 years of service.

In 1992, in a case called Machtinger, the Supreme Court of Canada dealt with a scenario where two employees had contracts saying they could be terminated on, respectively, zero and two weeks' notice, but they'd been employed for long enough to have a statutory minimum entitlement of 4 weeks. They were actually paid their stat minimums, but argued they were entitled to 'reasonable' notice (which was 7+ months in that case).

The SCC awarded it to them: The contractual provision was void, and therefore left intact the implied entitlement to reasonable notice of termination. So, ever since, that's been an easy question: A contract saying you get two weeks notice, if you're entitled to more than two weeks' notice, is void.

But the more interesting question is this: If a contract says you get two weeks' notice, and you get dismissed at a time when that would be enough to satisfy the statute, does the contract stand?

The courts in Ontario and BC have given clear 'no' answers to that question. In Alberta, it remains less clear.

British Columbia: Shore v. Ladner Downs

B.C. was ahead of the curve on this issue, with a fact pattern that arose even before the decision in Machtinger was released. Douglas Shore had a contract entitling him to 30 days notice of termination. When he was dismissed less than a year later, his statutory entitlements were only two weeks, and he was paid four weeks.

In 1998, the BC Court of Appeal found that Shore was entitled to reasonable notice: Though the facts were somewhat distinguishable from Machtinger, the SCC's rationale for voiding the contractual term wasn't changed by the mere fact that the contractual term happened to provide an amount that would have satisfied statutory entitlements at the time the contract was actually terminated.

The policy considerations applied in Machtinger, supra, would not be served if the contract were to be interpreted in favour of the employer so as to leave the individual employee responsible for determining, at the point of termination, whether the statutory minimum had risen above the notice period stated in the contract.  It is neither reasonable nor practical to leave the individual employee in the position of having to keep an eye on the relationship between the statutory minimum and the contractual term.

Ever since then, this proposition has been basically uncontroversial in British Columbia.

Ontario: Wunderman and Covenoho

It took much longer for Ontario to settle the question. In 2011, Justice Low decided what we call the "Wunderman" decision (Wright v. The Young and Rubicam Group of Companies).

Wunderman dealt with a more complex issue than Shore: The contract in question had a formula that scaled with length of service at a pace slightly different than the statutory minimums, and in a way that appears to have been intended to comply with the statute in all instances - but its author appears to have forgotten that Ontario's statutory severance accrues in partial years: Ultimately, there were brief windows around 8.5, 9.5, 10.5, and 18.5 years of service that could (under certain plausible conditions) have generated a contractual outcome marginally less than the statutory minimums.

Justice Low followed Shore: The validity of the term is assessed as at the time it is entered into as to whether it actually complies with the statute in all scenarios, and not with respect to any particular point in time over the life of the contract.

There is, in my view, no particular difficulty in fashioning a termination clause that does not violate either the minimum standards imposed by the Employment Standards Act or the prohibition against waiving statutory minimum requirements and there is no compelling reason to uphold a termination clause which the draftsman may reasonably be understood to have known was not enforceable either at all or under certain circumstances.

This matter did not go to the Court of Appeal, but it was a persuasive Superior Court decision. Many of us in Ontario (myself included, at that time) operated on an assumption that it was likely to stand up, because the premise was well-reasoned and consistent with appellate-level jurisprudence from B.C.

There was a hiccup in the decision in Ford v. Keegan, a decision with (as I have argued from time to time) a number of analytical problems, where Justice Price expressly declined to follow Justice Low.

However, if there was any doubt that Wunderman got it right, the Ontario Court of Appeal subsequently settled the debate in Covenoho: An employment contract providing for two weeks' termination notice was terminated by the employer in the first three months, when Ms. Covenoho's statutory entitlements were nil.

The Court of Appeal, relying on Shore and Wunderman, found that she was entitled to reasonable notice:

In determining whether the contract is in compliance with the ESA, the terms must be construed as if the appellant had continued to be employed beyond three months; if a provision’s application potentially violates the ESA at any date after hiring, it is void.

An Alberta Controversy

I'm unaware of any cases in Alberta considering this issue above the Provincial Court level. CanLII lists zero Alberta cases citing Shore, two ABPC decisions citing Wunderman (Makela, which distinguishes it, and Lui, which follows it), and only one decision citing Covenoho (again, Lui).

So, on the reported case law, there's only one decision in Alberta actually considering these lines of cases - a single 2019 ABPC decision by Judge Young.

However, there was also a recent ABPC decision by Judge Golden, Lawton, which upholds the validity of a simple '4 week notice' termination provision: "The severance without notice pursuant to the Employment Standards Code would be 1 week, not 4 weeks as negotiated." (There is no other discussion as to the impact of the Code, nor any references in the decision to the Shore lines of cases.)

So there's no binding precedent that a lawyer like me can take into a courtroom to say that a simple 4-week termination clause is illegal, regardless of when the contract is terminated.

Only one Provincial Court decision, a couple of extraprovincial appellate decisions...and a close analytical approach to show why it's right.

(If you're wondering "What about the other 11 jurisdictions?...the answer is that there are few citations of these cases outside BC and ON. In addition to the AB cases, there's a Nova Scotia Small Claims Court decision considering Wunderman; a preliminary decision in an unjust dismissal proceeding under the Canada Labour Code where Wunderman was raised in argument...and that's about it. This isn't all that unusual; the sheer volume of litigation coming out of ON, BC, and AB - based on population and economic activity - dwarfs the other common law jurisdictions.)

The Underlying Logic

The question, to the extent there's a question, is best framed this way: At what point is the validity of a contractual clause to be interpreted for purposes of statutory compliance? At the time it is entered into? Or at the time that its effect is engaged?

In other words, is a non-compliant contractual term void ab initio, or is it a valid term of the agreement until it restricts an employee from their statutory minimums?

There are a few reasons why the Shore interpretation has to be correct: Firstly, there's the underlying policy analysis from Machtinger about requiring employers to make their contracts enforceable at the front end. This is the primary basis for the outcome in Shore.

But beyond that general 'policy' issue, close technical interpretations of the underlying statutory language and contractual principles also require this outcome.

The Statutory Language

As I mentioned earlier, employment standards legislation use a voiding mechanism. The language isn't identical from Province to Province, but it's comparable.

Ontario: "Subject to subsection (2), no employer or agent of an employer and no employee or agent of an employee shall contract out of or waive an employment standard and any such contracting out or waiver is void."

BC: "The requirements of this Act and the regulations are minimum requirements and an agreement to waive any of those requirements, not being an agreement referred to in section 3 (2), has no effect."

Alberta: "An agreement that this Act or a provision of it does not apply, or that the remedies provided by it are not to be available for an employee, is against public policy and void."

So the specific words are different. (I encourage Ontario lawyers to ponder whether the ONCA's decision in Uber might have been easier with Alberta's language.) But in terms of how they operate, there's no material difference: An agreement that limits an employee's entitlements to less than the statutory minimum standard is simply void.

This is distinct from, hypothetically, language that would preserve employment standards minimums 'notwithstanding an agreement to the contrary'. The Legislature is actually stepping in to modify the terms of the contract itself.

And the voiding mechanism is based on the language of the agreement, not the effect of the agreement.

Contractual Principles

Every contract has an express or implied termination provision - an agreement (whether actually discussed or not) as to how the parties can extricate themselves from the relationship. In the employment relationship, the default is 'reasonable notice', but this implied contractual term can be displaced by express language.

When you void a contractual provision, what you're really saying is "Nope, this language doesn't form part of the contract, so the implied term will govern instead."

It's not the case that the implied term of reasonable notice forms some substratum of all employment contracts. It's either a term and condition that governs the contract, or it doesn't form part of the contract at all. This is conceptually misunderstood: Pay in lieu of notice isn't some obscure employment-specific pigeonhole doctrine; it's simply common law damages flowing from the breach of an express or implied contractual term. (The very language of 'pay in lieu of notice' is simply shorthand for the damages flowing from a failure to provide required notice. See my "Defining Wrongful Dismissal" paper, and the SCC's decision in Matthews.)

Therefore, as much as a decision like Wunderman seems weirdly technical, the alternative is frankly absurd: Imagine a framework that says "At 8 years, the written contract governs; at 8.5 years, the implied term is part of the contract; at 9 years, the written contract governs again for a month; then the implied term again forms part of the contract until the 10 year mark", etc. While there are conceivable circumstances where a material change in the law or facts could affect the ongoing validity of a contractual term, the idea of a contractual pendulum is incoherent.

(The kinds of cases where validity might change include scenarios where there was a statutory change: A previously-valid term might be invalidated by an amendment, in which case presumably the implied term of reasonable notice would insert itself. Or in a case where a fixed term contract turned into a de facto indefinite term contract, a termination clause that was going to be valid through the life of the fixed term contract might be scrutinized differently.)

One of the first lines of inquiry for a court in a wrongful dismissal action is "What contractual terms governed the employer's ability to dismiss?" The notion that the very contents of the contract can shift depending on the circumstances of termination...is quite absurd. Whether or not a particular term forms a valid part of the contract cannot and should not depend on whether the employer chooses to terminate the contract under circumstances where the clause's formula happens to meet applicable statutory requirements.

Other Implications

The point at which the statute serves to void a contractual term is significant in a few ways - not just in respect of enforceability, but also in respect of the requirements for an employer to amend a contract. Amending a contract requires 'fresh consideration' - something of value going each way that is a new commitment not present in the existing contract. This means that where an existing employee is entitled to 'reasonable' notice, and agrees to modify that entitlement to something that is less, without some other enhancement (bonus, raise, etc.), the agreement fails to actually modify the contract.

So a close look at the existing terms is important when amendment a contract.

Suppose I enter into an employment agreement with an American 'at will' termination clause, then the employer subsequently has me sign an agreement that provides only the statutory minimums (with no other amendments). Is this change an enhancement for me (changing "I get nothing" to "I get a little bit"?), or - in light of Machtinger - would this be a rollback of my 'reasonable notice' entitlement without consideration?

If we take the Shore approach, the answer is clear: The 'at will' clause was void from the moment it was entered into, and the implied term of reasonable notice formed a term and condition of the contract. Amending that to reduce notice entitlements would require consideration.

If we take the Ford v. Keegan approach, the answer is muddy. Would the 'at will' clause form a valid part of the contract until relied upon (notwithstanding that at any point after three months an attempt to rely on it would be patently illegal), such that a new contract that provides any notice would be an enhancement? Would the 'at will' clause be valid until circumstances actually existed that reliance upon it would be illegal? (And wouldn't that logic run into the same problem? If a 'stat minimum' clause still gives me zero at the time I sign it, is that really an enhancement over the at will clause? You can't quantify it as actually conferring any benefit without taking a forward-looking perspective to a timeframe when statutory minimum benefits would be owing - and where the at-will clause would be void.)

The Outer Edges of the Doctrine

There are a few interesting thought experiments to run about the kinds of scenarios that will void a contract.

Different periods of non-compliance

For example, the existing Shore line of cases stands pretty firmly for a proposition that a termination clause is rendered invalid by future non-compliance with statutory minimums, but what about past non-compliance? For example, if you have a contract that limits your termination entitlements to x weeks "per completed year of service", this can create a problem from 3-12 months, when statutory minimums provide an entitlement but the contract doesn't. Once you're past the 12 month point, assuming no other defects in the language, is the term still void?

In my view, this is an even stronger straightforward application of the Shore reasoning - provided that the term was entered into before the end of the first 12 months. If there was an actual block of time where the contract would have failed to comply with the statute, the clause couldn't have survived that period.

By contrast, if an existing employee with over a year of service gets promoted, and signs a contract with such language (taking into account earlier service) as part of that promotion, the concern doesn't arise.

Similarly there may be cases where a fixed term contract's early termination language language need not cover off entitlements that would only arise after the end of the fixed term. (Though an employer would need to be careful about extending such a contract.)

Different hypothetical reasons

For the sake of demonstration, let's look at Ontario's unique termination requirements: In addition to 'notice', Ontario has a contingent entitlement to statutory "severance pay", if the employee has more than five years of service and one of the two following conditions are satisfied:

(a) the severance occurred because of a permanent discontinuance of all or part of the employer’s business at an establishment and the employee is one of 50 or more employees who have their employment relationship severed within a six-month period as a result; or

(b) the employer has a payroll of $2.5 million or more.

$2.5 million isn't nearly as much as it was when this language was first implemented in 1987, and it covers a lot of employers these days.

But if I'm hiring someone in Ontario for my operation with a payroll that doesn't even approach that threshold, and where I have nowhere near 50 employees, is a contractual provision that ultimately serves to exclude statutory severance actually a problem? Hypothetically, x years down the road maybe I will be able to pass those thresholds.

Analytically, that seems like a stretch, and different from the hypothetical of "if you continue to be employed here for x years", because an indefinite term contract inherently contemplates that provision continuing to apply in x years.

But if I have a five-digit payroll, no plans to grow my business to that scale, and no expectation that it will ever reach that scale (and, in fact, it never reaches that scale), then it seems silly to invalidate a clause on the basis of failing to account for a contingency that was never reasonably contemplated.

But where's the line? Clearly, in a fact-pattern where my business takes off and actually surpasses a $2.5M budget, that clause can no longer be enforced. Perhaps the paradigm here is that the voiding mechanism in the ESA doesn't engage until the business actually achieves a threshold that could trigger severance obligations. Perhaps that occurs when the business first plans to achieve such a threshold.

Statutory changes

A few years ago, the Alberta NDP changed Alberta's Employment Standards Code to modify group termination requirements: When large numbers of employees are dismissed, different notice obligations applied depending on the scale of the group dismissal, going as high as 16 weeks.

If more than 50 employees were dismissed from a single location, 8 weeks' notice were required. Dismissal of over 100 would require 12 weeks' notice. For over 300 employees, 16 weeks' notice.

This was changed back after the UCP took government, but for a brief window of time, 8 weeks was no longer a hard limit to statutory minimum entitlements, and contractual formulas that were keyed to the statutory 'individual termination' formulas were not necessarily adequate to satisfy contractual entitlements.

Contracts can be drafted flexibly enough to take into account subsequent statutory changes, but if they aren't so drafted, the courts can't rewrite the terms of the contract to adapt to the changes. So a contract that gave an employee a flat termination entitlement of 8 weeks' notice may well have been legal when entered into, but if the group termination (s.137(1)) was engaged with entitlements of 12 or 16 weeks, there wouldn't have been any coherent interpretive approach capable of reconciling the language of the contract with the obligations under the statute.

Similarly, I've seen more than one contract coming out of large and institutional employers that specifically referred to the individual formula in the Code - s.56 - to quantify the required amount of notice, as opposed to referring to s.55, which is the one setting out the statutory obligation to provide notice of termination more generally. This is important, because during the period with the group termination provisions, s.55 directed the calculation to be made either with reference to s.56 or s.137(1).

So a contract that says "We're going to calculate your notice period in accordance with s.56" would pretty directly operate to contract out of s.137(1) (as it existed briefly), contrary to s.4 of the Code.

The Shore approach gives a compelling argument that, at least for employers potentially capable of engaging s.137(1), s.4 voids contracts entered into while the group notice requirements were effective; however, and more importantly, it likely also creates a compelling argument that, at the moment the NDP's amendments became effective, those changes had the legal effect of immediately voiding any clause that operated inconsistently with them.

(NB: There's some discussion of retroactivity in respect of Ontario's new legislation against non-competition agreements. Ontario's Ministry of Labour has made a statement that the change does not apply to pre-existing agreements. If that interpretation of the language is accurate (and I find it genuinely ambiguous), it's distinguishable from other employment standards in that what's actually prohibited here is the entering into the agreement, as distinct from providing employees with a substantive statutory right to compete. This interpretation would be difficult to apply in other contexts: For instance, if I hire an employee to work for me at $15/hour, and then the minimum wage goes up to $16/hour, I can't claim that my $15/hour agreement is still valid because it predated the change.)

Conclusion - The Formation and Resolution of Legal Schisms

Employment law often moves in a way where multiple approaches get adopted across the country, gradually working toward consensus or - sometimes - calling for clarity from the SCC.

While we now think of the 1960 Ontario High Court decision in Bardal as being beyond question nationally, that wasn't always the case: There used to be a doctrinal battle over whether 'reasonable notice' should be based on presumed intention of the parties at the time the contract was entered into, or based on circumstances present at the time notice of termination was given. The latter approach is, of course, Bardal, but the former approach derived from another 1960 decision that was upheld at the ONCA, Lazarowicz.

Courts periodically struggled with this divergence up until the 1990s. Courts generally leaned toward the Bardal approach, but some appellate courts were skeptical because it seemed to break from broader concepts of contract law. It appears to me that the issue stopped arising shortly after Machtinger - where the majority of the court reinstated a Bardal-premised notice period, and where the concurring reasons of Justice McLachlin (as she then was) explained why 'presumed intention' was not necessary for reasonable notice.

Or there's Matthews, addressing a question of whether the 'implied term' of reasonable notice requires actual notice or is contractually satisfied by pay in lieu of notice. As I reviewed in my paper, the question was pretty one-sided for many decades until the ABCA invented an alternative answer; but as of 2019 it had grown into a genuine schism in the law.

Then there's the question of the impact of express termination clauses on the duty to mitigate. Until Bowes v. Goss Power, you might have said that Ontario was an outlier in applying a duty to mitigate to contractual severance language. (Before the ONCA's decision in Bowes, I blogged about it a couple of times, arguing that the Ontario approach needed to change. After the decision, Bowes' counsel reached out to me and told me that his client had felt that I was the 'only one on the internet on his side'.)

Since Howard v. Benson Group out of Ontario and Quach v. Mitrux out of BC, we'd probably say there's a split on the duty to mitigate a breached fixed term contract. (IMHO: Ontario gets this wrong.)

Or there's the question of how to treat time with a predecessor employer, where Ontario recently expressly broke from the approach adopted in other Provinces. (IMHO: Ontario gets this right.)

Alberta hasn't yet picked a side on this one, and I think and hope that a properly-articulated argument at QB and/or the ABCA will persuade them that the Ontario and BC precedents got it right...but judges, and particularly Alberta judges, tend to dislike hypotheticals.

The reality, of course, is that - despite my use of the word here - the conflict isn't strictly hypothetical: It's a real conflict that exists between the language of the contract and the language of the statute, that exists prior to and notwithstanding any question of whether an operative difference later manifests.

*****

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.