More On Illegal 'For Cause' Termination Language

 Two years ago, I wrote an entry on the ONCA decision in Waksdale, finding that a contractual term for 'termination for cause' that offended employment standards legislation had the effect of invalidating the rest of the termination language, such that an employer could not rely upon the 'not-for-cause' provisions of the agreement.

There's a new one, Rahman v. Cannon Design Architecture, dealing with essentially the same question, and coming down the same way.

The main difference between the two cases is that, in Waksdale, the parties agreed for the purposes of the appeal that the 'for cause' language was illegal, and the 'not-for-cause' language was not otherwise problematic; in Rahman, the employer argued (and the motions judge accepted) that the 'for cause' language should be interpreted as ESA-compliant because that's how both parties (including a relatively sophisticated plaintiff) would have understood it.

So in some ways, this case is more about 'what brings a clause within the Waksdale framework' as opposed to questioning the Waksdale approach up front.

The Court of Appeal concluded that the "plain wording" of the provision ran afoul of the ESA, and that it was a mistake to try to cure that by looking to the sophistication of the plaintiff. (This conclusion probably has significance across a larger class of cases, but it shouldn't surprise any employment lawyer.)

But while this case wasn't really a direct invitation to the ONCA to reconsider Waksdale (there's a specific process for that), the subtext of the Court's decision has a bit of a "Did we stutter?" feel to it? Many judges, faced with a precedent with which they have significant concerns, would feel inclined to express those concerns even while following it, to signal to the bar that this question may still be up for further discussion in the right case.

(It's interesting to note that the panel that decided Waksdale was predominantly from a commercial law background - two practicing in major national firms, one splitting early in her career from a major firm to a boutique litigation firm where she practiced commercial and employment law. The panel in Rahman, by contrast, was academic-heavy. Justice Gillese used to be Dean of Law at UWO; Justice Harvison Young was Dean of Law at Queen's; Justice Trotter was another Queen's professor and briefly Acting Dean after AHY's appointment to the bench - though, to be fair, he had a number of years of practice under his belt before he joined academia.)

There's another recent development that I may bring up from time to time in 'judicial comity', which makes these decisions more relevant in Alberta than they may previously have been.

The Merits

I argued before that I think Waksdale got it right. The more I think about it, the more I think this analysis has to be the right one.

Practically speaking, one of the more compelling issues here is that it's the only answer that gives employers an incentive to ensure statutory compliance at the front end - a consideration that has guided the SCC before in cases like Machtinger and Shafron.

But I'm more interested in the analytical approach.

I continue to approach the issue from a perspective that silos off the different contractual termination events. The default at common law is that there's 'just cause', voluntary resignation, and literally everything else: The 'everything else' category entitles employees to notice; however, resignation or termination for cause do not.

Many express contracts follow a similar matrix. Some do not. Some have specific terms dealing with sale of the business, winding down of the business, death of the employee, death of the employer, termination following employee notice, or various other contingencies.

For example - and this is true - there's an absolutely AWFUL Law Depot precedent that employer clients keep bringing to me that has weird illegal language dealing with termination due to business discontinuance - the language is borderline incoherent and appears to be drawn from a fixed term contract (though the rest of the contract is usually structured as an 'indefinite term' contract), but when I work through the language I come to the conclusion that its intended effect is that, if the employer shuts down, the employee's contract simply comes to an end with no further obligations.

Setting aside, for the moment, the drafting weirdness of a clause with that effect, let's imagine we litigate such a matter - so a not-particularly-hypothetical contract with the following termination structure:

1. You can resign on x weeks written notice
2. If you engage in sufficiently serious misconduct, we can fire you for cause without notice.
3. We can fire you at any time without cause on provision of y weeks notice.
4. If the business shuts down, your employment ends at the shutdown date and we owe nothing further to you.

Assume, for the sake of argument, that the second and third points don't raise statutory concerns on their own. But the fourth obviously does.

This creates four different silos without overlap, setting mutually exclusive expectations of the kinds of events that will trigger termination and the resulting entitlements. On the face of the contract, if the employer terminates the contract per para 3, you have y entitlements; but if the employer terminates the contract per para 4, you have no entitlements. (In the event of an improper invocation of paras 2 or 4, setting aside validity questions, para 3 guides an employee's damage award only by virtue of Hamilton v. Open Window Bakery. It doesn't change the nature of the termination itself.) 

So here's the problem: If you say that para 4 was void ab initio, as employment standards legislation does, then what does that do to the matrix? The parties intended a specific (void) result to flow in the event of a business shutdown; by voiding that provision, you need to figure out what to do with terminations that would/should/could have been within that silo.

If you just delete the paragraph, that creates what would be a coherent framework, but it's a different framework than the one the parties agreed to, where now the employer is required to provide y notice in the event of a business shutdown. That simply does not give effect to the bargain struck by the parties; it's substantively rewriting the bargain.

Admittedly, it'll usually be a less severe impact than striking the whole framework in favour of the 'implied term of reasonable notice', but that's beside the point: When we rely on the implied term, we're expressly not relying on the intention of the parties. Rewriting the bargain in a way to get closer to the intention of the parties is precisely what the SCC rejected in Machtinger.

For the same reason, you can't just 'employment standards minimum' the invalid language, because, again, Machtinger. And the only plausible alternative would be to say that a business shutdown triggers entitlements to reasonable notice - which has an incoherent impact, again, due to Open Window. There's no arguable way of interpreting the rest of the language that doesn't practically result in putting paragraph 3 notice on a paragraph 4 event, changing the impact of other language than the clause being voided.

Simply put, you can't remove the void paragraph without changing the meaning of the other language. As a result, the different clauses can't be read independently of one another, and therefore can't be severed.

The More Typical Case: The Three-Silo Contract

Both Waksdale and Rahman deal with deficiencies in the 'just cause' language - raising what appears to be a different factual question from my 'four-silo' example by applying it to the more frequent contractual scenario that is terminable only by resignation, termination for cause, and not-for-cause discretionary termination.

This is, to some extent, an Ontario-ism, in that Ontario's ESA uses different language than 'just cause' to disqualify employees from their statutory minimums. (The precise nature of the distinction is somewhat arguable, and there's some recent case law, Render, suggesting that it's actually a pretty big gap. I have my doubts about that interpretation, but there's no reasonable dispute that there is a gap.) However, in other Provinces there's some potential relevance where a contract tries to lean on the definition of 'cause'.

So let's take a case like Render, where the employee engages in misconduct that is found (a) to be just cause at common law and (b) not to disqualify the employee from ESA minimums. Without a written contract governing entitlements, as in Render, the outcome is that the employee gets his/her ESA minimums, but not common law pay in lieu of notice. (I haven't worked through it entirely yet, but I'm not 100% convinced that this is quite right. Let's set that aside.)

If you imagine that Render had a contract like Rahman's, which silo defines the employee's entitlements? At face value, the employee gets nothing, but that's void because of the ESA, so the employee is entitled to...what? Again, correcting the deficiency in the language and providing only ESA minimums is exactly the kind of thing that the SCC has said time and time again 'we're not doing that'. Again, by simply removing the contractual treatment of cause, you're changing the meaning of other terms in the contract, and entitling the employee who has committed serious misconduct to the same benefits that occur on discretionary termination - rewriting the contract.

The only solution, from a contractual interpretation point of view, is to say that the termination provisions were void ab initio. In this instance, that puts the employee into the same spot Render actually occupied, getting statutory minimums only but nothing more - actually the best potentially-legal outcome for the employer. (Rahman's contract, under some circumstances, would give more than the statutory minimums. Invalidating the entire contract and replacing it with the framework applied in Render wouldn't actually help an employee who has committed acts of just cause.)

It Matters in Alberta

Alberta doesn't have that same dichotomy between 'statutory cause' and 'common law cause', really. Alberta's regulations emulate the common law standard, so while one might hypothetically interpret the regulatory standard as different from the common law, using the language of 'just cause' isn't likely to invalidate a provision.

However, it's also not too unusual to see a contract using the 'just cause' provisions to list off the kinds of things the employer regards as constituting cause, in a way that seems to say, "If you commit x act at all, we have the right to fire you without any notice."

That's going to engage a Waksdale-type question: Does that language, in and of itself, contract out of the Code? The standard of 'just cause' is highly contextual; it's impossible to say at the beginning of the relationship that any particular act will certainly justify termination without notice - so if I have a contract saying "We can dismiss you not-for-cause on payment of statutory minimums, or if we dismiss you for dishonesty you get nothing", then I get dismissed on a not-for-cause basis...I'd challenge that based on the Waksdale reasoning.

Historically, Alberta hasn't been shy about saying, "Yeah, other Provinces are saying x, but we disagree." (My Alberta Schism paper was on just such an example.) It's almost akin to American exceptionalism at times, in the sense that Alberta's courts almost take pride in cultivating their own distinctive jurisprudential culture.

And it wouldn't altogether surprise me to see the ABCA doing the same thing here.

But there's another development from the SCC that makes case law coming out of other appellate courts a little harder to hand-wave away by saying "That's not the law here."

In R. v. Sullivan, the SCC recently considered the doctrine of 'judicial comity' or 'horizontal stare decisis' in a new way.

This is a concept that's gotten a bit of discussion at the SCC before in terms of reticence (sometimes from dissent) in overturning its own precedents, and throughout the case law, you can find this occasionally arising in a sense of "We're not bound by decisions of the same level, but we should stick to them unless there's a pretty compelling reason not to." But they haven't historically talked much about it at the lower level - if a question goes to the SCC with conflicting decisions below, they tend to decide which approach they prefer, not slap the hand of whichever conflicting approach came second.

In Sullivan, the SCC is now clear that - at least intra-Provincially - decisions of the same court are more than merely 'persuasive'. Decisions by the same court, or courts at the same level within a Province, are binding unless (a) there's subsequent appellate intervention on the point, (b) the prior decision was reached out of carelessness, or (c) the decision was not fully considered (taken in exigent circumstances).

Reading the whole Sullivan decision, while the court uses language of the 'same level' of court, it's pretty clear that they're not extending the principle interprovincially: Constitutional principles related to Federalism make it impossible for, say, an Ontario court to bind Alberta's.

But the principles leading them to make horizontal stare decisis binding within the Province - a discussion of judicial comity and rule of law - are not limited by those constitutional considerations, and I would argue that that extra-jurisdictional precedents now have more weight. Not that Alberta courts are bound to follow Ontario precedents, but that, as a matter of judicial comity, Alberta courts are at least required to give serious consideration to another Province's decisions before declining to follow as a matter of exceptionalism. (Of course, this assumes similar principles apply - consideration of the same common law principles, for example, or of analogous statutory language. Relevant distinguishing features are always a good enough reason to not follow a given precedent.)

The Waksdale issue is something that reasonable lawyers can disagree about. I've called it a close case before. But wherever you come down on the analytical questions, it's hard to deny that (a) the consequential impact of demanding employers use legally-compliant language is consistent with the long-standing approach of the SCC; and (b) the ONCA's conclusion that the termination provisions of a contract aren't severable is at minimum a coherent interpretive approach.

So I'd suggest that the ABCA, while not bound by the ONCA, at minimum has to regard it as persuasive, regarding such a question with due regard for judicial comity and legal predictability: The creation of a new schism should be limited to scenarios where you can compellingly conclude that the extrajurisdictional court got it wrong, and not scenarios where there are two pretty reasonable-looking approaches and you simply prefer the other one.

*****

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.