The Right to Strike

The lines are drawn for the next great battle of contemporary labour law.

For all of the issues that have arisen in recent labour disputes in Ontario and in the Federal sphere, including the Province designating the TTC as an essential service and the Feds legislatively interfering labour disputes involving Air Canada and Canada Post, this case has arisen in Saskatchewan, and has provided - at the trial level - an answer to a question whose spectre was raised 11 years ago by the Supreme Court in the Dunmore decision:  Does s.2(d) of the Charter of Rights and Freedoms guarantee a right to strike?

The History

In the early Charter jurisprudence, the Supreme Court pretty thoroughly rejected the notion that s.2(d) (freedom of association) protected union-related rights.  However, in Dunmore the Supreme Court overturned those earlier cases, accepting that the freedom of association included the freedom to join unions.

Dunmore was important for a few reasons.  It also has become a precedent for the Charter requiring what is characterized as 'positive government action'.  Traditionally, Charter rights only restricted what a government could do.  Freedom of expression prevents someone from being silenced, but does not require anyone to provide a soapbox.  It means that the government can't force my blog to be shut down (without good reasons), but doesn't require the government to provide hosting for my blog, or to force Blogger to continue to host it.

So how does that apply to a right to join a trade union?  Under a traditional Charter analysis, it would mean that the government can't make it illegal for a group to join a union.  Early in the labour movement, but in days long past, this was the case.  But with most groups these days have statutory rights to join unions, the question took on different dimensions:  Dunmore was about the exclusion of agricultural workers from the labour relations regime, about saying to society at large, "Everyone can join unions...except these guys."  The Court in Dunmore concluded that this had a chilling effect, that the exclusion of one group in itself sent a message to employers that agricultural unionization efforts would be illegitimate, and by doing so the government was interfering with such unionization efforts.  In response, the legislature enacted a law, the AEPA, giving agricultural workers the right to join unions and make representations to employers.  We'll come back to that in a minute.

The next question was what was included with this right to join a union.  Can the employer proceed to ignore  the union anyways, in the absence of a statutory regime telling them otherwise?  Does s.2(d) include not just a right to join a union, but also a right to collective bargaining?  And if so, does that impose a positive obligation on government to protect such a right?

The second case was B.C. Health Services in 2007, and the answer to the question was Yes.  The British Columbia government had legislatively overridden several public sector collective agreements, and the Supreme Court found that this interfered with the associational rights of the individuals involved - a meaningful right to associate means protection of the objectives for which people associate, including collective bargaining.  The case reinforced Dunmore and referenced the discussion from that case of 'underinclusive legislation'.  I've tended to argue that the discussion of positive government action in B.C. Health Services is somewhat misplaced, since we are talking about a case involving the government actually legislatively undermining collective agreements, but I (though I'm probably alone in this) really don't read Health Services as meaningfully expanding the "positive obligations" scope from the extremely limited doctrine set forth in Dunmore.

Then, there was the third case, Fraser, dealing with the AEPA in Ontario.  The objection was that the AEPA didn't obligate employers to actually negotiate in good faith when receiving representations from the union.  The Supreme Court's response, in a remarkable feat of legislative interpretation (see my commentary here), was "Yes it does", finding that good faith obligations could be read into the statute, and that s.2(d) may guarantee a meaningful right of collective bargaining but does not guarantee any particular mode or structure for that collective bargaining.

I am somewhat disturbed by how far Fraser comes from Dunmore's 'the government can't substantially interfere with the right to organize' to 'the government is obligated to enact statutes imposing good faith obligations on employers'.  But that's not for today.

What Now?


Yesterday, the Saskatchewan Court of Queen's Bench released a decision finding that s.2(d) constitutionally guarantees the right to strike.  The facts involve a statute enacted which essentially imposes a wide-spread 'essential service' model, stripping public sector employees of the right to strike.

The Court noted the role of the strike in labour relations.  In the private sector, a strike imposes economic pressure on both sides - the employees aren't making money, but the employer isn't getting labour.  In the public sector, the pressure on the employer is different, usually being political in nature.

It's a lengthy decision, but I will deal with it in very brief terms for now, at the risk of oversimplifying:  The Court views the strike as being a keystone of collective action - without the right to strike, to exert collective pressure against the employer, the right to collectively bargain is meaningless.  Thus, s.2(d) must extend to the right to strike.

Further, Canada's international obligations and participation in ILO conventions tend toward the recognition of a right to strike.

My Thoughts

I'm not an expert in international law.  There are people far smarter than me who have argued either way on this point in the wake of B.C. Health Services, but I dislike the invocation of international law in this context.  It is difficult to accept that an international treaty, which governments can enter into or rescind, can inform the interpretation of the Canadian constitution.

In any event, even if my distaste for the Supreme Court's approach in Health Services is unsubstantiated, the Court in this case takes it further.  You see, the right to strike is not expressly set out in the applicable ILO convention, but rather has been read in by the ILO over time.  Namely, in 1971, the ILO appointed a Committee to inquire into a complaint against Greece, which concluded that the right to strike was implicitly protected by convention 87.  The Saskatchewan Court here pointed out that Canada ratified the treaty in 1972, after that decision made it clear that the Committee on Freedom of Association was taking that position, and therefore it informs the contents of our international obligations, which in turn (following B.C. Health Services) inform the content of our Charter rights.

Okay, let's parse this for a second:  We have freedom of association.  Because, in 1972, we ratified a convention agreeing to protect collective bargaining, that gets lumped in with freedom of association.  And because, in 1971, an international organization concluded that the right to strike was implicitly protected by a right to collectively bargain, that means our constitution must therefore guarantee a right to strike.  Even if I were to put aside the heavy deference on an international organization's interpretation which was not at all informed by Canadian legal principles, this is starting to get a little disturbing for other reasons, but I'll get to that in a second.

The Court also refers to the presumption of conformity with international law, which is a principle of statutory interpretation suggesting that, in the absence of an irresistably clear intention not to comply with international obligations, a law should be read as being consistent with international obligations.

This is true.  I'm quite fond of that principle, actually.  I like international cooperation and multi-lateralism, and I'm fond of the fact that a government is unable to subtly legislate around its international obligations - it can do so, and should be able to do so, but has to admit openly that that's what it is doing.

But it's a rebuttable presumption, premised on the notion that "If the government really wanted to ignore its international obligations, it could have said so expressly."  Moreover, if that's really what the government wanted to do, they can just go and amend the law to do so.  When we're talking about Charter interpretation, that notion no longer exists.  The Charter is there.  It has been there since 1982, and it's nigh-impossible to amend.  If we were talking about international obligations that post-date the Charter, I'd be deeply worried about the Court saying "Well, if they didn't want us to interpret the Charter to include this international obligation, they should have put that in the fine print of the Charter."  Such a conclusion would be absurd.

In this case, of course, the international obligations predate the Charter, but if that's the deciding factor then that worries me for different reasons:  Unlike that of our southern neighbours, the Canadian constitution is a "living tree".  It grows and changes as the society it protects grows and changes.  The intended meaning upon drafting isn't completely irrelevant, but nor is it determinative.  American constitutional interpretation today is rather muddied by the fact that they feel beholden to the ideals, values, and intentions of those who drafted the document, well over two centuries ago.  Two hundred years from now, when our society (and economy and workforce) has grown and evolved in ways that we can't necessarily imagine now, are we going to have to interpret our ongoing constitutional rights and obligations based on what a group of non-Canadians said about a conflict in Greece in 1971?

The answer, almost certainly, is No.  That would be anathema to the entire history of Canadian constitutional jurisprudence.  We leave it to our Courts to interpret and determine the contents of the Charter, and we trust them to consider these questions in meaningful and thoughtful ways.  They cannot defer on questions of constitutional interpretation, and even if you grant that Canada's role in the international community may have some interpretive value, it simply cannot be the case that the Courts' hands can or will be tied in context of constitutional interpretation by the conclusions of an international organization.  Canada can decide for itself whether or not meaningful freedom of association requires a right to strike.  If our Courts find that it does not, that will not abrogate from the government's international obligations to recognize a right to strike anyways...but that's kind of my whole point:  It is an *international* obligation, one that we have taken on and which is subject to our own sovereign statehood.  That alone ought not to make it a constitutional obligation.

Furthermore, I'm not certain that I agree with the conclusion that a strike is necessary for meaningful collective bargaining.  There are alternative models, including interest arbitration, for resolving irreconcilable disputes in bargaining.

There are problems with such arbitration, of course.  Firstly, defining the mandate of the arbitrator can be difficult, directing them in what factors they should consider in coming to a decision.  Secondly, widespread arbitration can result in a disconnect from free market pressures and bargaining power, resulting in wages which are either inordinately high or low with regard to market conditions, because arbitrated results tend to focus strongly on "What are other people getting?".  (That being said, negotiated results often have that dimension, too.)  Arbitration also tends to resist innovation - an arbitrator doesn't want to adjudicate something new between the parties that doesn't exist elsewhere.

Yet I'm not sure that any of these problems are such that they can be said to interfere with meaningful association by definition.  While a statute that puts economic concerns of the employer at the forefront of an arbitrator's mandate would likely do so, a balanced formulation of the mandate is not impossible.  As for the prospect of a break from market conditions, this could be built into the mandate to some extent.  Not a perfect solution, but it's hard to imagine wages skyrocketing out of control, or unionized employees suddenly becoming drastically underpaid.  And the issue of stagnation is troubling, but nor is it one-sided.  If a party wants something new and different, they can still bargain for it, and bargaining may require compromise, on both sides.  A no-strike model would likely have prevented Ford from bullying CUPE into compromising its job security provisions.

In light of Fraser, noting that no specific labour relations model is necessary, I think that the existence of no-strike alternative models which have been effectively used in some contexts will make it difficult for this decision to survive its appeals.

What next?

This will be appealed, likely all the way to the Supreme Court in time.  If it holds up, the consequences will be widespread.  Many of the current Federal government's actions will be seen as ultra vires (outside of their power), interfering with collective bargaining processes.  Back-to-work legislation in general will become presumptively unconstitutional, unless it can be justified under section 1 of the Charter.  Essential services legislation will be the same - most likely, the TTC's essential services designation will be done away with.

I don't like the approach taken by the Federal government to labour issues.  I do think that, where the government is going to step in, a fair alternative is necessary, and that has not been happening.  But, with fair alternatives available, I dislike the notion of an outright prohibition of government interference in labour disputes.  Labour disruptions can and often do have major impacts on the public interest, and I'm not laissez-faire enough to say that we should always just leave it to market pressures to decide who's right.

*****

This blog is not intended to and does not provide legal advice to any person in respect of any particular legal issue, and does not create a solicitor-client relationship with any readers, but rather provides general legal information. If you have a legal issue or possible legal issue, contact a lawyer.

Fraser v. Ontario - Is the pendulum about to swing back?

The Supreme Court of Canada last week released its decision in Fraser v. Ontario, which has been long-awaited in labour law circles, on the question of the extent of the Charter right to collective bargaining.

To highlight the importance of this decision, note that the decision involved fifteen interveners, 33 lawyers, and 9 judges releasing 4 sets of reasons in 369 paragraphs.

There is much to be said about this decision. Many of the debates raised could be parsed in detail; I could write a full-length paper on the nuances of the distinction between ‘freedoms’ and ‘rights’ that Rothstein and Charron attempt to raise and the majority dismisses. But in this entry I will focus on the broad strokes of the decision.

A Brief History of Labour Relations under the Charter

Traditionally, the ‘standard’ labour relations regime follows what we call the ‘Wagner’ model: If the majority of employees in a bargaining unit support a union, that union gets the exclusive right to represent all employees in the bargaining unit in negotiations with the employer. There are certain exclusions from the Wagner model, including a traditional exclusion for agricultural workers in Ontario.

There were a series of early Charter cases where the Unions tried to raise “freedom of association” (s.2(d)) to challenge limitations on labour relations regimes. The Courts rejected these arguments: Freedom of association had nothing to do with forming unions.

Then the facts changed. In the 1990’s, Ontario’s Rae government removed the exclusion for agricultural workers. This was short-lived, and was reinstated by the subsequent Harris government. This was challenged in a case referred to as Dunmore, and the Supreme Court overturned its own prior jurisprudence, finding that s.2(d) of the Charter did, in fact, protect against government interference with people’s formations of employee associations and unions.

Remember: The Charter of Rights and Freedoms is a constitutional document which limits the powers of government. All the jurisdiction before Dunmore was clear that it could not be used to compel positive government action. Freedom of speech doesn't obligate the government to provide me with a soapbox; it just means that the government can't shut down my blog without good reason. Dunmore expressly maintained that proposition. However, the Supreme Court decided – rightly or wrongly – that the exclusion of agricultural workers in fact created a ‘chilling effect’, delegitimizing attempts by agricultural workers to organize into unions, making it harder for them to do so than if there was no labour relations protection at all. Accordingly, the exclusion was unconstitutional.

So the Ontario government enacted the Agricultural Employees Protection Act (“AEPA”), which gave employees the right to form associations and make representations to the employer. The employer is obligated to listen to and acknowledge receipt of those representations. That’s it.

So when a group of mushroom farm employees in Leamington were thoroughly ignored by their employer, Rol-Land Farms, they brought a constitutional challenge to the AEPA, saying it still didn’t give them a meaningful right to organize. This was the Fraser case. It went before Justice Farley, who held that there was no constitutionally protected right to collective bargaining in good faith, and that the AEPA had been enacted in accordance with the Supreme Court’s requirements in Dunmore.

Fraser appealed to the Ontario Court of Appeal. But before it was heard, things got interesting.

The appeal was delayed pending disposition of the Health Services case from B.C. at the Supreme Court: The government of British Columbia had legislated around and over the terms of various health care sector collective agreements, and the question became whether or not this violated s.2(d). The Supreme Court found that it did, and that there was a right to collectively bargain in good faith.

If that sounds weird, it should. I have the freedom of speech. That doesn’t compel others to listen. I have freedom of religion. That doesn’t compel you to provide me with a church. Yet saying that there’s a ‘right’ to have good faith collective bargaining suggests that the Charter itself requires employers to bargain in good faith with me. But, generally, it doesn’t: The Charter doesn’t affect private actors.

It makes some sense that the B.C. government might not be able to legislate over existing collective agreements. It kind of undermines the purpose and function of the employee association, and is clearly positive governmental action subject to Charter scrutiny.

But to go a step further and say that the Charter actually requires governments to legislate an obligation for employers to bargain in good faith…well, that is something different. And that is what the Fraser case became about: Basically a question of “Is that really what the Supreme Court said in Health Services?”

My Two Cents: The 2008 Hicks Morley Moot

It was around this time, in my third year of law school, that I participated in the Hicks Morley moot at Queen’s. Frequently, a moot involves addressing an issue answered by the Court of Appeal with a strong dissent. The appellant has to distill and argue the dissenting judge’s argument, and the respondent has to distill and argue the majority argument. This moot was more interesting, however, as the moot was of the appeal from Justice Farley’s decision, post-Health Services. A shift in the law meant that the moot was treading entirely new ground, that Justice Farley's conclusions of law were outdated and that we were looking at a fully novel legal question.

I was happily arguing the government side, and dealing with the s.2(d) issue. The structure of my argument was essentially this: The suggestion of a carte blanche proposition that everyone is entitled under the Charter to a Wagner-esque labour relations scheme is simply wrong, and is a misstatement of Dunmore and Health Services.

Quite the contrary; Dunmore was clearly a very narrow decision, expressly decided on the very unique factual situation where the existing legislation was shown to actually, in fact, have a chilling effect on organizing campaigns. Quite expressly, Dunmore would have permitted the legislature to scrap labour relations protections altogether. (Political suicide, certainly, but well within their constitutional powers.) The Supreme Court in Health Services referred to Dunmore as supporting the proposition that affirmative state action could be required in narrow circumstances where there was an evidentiary foundation to support the proposition that the government’s actions themselves were responsible for the impossibility of meaningful association.

Health Services should be interpreted similarly narrowly. “The Charter applies only to state action. One form of state action is the passage of legislation. In this case, the legislature of British Columbia has passed legislation applying to relations between health care sector employers and the unions accredited to those employers.” (Paragraph 88.) Health Services does not purport to reverse or modify the general rule that the Charter cannot force positive government action, but adopts the general rule subject to the narrow nuanced exception in Dunmore.

Read in that context, the Court’s more sweeping statements regarding the contents of the right to collectively bargain (such as “The parties have a duty to engage in meaningful dialogue and they must be willing to exchange and explain their positions. They must make a reasonable effort to arrive at an acceptable contract” at paragraph 101) must be read simply as explaining the content of meaningful association and simply cannot be read as imposing large-scale affirmative obligations on government (or private actors) to ensure widespread meaningful association.

So the only question is whether or not the AEPA creates the same chilling effect as the statute in Dunmore did, and I argued that there was not (or, at least, was not yet) the evidentiary foundation to support the existence of such a chill. Justice Farley had not made the factual findings necessary to ground such a claim, and had in fact noted that it was too soon to tell what the effects of the AEPA were.

Eventually, the Ontario Court of Appeal essentially found that Dunmore plus Health Services equals an affirmative obligation for legislatures to enact comprehensive labour protections for all, subject to s.1 of the Charter, and accordingly the AEPA was unconstitutional.

The SCC Decision

The Court was divided here: Of nine judges, there was a majority decision by five of them, two sets of concurring reasons (sort of…really dissenting) by three of them, and a lone dissent from Justice Abella. Justice Abella was the only one arguing that the appeal should be dismissed and the AEPA found unconstitutional…but the other 8 essentially had a very important 5-3 split.

Let’s start with the concurring reasons. Justice Deschamps argued that Health Services should be interpreted narrowly, and that it is difficult to begin imposing affirmative obligations on government on the basis of the Dunmore analysis; these cases are about economic inequality, but the courts have always refused to recognize economic inequality as a s.15 (anti-discrimination) ground, and the Dunmore analysis was a convoluted sidestep to allow that continued refusal. Justice Deschamps would prefer to open up the s.15 analysis and resort to the ‘omission’ analysis in Vriend v. Alberta. (Alberta couldn't exclude sexual orientation as a ground of protection in its Human Rights legislation.)

Justices Rothstein and Charron argued that Dunmore and Health Services were wrongly decided, that they broke with established jurisprudence and are unworkable and undesirable and should be reversed.

The majority, however, rejected these arguments. Health Services is still good law, did not constitute a break from existing jurisprudence, and should not be overturned without good reason. It is too soon to tell if it is, indeed, unworkable.

The majority essentially said that a full-blown Wagner model is not necessary, but the legislative regime is required to give all necessary protections to ensure a meaningful right to collective bargaining. They looked at the AEPA and decided to read in an obligation on the employer to consider proposals in good faith.

Sections 5(6) and (7) are critical. They provide that the employer shall listen to oral representations, and read written representations, and acknowledge having read them. They do not expressly refer to a requirement that the employer consider employee representations in good faith. Nor do they rule it out. By implication, they include such a requirement.

The argument for such an interpretation is three-fold: (1) A statute should be interpreted in such a way as to give effect to its purposes, and there’s not much purpose in telling an employer to listen to a proposal if they aren’t being required to consider it; (2) a statute is presumed to intend to comply with the Charter; and (3) the Minister said that the government of Ontario intends to meet its obligations to grant meaningful freedom of association.

If I may comment personally, I find this argument to be disingenuous. Firstly, the majority is picking and choosing its principles of statutory interpretation. The legislature’s word choice is important, and had they intended to impose good faith obligations such as in the Labour Relations Act, they would have used similar language. The fact that the legislature only requires the employer to “acknowledge” the representations has meaning, which the majority is ignoring. In fact, it seems that the majority is finding ambiguity where none exists: The language simply does not bear the interpretation given to it by the majority. No reasonable person would look at the language and suggest that it actually creates an obligation on the employer to bargain in good faith. Rather, the obligations it places on the employer are quite explicit, unambiguous, and minimal.

Further, the ‘obligations’ the Minister promised to meet were the ones set out in Dunmore – his language is quite express – and this predated Health Services. It cannot be the case that this rather puffed up statement of intention to comply with the Charter gives the Courts carte blanche to read in whatever remedial language they like.

The Minister also said that the AEPA was not intended to “extend collective bargaining to agricultural workers”, but the majority chose to interpret their way around that.

According to the majority, ultimately, the AEPA (“correctly interpreted”) protects the right to have employee submissions considered in good faith, and therefore is constitutional. The majority further notes, as I argued, that the AEPA has not been fully tested to see if the protection it gives to freedom of association is ‘meaningful’.

My Analysis

I smell change on the wind again. Ten years after Dunmore, the pendulum has just hit its apex on the swing towards labour. This is a weak majority decision by the Supreme Court, with certain irrationalities in the decision that makes the whole thing smell a little of horsetrading (and makes the whole thing rather bad law), and the next few judges to retire will have their replacements appointed by Mr. Harper, meaning that they likely won’t be labour-friendly, for better or for worse.

I mentioned irrationalities: The suggestion that Health Services does not break from established jurisprudence is revisionist. It is said to hold true to Dunmore (which was, itself, a massive break from established jurisprudence), but in fact was a significant expansion from Dunmore. Indeed, this decision interprets Health Services quite broadly, finding that there is a carte blanche obligation on governments to impose labour relations regimes which protect good faith bargaining.

I would also object to the fact that the majority seems to presuppose, without the necessary analysis, the unconstitutionality of the AEPA in the absence of the language being read in.

No, seriously, step back and look at the second rationale for reading in a good faith obligation: Legislation should be presumed to be intended to comply with the Charter. Therefore, we will interpret the language in such a way as to comply with s.2(d), rendering it unnecessary to proceed with a s.1 analysis.

But without a s.1 analysis, how can you say that the alternative interpretation would be unconstitutional? In fairness, of course, the s.1 analysis wouldn't have saved the alternate interpretation.

Yet the decision isn’t fully labour-friendly. Ultimately, the UFCW lost. They were going for a full-blown Wagner scheme, and they didn’t get it. They now get to go the employer and say “See, you have to consider these representations in good faith”…then when the employer comes back and says “Okay, we’ve considered them carefully, and decided that we are not prepared to agree to them; now go away”, their only recourse is to a novel and untested administrative tribunal process. The Court is not prepared to dictate how the substantive s.2(d) rights must be addressed, which gives the government some flexibility.

So this decision is an expansion on labour rights, but a tempered one, like when Vikings Rage pauses near the top of its highest swing before coming back crashing down the other way.

Also, the expansion itself may carry the Dunmore doctrine beyond the weight it can reasonably support. It goes from the Court telling the government “You can’t legislate contrary to the Charter” to the Court dictating public policy. This is so deeply at odds with the history of Charter jurisprudence that it cannot hold. Particularly in light of the door being left open for future arguments about ‘unworkability’, I think we will see those arguments being made into the future, and I think the Courts will scale the doctrine back over time.

Any thoughts from readers? Where will s.2(d) go next?

And a further question: I would invite perspectives on whether or not the expansion of s.2(d) opens up the door to challenges to the Rand formula.

*****

This blog is not intended to, and does not, provide legal advice to any person in respect of any particular legal issue, and does not create a solicitor-client relationship with any readers, but rather provides general legal information. If you have a legal issue or possible legal issue, contact a lawyer.