Bill 32: The Constitutionality of Limiting a Union's Political Activities

This week, Professor Doorey published an article I wrote about Alberta Bill 32 on his Canadian Law of Work Forum.  I'm not going to reproduce it here - just click the link:  http://lawofwork.ca/bill32-charter/

Consider this post a bit of a meta-narrative, where I'll prattle on about some other perspectives not fully addressed in that article.

Bill 32 which passed on Tuesday and received royal assent yesterday, is attracting national attention, particularly because of its treatment of political speech. It's getting attention for other reasons, too - my social media feeds have been filled with CUPE ads drawing attention to its overtime reform, and there's concern about limitations to secondary picketing - but in most of these ways it's more or less undoing changes by the Notley government, if perhaps going a bit further (not to say that some of these aren't constitutionally questionable in their own right).  The 'political activity' issue is probably Bill 32's most revolutionary and controversial approach.

In a nutshell, I argue that separating out a union's political and representative activities is impossible, and that a union's political action is either an extension of its representative capacity, or else is integrally connected to its representative capacity.  (In truth, there are probably elements of both, depending on the circumstances and the type of political issue.)  I leaned heavily on the Supreme Court's 29-year-old decision in Lavigne, where the court seemed deeply skeptical of the viability of such a separation.  It was largely obiter there, and not precisely on point to the constitutional issues raised by Bill 32, but the analysis of the public policy issues are closely analogous.

My argument is that Lavigne provides a solid foundation for today's courts to recognize the deep integration of a union's political and representational activities, and that the damage done to the American model of labour relations by trying to create such a dichotomy is even more evident today than it was in 1991.

The Feasby Article

Now, let's introduce the counterpoint:  In a recent ABlawg entry, lawyer Colin Feasby argues that these legislative reforms are not obviously constitutionally objectionable on their own, but that the constitutional analysis may be tainted by its "naked partisan purposes".

A constitutional challenge to Bill 32 would be a difficult case. A law requiring union members to opt-in to having their dues used for political purposes, when viewed apart from its context, is not inherently objectionable and can be defended under section 1 of the Charter as promoting the legitimate purpose of enhancing union members’ freedom of expression and association. Bill 32 only appears to be constitutionally objectionable when viewed from a structural or contextual perspective considering both its partisan purposes and its adverse effect on political competition.

Feasby's academic and practical constitutional law credentials dwarf mine, and I have little to add to his detailed discussion of how - and at what stage of the analytical framework - the government's partisan motivation should be evaluated.

In preparing my own article, while this problem absolutely came up on my radar, I deliberately stayed away from discussion of the bona fides of the government's pressing and substantial objective, instead choosing to focus on whether or not the contents of Bill are, to use Feasby's terminology, 'inherently objectionable'.

Proportionality

This is where I most differ from Feasby, and also - because it deals with the ground-level realities of labour relations - where I am most qualified to disagree with him. The difference between our positions most significantly boils down to a simple question: Is the ostensible objective of the legislation (individual employee choice) proportionate to the interference in Charter rights it generates?

To be clear, this question arises at the last stage of a Charter analysis. It only matters if a Charter right is infringed upon, in a legislative initiative furthering a pressing and substantial objective, in a minimally impairing way. (I also doubt that Bill 32 is minimally impairing, but there are much easier fixes for that than the proportionality problem.) Feasby appears to acknowledge that Bill 32 infringes upon Charter rights - though he appears to think that it's primarily freedom of expression that is engaged, whereas I regard the infringement upon freedom of expression as a secondary issue to freedom of association, for reasons that I'll highlight a little further on.

Feasby takes a deferential stance on this.  Whereas I might even challenge the assertion that individual employee choice is a pressing and substantial legislative objective in the context of a majoritarian labour relations model, Feasby argues that the objective actually advances other Charter rights of the individual bargaining unit members, and concludes that the legislature is entitled to deference in how it balances Charter rights.

There are two problems with this position. Firstly, the idea that individual employee choice in payment of dues engages Charter rights...is precisely what the majority of the Supreme Court of Canada rejected in Lavigne.  To be fair to Feasby, the interpretation of the Charter, and in particular s.2(d), has changed dramatically over the intervening decades; still, his suggestion that requiring an opt-in for dues enhances "union members’ freedom of expression and association" is one that runs directly contrary to the current canon.

The second and more important issue is that the deference to be accorded to the legislature is far from absolute.  The question is whether or not the legislature's decision falls within "a range of reasonable alternatives".

So while Feasby appears to jump from 'the legislature is entitled to deference' directly to a conclusion that the Bill is not 'inherently objectionable', I would respectfully suggest that he has skipped the step of considering whether or not the balancing of interests, in this case, actually falls within that range of reasonable alternatives.

Expression as a Peripheral Issue

I suspect that Feasby is falling into a very specific trap here - that is, the trap of presupposing that the distinction created by Bill 32 between political and representational activities is a viable distinction at all, and that you could hamstring a union's ability to do the former without impacting the latter.  In which case it probably looks like a 'sum zero' balancing of interests to him - the union doesn't have the money in its coffers to donate to an NDP-supportive PAC; and the employee has that same money to choose to donate to political causes, or not.

I still think an expression analysis calls for further consideration of what the Bill actually does.  If the Bill limited this adverse treatment to non-member dues - i.e. the dues required by virtue of the statutory entrenchment of the Rand formula - then I'd barely think there's an expression interest engaged at all.  Just the government tying a string to that legal obligation that the compelled payment can't be used for political purposes.

However, the Rand formula doesn't account for all the dues, initiation fees, etc. collected by trade unions. Union members are liable for dues and initiation fees regardless of whether or not they are within a Rand formula bargaining unit; they undertook to pay those fees voluntarily when they joined the union. (One might be forgiven for thinking that all union dues are 'mandatory':  Even if you're aware that the Rand formula isn't universal, you're likely aware that other workplaces have different types of union security requirements, like union shops and closed shops.  However, union shops and closed shops, unlike the Rand formula, are not creatures of statute. Furthermore, union members who are not part of any bargaining unit may still be liable for dues and/or initiation fees, simply by virtue of their membership - just like any other dues, fees, or subscription costs we may pay to organizations we deal with.)

And because Bill 32 extends beyond the Rand formula, including dues and initiation fees charged to members who voluntarily join the union and undertake to pay those fees, the government creating further procedural and substantive constraints that prevent the union from using those revenues to fund speech...is deeply problematic. It's no different than if it were in private industry: Imagine telling Bayer that it can't use the proceeds of Aspirin sales to finance lobbying unless it gives consumers the option to not pay that portion of the cost.  It's a very deep cut into s.2 rights.

But the reason I don't look at expression as the primary Charter right engaged here is because it's sort of curable.  If the Legislature were to limit this 'political activity' treatment to non-member dues (as in the US), then Feasby would probably be right that any remaining incursion into freedom of expression would be justifiable under s.1.

The Incurability of the s.2(d) Issue

Freedom of association, however, is a more fundamental problem in this context. If you accept - contrary to the assumptions underlying the Bill - that a union, as a consequence of being limited in its political activities (as defined within Bill 32), will be deeply impaired in its ability to carry out its core representative functions...then we're into a wholly different analysis: Now we're balancing an abstract advancement of individual choice by not being required to contribute to a union's political speech, against a material impairment of the ability of workers to collectively pursue bargaining interests in the first place.

And I'm not out in left field when I suggest that a union's political activities are integral to its representational mandate - the Lavigne decision is filled with subtextual (and sometimes textual) acceptance of this premise. The political activities of unions are both integral to their fundamental representational roles, and indispensable in our broader national discourse.

The collective action protected by s.2(d) isn't strictly limited to bargaining against one's employer, but also to challenging the employer's influence over public policy. And opt-outs (or worse, opt-ins) are anathema to such collective action: We get into a phenomenon most often described as a 'free rider' issue, but I actually prefer to describe it as a prisoner's dilemma:  The 'free rider' suggests against people who opt out of union dues while continuing to take the union's benefit, whereas the prisoner's dilemma highlights that, where such option is available, individual non-payment is always the rational choice, even though collective non-payment hurts all participants. Put simply, in most contexts, if every union member pays dues, everyone's better off.  However, regardless of whether or not everyone else pays dues, I am also better off not paying dues. (If everyone else pays dues, and I don't, then my withholding of dues won't materially impact the union's representation of my interests, but I also get to keep my money. If everyone else withholds dues, and I pay them, then my dues will not meaningfully enable representation of my interests, and I'm also out that money.)

A glance south of the border is instructive here. They long operated on the basis of such a dichotomy (though even that was just for non-member fees), and their system of organized labour has effectively collapsed. Of course, correlation does not equal causation, and n=1, but the reduced union participation in public discourse has had a palpable impact on the quality of that discourse. While the differences in our political cultures are myriad and complex, it isn't a coincidence that America has (a) bifurcated political and representational activities of unions; (b) transitioned to challenge the viability of trade unions altogether through right-to-work legislation; (c) heavily declined in union representation rates; and (d) become a veritable plutocracy, where the purchase of political influence is not only ubiquitous, but constitutionally protected.

This is what Justice Wilson described as a "disproportionate weakening of the union voice", citing a 1985 text by Harvard Professor Laurence Tribe:

In theory, of course, employers and employees are supposed to get evenhanded treatment in the legal administrative framework within which labor disputes ‑‑ including those involving free speech ‑‑ are handled.  In reality, however, workers have found themselves at a substantial disadvantage.  When unions speak out on political matters, for example, they must (upon request) refund to dissenting members the prorated cost of such activity.  Corporations do not have this problem; corporations may speak out on political subjects in spite of shareholder dissent.  Corporations also speak with a far louder voice, heavily outspending labor on the dissemination of their views.  Indeed, the proof of this imbalance of power can be seen in the results:  "the failure of labor to pass any legislation affecting the basic structure of private sector bargaining since 1935," and the decline in the rate of union representation of American workers from 35 percent in the 1940s to barely 20 percent in 1980.

This has grown worse since the 80s. Union representation overall is down to about 10%; worker rights are trampled on in ways that seem astonishing to a Canadian observer; and yet the average American voter seems utterly convinced that anti-union and other anti-worker policies - like 'right-to-work' and 'at-will employment' - operate to their benefit.

And Bill 32 is worse than the American model in at least two ways: Firstly, it creates an 'opt-in' instead of an 'opt-out', which has an immediate and severe impact on participation. (Back in law school, we held a referendum on whether to keep the LL.B. designation or switch to the J.D. My recollection is that some 30% of my class preferred to keep the LL.B...but the transition came with an opt-out for my year, allowing people to receive an LL.B. if they filled out and submitted a simple one-page form to student services. Of a class of approximately 160, 3 of us submitted the form. Similary, a UCP MLA last year introduced a private member's bill to change organ donation to 'opt out', noting the very high percentage of Canadians who support organ donation, but the very low percentage who actually take the steps to donate their organs.)

Secondly, it extends the option to union members, not just to non-union members working in unionized environments.

In this light, the proportionality issue is likely to be straightforward: The abstract individual freedoms to not pay the union representing you is not at all proportionate to the broad damage that would be done to the pursuit of collective bargaining objectives.

*****

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.