Over the last few weeks, the Supreme Court re-wrote yet another part of the Constitution ― this time, the Charter’s freedom of association provision. Section 2(d) now means that labour unions have a constitutional right to participate in a “meaningful process of collective bargaining,” created in Mounted Police Association of Ontario v. Canada (Attorney General), 2015 SCC 1, and in many if not in all cases, to strike in order to achieve their goals within that process, created in Saskatchewan Federation of Labour v. Saskatchewan, 2015 SCC 4. After the first of these decisions, Omar Ha-Redeye wrote that “everything we thought we knew” ― everything, that is, that the Supreme Court itself had been telling us ― “about labour law in Canada and s. 2(d) changed.” And then, a couple of weeks later, it changed some more. Mr. Ha-Redeye’s posts, over at Slaw (here and here), set out, with some sympathy if I read him right, the labour law story. In this post, I want to say something about the assumptions behind that story. As will quickly become obvious, for my part, I have no sympathy whatsoever for what the Court has done.
The foundations of the Supreme Court’s freedom of association jurisprudence, and especially its recent decisions, are shaky in several respects. I will skip over its cavalier approach its own recent precedents, which Mr. Ha-Redeye describes, pointing out that “when the law and interpretation of the constitution shifts so drastically, it also leads to some very creative legal writing,” including an “interpretation of existing case law” that is “open to dispute.” Geoff Plant, at Plant’s Rant, has also criticized the court (here and here) ― harshly and, in my view, with good reason. I, instead, will focus on the Court’s economics, methodology, and view of its own powers.
The Court’s utter failure to understand the economics of trade unionism taints its interpretation of the s. 2(d) right. The Court believes that “[w]ithout the right to pursue workplace goals collectively, workers may be left essentially powerless in dealing with their employer or influencing their employment conditions” (Mounted Police, para. 68). Trade unions, and the imposition on employers of a duty to bargain with them, are in its view required to redress this imbalance of power. This is simply not true. A majority of workers in Canada are not unionized. Are they all at the mercy of their employers, unable to obtain reasonable employment conditions? Of course not. At the very least, judges ought to be aware that Bay Street lawyers ― including associates ― are not exactly starving because they are not unionized. Nor are, say, executives, or the employees of Google. In a competitive market, an employer will not be able to impose unfair employment conditions on workers, no more than the buyers of goods are able to impose their conditions on the sellers. Sure, the labour market is not perfectly competitive. But nor is any other. The Court does not even attempt to explain why the labour market is somehow uniquely unbalanced.
Admittedly, it is true that in a competitive market, a worker by himself has almost no ability to “influence his employment conditions” ― but the exact same thing is true of the employer. Employment conditions for a given category of worker, like the price of any good in a competitive market, are those at which the market reaches equilibrium. They are outside the control of anyone ― worker or employer. So the Court is right to say that trade unions are necessary to give workers power to influence their working conditions. But that’s because in order to give anyone such power, the labour market must be distorted and made uncompetitive. This is exactly what trade unions do, by cartelizing labour supply and reducing the competition between workers that helps, like competition between suppliers in any market, keep the prices at equilibrium. In economics terms, labour unions extract monopoly rents from their employers ― who, of course, pass on the cost to consumers, if the employer is a private firm, or to taxpayers, if the employer is the government. (Another consequence of the artificially inflated cost of workers which the unions create is, as in any other market, the demand for labour goes down ― in other words, fewer workers will be employed. The Court seems to be blissfully unaware of this.)
Now some labour markets may be particularly uncompetitive. For instance, for some jobs, the government is (almost) the only employer, and might be able to exercise a monopsony power to drive down the price it pays for labour below what it would be in a competitive market. For these specific occupations, there might be a case for a prima facie right to bargain collectively, though even there, the costs it generates might be such that restrictions to that right would be well justified. But they are exceptions, and cannot justify extending, as a matter of constitutional law, a right to bargain collectively, and to strike, to all workers.
This discussion leads me to a second problem with the basis of the Supreme Court’s section 2(d) jurisprudence, which I referred to, above, as a methodological one ― its lack of grounding in the sort of social science evidence which it has been favouring in some of its recent decisions, such as Canada (Attorney General) v. Bedford, 2013 SCC 72, [2013] 3 S.C.R. 1101. As support for its claim about the purported imbalance of bargaining power, the Court cites not some economic study, but an almost-octogenarian decision of the U.S. Supreme Court. Indeed, the lengthy list of “authors cited” in the Court’s two recent decisions (which includes citations by the dissenters as well the majority opinions) consists largely of labour law scholarship, with a sprinkling of industrial relations studies, and not a single economics one. It is bad enough when judges, in Richard Posner’s words “duck, bluff, weave, change the subject” because “they don’t understand the activity from which a case before them has arisen” (Reflections on Judging, 85-86). But it is worse when think they understand it, but are in fact, proceeding on the basis of ideas that are little better than folklore. To quote Judge Posner once more, “[w]e need evidence-based law across the board, just as we need evidence-based medicine across the board, and not a combination of scientific and folk medicine” (62).
Now the Court is probably not alone to blame for this state of affairs. If the governments that were defending their approaches to labour law failed to adduce relevant economic evidence, or indeed any economic evidence, they missed an opportunity to force the Court to consider it. Moreover, at the s. 1 justification stage, the governments did not press cost considerations as the objectives behind the statutes whose constitutionality was challenged, focusing rather on the need for effective service delivery. Perhaps they had a sense that merely financial reasons would not be regarded as sufficient to justify infringements of rights. And indeed they often shouldn’t be. But the cost of union rights is a different matter than, say, the expense of printing ballot papers which courts surely could, under s. 3 of the Charter, require a government to incur.
That’s not so much because of the amounts in issue, but because making the Court consider the cost of its decisions would have required it to confront the third, and perhaps most fundamental problem with its reasoning ― its misapprehension of its constitutional role. As Justices Rothstein and (in Saskatchewan Federation of Labour) Wagner point out, labour law involves balancing competing interests, which is something legislatures are often better than courts at doing. But the issue is, if anything, even more fundamental. To the extent that public sectors unions are involved ― and they are the ones who drove the litigation that led to these decisions, and their main, if not their only, beneficiaries ― one of the competing interests is that of the taxpayers. The Supreme Court’s intervention in the field of labour law makes public sector workers much more expensive to employ which, at least in the short and medium term, means substantial public expenditures. The issue is not simply one of institutional competence. It is one of basic legitimacy. Allowing the citizens’ elected representatives to decide how they will be taxed, and how their money will be spent is, at least historically, the first and most basic function of a democratic constitution. Charles I lost his head over this issue, James II his crown, and Britain, its American empire.
Still, institutional competence matters too, and indeed it is directly connected to the issue of democratic legitimacy. As the Supreme Court itself had recognized in Irwin Toy Ltd. v. Quebec (Attorney General), [1989] 1 S.C.R. 927,
[w]hen striking a balance between the claims of competing groups, the choice of means, like the choice of ends, frequently will require an assessment of conflicting scientific evidence and differing justified demands on scarce resources. Democratic institutions are meant to let us all share in the responsibility for these difficult choices. Thus, as courts review the results of the legislature’s deliberations, particularly with respect to the protection of vulnerable groups, they must be mindful of the legislature’s representative function.
The idea that legislatures are entitled to some deference (the extent of which, admittedly, has never been made very clear), in making decisions involving a balancing of competing claims, especially competing claims on scarce resources, has been a recurrent ― and perhaps a strengthening ― theme of the Court’s jurisprudence since then.
The defenders of the Courts’ rulings might argue that unions, perhaps especially public sector unions, are in fact “vulnerable groups,” and that the need to protect them against the government’s depredations justifies setting aside the deference to which governments might otherwise be entitled. Perhaps that’s what the Court thinks, given its acceptance of the imbalance of power myth. But experience (which, admittedly, it would be best for governments to support with empirical evidence!) disproves this idea. Public sector unions have frequently been able to hold governments to ransom, and obtain, in exchange for “labour peace,” substantial benefits and advantages for their members. This makes perfect economic sense, of course. Governments aren’t giving away their own money, but taxpayers’, when accede to the civil servants’ demands. Moreover, by paying civil servants with fringe benefits and pensions, they are able to do so relatively untransparently, and often to peddle to cost to future generations rather than imposing it on current voters. But to the extent that current voters have to pay, the government, unlike a firm selling its products on a competitive market, can make them pay more or less as much as it asks. The idea of civil servants’ unions as vulnerable groups, or “discrete and insular minorities,” is groundless.
The Supreme Court’s recent decisions on the scope of s. 2(d) of the Charter are based on economic myths, unsupported by appropriate evidence, and derogate from the Court’s normal constitutional role. They are also inconsistent with the Court’s jurisprudence in other ways, notably in that they constitutionalize organized labour’s economic rights and impose constitutional burdens on private parties, something I will discuss separately. They call for a response, and indeed a reversal, in one way or another ― something I will also discuss separately. In the meantime, they need to be denounced, loud and clear.
Double Aspect 