Unconstitutional and Unconstitutional

Why delegating plenary taxing powers to the executive is wrong as a matter of constitutional principle and constitutional law

The government’s fortunately short-lived proposal to arrogate to itself the power to make regulations “that have the effect of repealing or imposing a tax, decreasing or increasing a rate or an amount of tax or otherwise changing the incidence of tax” generated a flurry of discussion about aspects of the constitution that are both fundamental and obscure. The most impressive contribution to this conversation is that of co-blogger Mark Mancini. Mark argues that, while a sweeping delegation of the power to tax to the executive is bad policy, it is not unconstitutional. Specifically, he addresses two arguments about it constitutionality: one based on section 53 of the Constitution Act, 1867, and one based on the unwritten principle of democracy.

For my part, I am not convinced by what Mark says about section 53, and I think that the principle of democracy is not the most important one to think about here. In my view, the Supreme Court’s interpretation of section 53 does not support ― and indeed give reason to challenge ― a delegation as sweeping as that which was apparently contemplated. The principle of responsible government ― not just democracy writ large ― also calls it into question. Before getting to these arguments about constitutional law, though, I think it’s important to emphasize that a plenary delegation of taxing powers is unconstitutional in a somewhat different sense.

Government action can be meaningfully said to be unconstitutional even if it contradicts no rule of binding constitutional law that could be enforced by the courts. This is most obviously so in the case of a breach of constitutional convention (assuming, that is, that the orthodox distinction between convention and law still holds), but arguably even in the absence of a violation of a precise rule, if government acts contrary to fundamental principle. It is in this sense that the governments (and Parliaments) of the United Kingdom and of New Zealand can be said to act unconstitutionally. The constitutions of these polities are not entrenched and judicially enforceable, but they are no less real, and susceptible of being contravened in a way that calls for denunciation in constitutional terms.

One of the fundamental principles of the Westminster constitutions since at least 1688 is that of Parliamentary control over taxation. Mark refers to the post-Glorious Revolution constitutional settlement by saying that “if the Bill of Rights of 1688 meant anything, it meant that Parliament came into its own as the controller of the executive; it became a sovereign body” ― but that’s not quite right. The references to Parliamentary control of the executive in the Bill of Rights are more precise than a general assertion of sovereignty. They do not focus on Parliament’s power to make laws ― that was a given, and the Crown’s inability to make new law was recognized in the Case of Proclamations 80 years earlier. Nor do they involve a general control of the executive ― that would only come with responsible government, which developed over a long period of time starting decades after the Glorious Revolution and not taking a final form until the 1830s.

What the Bill of Rights 1688 did do was to impose firm prohibitions on the Crown “suspending laws”, “dispensing with laws”, and “levying money for or to the use of the Crown by pretence of prerogative, without grant of Parliament, for longer time, or in other manner than the same is or shall be granted”. Now, “consent of Parliament” overrides these restrictions, as it obviously does that on the Crown’s law-making power. Acting “by and with the advice and consent” of Parliament, the Crown can make and change law, and it can impose and abolish taxes. The question, though, is whether this consent can be given prospectively, in advance, and in the form in effect of a blank cheque. After all, granting the Crown, acting on the advice of its Privy Council (and, in practice, of the cabinet) rather than of Parliament, the power of “repealing or imposing a tax, decreasing or increasing a rate or an amount of tax” amounts to nothing else.

In my view, the principle behind article 4 of the Bill of Rights ― the one dealing with “levying money without grant of Parliament” ― requires specific authorization on an ongoing basis. Parliament sought, and succeeded in gaining the ability, to actually keep tabs on the executive’s finances. It did not do so to simply let the executive run itself as if 1688 hadn’t happened. “The Crown can imposes whatever taxes and imposts it pleaseth, for ever and ever” would not be consistent with the purpose of article 4, and the contrary idea wouldn’t have occurred to anyone until the development of responsible government, and indeed well after. But even now, it is not a sound idea. Parliamentary scrutiny of taxation must be constant to be effective. It cannot just happen once in a blue moon, and the vagaries of question time are not a sufficient substitute for accountability mechanisms focused on taxation and spending.

The proposed delegation of taxing power to the executive was not, of course, for ever and ever. But it would have lasted almost half the duration of a normal Parliament, and longer than hung Parliaments typically survive in Canada. And it was, of course, quite uncabined ― the executive really would have been able to do anything it pleased. In my view, it is absolutely contrary to the principle and spirit of article 4 of the Bill of rights 1688, and so not merely stupid, but actually unconstitutional, at least in the sense of being inconsistent with the constitution’s underlying commitments. Whether the courts would have been able to do anything about this is a separate question, and a moot one at this point.

Despite its mootness, I turn now to the question of the constitutional legality of the government’s proposal. As noted above, the key constitutional provision here is section 53 of the Constitution Act, 1867, which provides that “Bills for appropriating any Part of the Public Revenue, or for imposing any Tax or Impost, shall originate in the House of Commons”. The question is whether taxes imposed pursuant to delegation, and one as broad as the one described above, meet this requirement.

It may be worth pointing out that the exact status of section 53 is somewhat mysterious. The Supreme Court has long held, as Justice Iacobucci put it in Ontario English Catholic Teachers’ Assn v Ontario (Attorney General), 2001 SCC 15, [2001] 1 SCR 470, that “[s]ections 53 and 54″ ― of which more shortly ― can be amended by Parliament”. [68] Yet Justice Iacobucci went on to say that “there is a constitutional guarantee of ‘no taxation without representation'” ― for which section 53 is (rightly) taken to stand ― “in Canada”. [70] I’m not sure how these two statements are to be reconciled. In any event, the position seems to be that, at least so long as section 53 has not in fact been amended, failure to comply with it will result in the invalidity of non-compliant legislation, rather than being taken as (pro tanto) implied repeal. 

So would the proposed delegation comport with section 53? In OECTA, Justice Iacobucci offered the following general principle for assessing delegations of the power to tax:

The delegation of the imposition of a tax is constitutional if express and unambiguous language is used in making the delegation. The animating principle is that only the legislature can impose a new tax ab initio. But if the legislature expressly and clearly authorizes the imposition of a tax by a delegated body or individual, then the requirements of the principle of “no taxation without representation” will be met. In such a situation, the delegated authority is not being used to impose a completely new tax, but only to impose a tax that has been approved by the legislature. [74]

Justice Iacobucci then went on to explain why the delegation at issue ― a grant of power to a Minister to set the rates of a school tax ― was acceptable:

The [impugned statute] … expressly authorizes the Minister of Finance to prescribe the tax rates for school purposes.  When the Minister sets the applicable rates, a tax is not imposed ab initio, but is imposed pursuant to a specific legislative grant of authority.  Furthermore, the delegation of the setting of the rate takes place within a detailed statutory framework, setting out the structure of the tax, the tax base, and the principles for its imposition. [75]

There is, then, a crucial distinction between the imposition of taxes ab initio and the imposition of “a tax that has been approved by the legislature”. Justice Iacobucci’s discussion of the case before him at least strongly suggests that, to count as “approved by the legislature”, the tax ― at least its purpose, but probably also (some of?) its “structure”, “tax base”, and “principles for its imposition” ― has to be described with some specificity.

The proposed delegation of a blanket authority to impose new taxes and to “chang[e] the incidence of tax” is too vague to meet these requirements. It contemplates that taxes might be created, but does not explain to what end they must be levied or on what principles. It amounts to an authorization for the executive to create taxes ab initio ― but OECTA suggests that such an authorization cannot be given, at least, without repealing section 53 of the Constitution Act 1867, and perhaps at all.

Mark writes that, historically, the Supreme Court “has permitted extremely broad delegations of power—especially in crisis situations—so long as the executive remains responsible to Parliament for the exercise of these extraordinary powers”. He recognizes that the leading cases on this, In re Gray, (1918) 57 SCR 150, and Re: Chemicals, [1943] SCR 1 were not decided in the context of taxation, but argues that the principle they stand for, which is that (to quote Mark) “so long as Parliament retains control over the delegated power—so long as it does not ‘abdicate’ its power (Gray, at 157) there is no legal concern”, is applicable.

I’m not so sure. Taxation really is different from other types of legislation. This is where section 54 of the Constitution Act, 1867 comes in. It provides that

It shall not be lawful for the House of Commons to adopt or pass any Vote, Resolution, Address, or Bill for the Appropriation of any Part of the Public Revenue, or of any Tax or Impost, to any Purpose that has not been first recommended to that House by Message of the Governor General in the Session in which such Vote, Resolution, Address, or Bill is proposed.

I feel on shaky ground here, and would welcome correction, but I wonder if the consequence of this provision is not that unlike with normal legislation, where ― in theory, since in practice the executive is actually driving the legislative agenda ― Parliament is indeed free to resume control, when it comes to tax matters, delegation to the executive is a one way street. Once the executive gets its hands on a broad delegated authority to tax, it need not to “recommend” any legislation undercutting this authority by levying taxes not created by regulation to the House of Commons, and Parliament is then handcuffed for as long as the delegation runs. (This also makes delegation of taxing authority to the executive very different from delegation to, municipalities ― municipalities aren’t able to control the enactment of new tax laws by provincial legislatures.)

Let me finally address the other point Mark makes, about unwritten constitutional principles. As explained here not long ago, I am much less skeptical about the use of such principles in judicial decisions than many of my fellow scholars, including Mark. That said, I agree that the principle of democracy is vague ― democracy can take any number of different forms, and we must be careful to implement the specific form of democracy provided for by the Canadian constitution, and not some idealized version of what that principle might mean.

Yet here the relevant principle is not democracy generally, but the particular form of democracy that is at the heart of the Canadian constitutional order: responsible government. In turn, money votes, of which votes on tax bills are one (but not the only) sort are at the heart of responsible government. Winning such votes is how a ministry demonstrates the continued confidence of the House of Commons. Delegation of taxation powers to the executive allows it to avoid these votes, and so arguably undermines, although admittedly it does not completely subvert, this fundamental principle.

What, if anything, the courts might do about this is not an easy question. Courts are sometimes ― although not always, as I have argued in the post linked to above ― reluctant to enforce constitutional principles against legislation. But two precedents are worth thinking about. First, there is Justice Beetz’s warning, in Ontario (Attorney General) v OPSEU, [1987] 2 SCR 2, that there may be limits to a provincial legislature’s ― or Parliament’s ― ability to “do anything it pleases with the principle of responsible government itself”. (46) Justice Beetz is evasive as to the extent and source of these limits, but he does suggest that the legislatures (and Parliament) may lack “power to bring about a profound constitutional upheaval by the introduction of political institutions foreign to and incompatible with the Canadian system”. (47) And second, there is the much more recent Reference re Senate Reform, 2014 SCC 32, [2014] 1 SCR 704, where the Supreme Court found that an entrenched “constitutional architecture” limited the ability of Parliament to bring about constitutional change by ordinary legislation. If I am right that this architecture consists of constitutional conventions, it may well protect the principle of responsible government against fundamental interference, as Justice Beetz suggested.

In short, the delegation of plenary taxing authority to the executive is doubly unconstitutional. It is unconstitutional, first, in the British sense of the word ― as contrary to the constitution’s logic and fundamental commitments. It is unconstitutional, second, in the Canadian sense of the word, as contrary to an express provision of the constitution, and arguably also to its legal underlying principles.

It was not merely stupid, or a bad policy. It was an attempt at a serious breach of the basic rules of our political order. As Keith Whittington has recently written over at the Volokh conspiracy, “[t]he normal logic of political rent-seeking and incompetence does not magically disappear in a crisis, though we might have to be more tolerant of such political failings in order to deal with a fast-moving situation”. The now-defunct proposal was not merely rent-seeking, but a power-grab, perhaps an unprecedented one. The present moment may mean that punishment for it must be delayed, but it ought to count against its perpetrators.

Stupid. But Constitutional.

The Globe and Mail reports that the government is seeking to introduce wideranging methods to permit the Cabinet to raise revenue. However, this report has now evolved, and the proposed measures have been walked back. But the original Globe article said:

One section of the bill grants cabinet the power to change taxation levels through regulation, rather than through legislation approved by Parliament. It states that cabinet will have this power during the period “before 2022.”

“For greater certainty, a regulation made under this section may contain provisions that have the effect of repealing or imposing a tax, decreasing or increasing a rate or an amount of tax or otherwise changing the incidence of tax,” the bill would have stated.

Let’s assume that this reporting was accurate. Let’s also assume that there are more provisions in the bill that set out some more detail on the tax (based on the words “[f]or greater certainty”). In my view, and despite opposing arguments from unwritten principles, I think this Bill would have likely been constitutional. I first address my argument that s.53 of the Constitution Act, 1867 would likely not have been abridged; and second, that the presence of unwritten principles does not change this conclusion.


While this Bill has now been walked back (and probably for very good reason), the old proposal would have been constitutional, because (at least at face value) it clearly delegated taxing power.

Let’s start with the basic point. Section 53, as noted in the seminal Eurig Estate case, encodes the principle of “no taxation without representation, by requiring any bill that imposes a tax to originate with the legislature” (Eurig, at para 30). The restriction here is simple: s.53 prohibits the executive from imposing new taxes ab initio “without the authorization of the legislature” (Eurig, at para 31).

Notably, however, this does not mean that the executive cannot raise taxes. Merely, the executive’s ability to do so is parasitic on clearly-delegated legislative authority. As John Mark Keyes notes in his work Executive Legislation, “[s]ection 53 does not set up an absolute bar to the delegation of taxation powers” (at 122). If it is clear that Parliament has delegated taxing authority to some executive actor, there is no reason to impugn the delegation, constitutionally. This means that executive legislation raising revenue will be constitutionally proper if it does two things: (1) the legislation is enacted pursuant to a delegated power; (2) it is clear that the delegation is a delegation of taxing authority.

Most of the conceptual work is done at the stage of determining whether the delegation is clear. And on that note, the Supreme Court has spoken: consider its opinion in the Ontario English Catholic Teachers Assn case, at para 74:

The delegation of the imposition of a tax is constitutional if express and unambiguous language is used in making the delegation. The animating principle is that only the legislature can impose a new tax ab initio. But if the legislature expressly and clearly authorizes the imposition of a tax by a delegated body or individual, then the requirements of the principle of “no taxation without representation” will be met. In such a situation, the delegated authority is not being used to impose a completely new tax, but only to impose a tax that has been approved by the legislature. The democratic principle is thereby preserved in two ways. First, the legislation expressly delegating the imposition of a tax must be approved by the legislature. Second, the government enacting the delegating legislation remains ultimately accountable to the electorate at the next general election.

The point of the clarity principle, then, is to ensure that the executive is actually acting pursuant to lawfully delegated authority. So long as the delegating provision is clear, there is no constitutional basis to assail it.

Additionally, and as noted above, I am making an assumption that this is not the only operative delegating provision. In other words, it may be a requirement that a bare delegation of taxing authority must be couched in language that sets out the tax’s “structure, base and principles of imposition” (see Keyes, at 124; see also Ontario English Teachers Association, at para 75). I am assuming that this is the case here. But if my assumption is wrong, this becomes a closer case. If the delegation says it is delegating a tax, is that enough on the Supreme Court’s terms? Or is a framework a requirement?

If only the word “tax” is required, or if the taxing power is cabined by other provisions (as it appears to be in this case), then the case for constitutionality is strong. As such, this statute seems to clearly delegate power to the executive to take any number of actions with respect to taxes. Since that authority is lawfully delegated, it likely cannot be impeached in a constitutional sense. And so long as the executive remains responsible for these powers, there is no sense in which it could be said that the executive is evading parliamentary scrutiny.


More broadly, the Supreme Court’s comments on delegation also support the constitutionality of this measure. Though these comments do not relate to taxation, they do underscore the broader context of how the Court has historically viewed delegated power. In short, the Court has permitted extremely broad delegations of power—especially in crisis situations—so long as the executive remains responsible to Parliament for the exercise of these extraordinary powers. The same goes in this situation.

I highlight two cases to this end. In Re Gray, the context was WWI. Under the War Measures Act, Parliament granted power to the executive under a so-called Henry VIII clause; the power to amend or repeal laws, delegated to the executive. The Court upheld this delegation. It said, even though the delegation was extensive, Parliament has not abandoned control over the executive carrying out these powers, and the Ministry remained “responsible directly to Parliament and dependent upon the will of Parliament for the continuance of its official existence” (Gray, at 171). Therefore, so long as Parliament retains control over the delegated power—so long as it does not “abdicate” its power (Gray, at 157) there is no legal concern.

Similarly, in the Chemicals Reference, another broad delegation was at issue. The delegated power permitted the Ministry, in service of WWII efforts, to make rules allowing censorship, control of transportation, forfeiture and disposition of property, and arrest and detention. Again, the Court upheld the delegation :

Parliament retains its power intact and can, whenever it pleases, take the matter directly into its own hands. How far it shall seek the aid of subordinate agencies and how long it shall continue them in existence, are matters for. Parliament and not for courts of law to decide. Parliament has not abdicated its general legislative powers. It has not effaced itself, as has been suggested. It has indicated no intention of abandoning control and has made no abandonment of control, in fact. The subordinate instrumentality, which it has created for exercising the powers, remains responsible directly to Parliament and depends upon the will of Parliament for the continuance of its official exist­ence (Chemicals Reference, at 18).

While these cases might not be directly applicable in the taxation context, they do shed light on the underlying theory that was also present in the Ontario English Catholic Assn case. That is, so long as Parliament controls the delegation and the executive is responsible for the exercise of delegated powers, there is no way to impeach the delegation of power.


I do want to address one potential argument, that is primarily made by Alyn (James) Johnson, in his delegation piece in the UBC L Rev. That argument is based on unwritten constitutional principles (and perhaps constitutional architecture) set out in cases like the Secession Reference and the Senate Reference. One might make the argument that constitutional architecture—the structure and separation of the legislature and the executive—should serve here to prohibit the legislature from delegating its power away in this fashion to the executive. Additionally, Johnson makes the argument primarily based on the principle of democracy: he contends that a “marginalized legislature delegating un-cabined power to willing executive instrumentalities is incoherent and unprincipled.” (Johnson, at 823). More specifically, legislatures are a place for discussion and deliberation; they are fora for democratic contention; but if delegation is widespread, the political/democratic process is lost, and people lose “authorship” over laws (Johnson, at 879-880). Moreover, one could make an argument from the separation of powers: it fundamentally transforms the functions of each of the branches for widespread delegation of this sort to be permitted.

My initial impetus is to be skeptical of unwritten principles and arguments from constitutional structure. For one, the role of unwritten principles is somewhat limited: they may have “normative force” (Quebec Secession Reference, at para 54) they also cannot be used to attack the content of legislation (or so the Court held with respect to the Rule of Law: see Imperial Tobacco, at para 59). In whole, while it could be true that unwritten principles could strike the content of statutes, their role appears to be limited; they cannot, for example, “dispense with the written text of the Constitution” (see Quebec Secesstion Reference at para 53; see also literature questioning the extent of use of unwritten principles: Jean Leclair, “Canada’s Unfathomable Unwritten Constitutional Principles” 2002 27 Queen’s LJ 389 at 400).

Moreover, unwritten principles arguments lack the coherence and structure of traditional doctrinal arguments, and in my view, can be used to support whatever outcome a person wishes. For example, in my view, the principle of “democracy” for example, endorsed by the Supreme Court, might just as well support a Parliament taking an expansive view of its ability to delegate, and delegating widespread authority to the executive. After all, the Court has said that “regulations are the lifeblood of the administrative state” (see Hutterian Brethren, at para 40), and if the Bill of Rights of 1688 meant anything, it meant that Parliament came into its own as the controller of the executive; it became a sovereign body: “each successive delegation of legislative power has been a fresh recognition of that sovereignty” each delegation “a victory at the expense of the Crown” in which the Crown gives up pretensions to legislate by itself (see C.T. Carr, “Delegated Legislation: Three Lectures” at 48-52; see also A.V. Dicey, at Introduction to the Study of the Law of the Constitution, at 6 ). What we are talking about is a sovereign Parliament, and as the Supreme Court has recognized, “parliamentary sovereignty remains foundational to the structure of the Canadian state: aside from constitutional limits, the legislative branch of government remains supreme over the judiciary and the executive” (Pan-Canadian Securities Reference, at para 49). If that is the case, Parliament can just as well delegate its power away if it is sovereign.


In conclusion, I am not trying to say that this law is a good idea. Indeed, there are abstract worries we might think about: what about the separation of powers? What about the institutional functions of each of the branches of government? There are also significant policy reasons to dislike the old proposal. Are these powers proportionate and strictly tailored to their purpose, for example?

Nonetheless, I believe the legal case for a statute of this sort was at least facially strong.

De la connerie

Je voudrais revenir sur une chronique qu’a publiée hier Alain Dubuc dans La Presse. Faisant allusion à la une récente de Libération qui, s’adressant à l’homme le plus riche de France, qui aurait demandé la nationalité belge dans le but de payer moins d’impôts une fois établi dans le plat pays, hurlait “[c]asse-toi, riche con!”, M. Dubuc se demande si les québécois aisés seront, eux-aussi, tentés de “se casser” en réponse aux augmentations d’impôts prévues par le Parti québécois. (Étant donné la position minoritaire du gouvernement du PQ, il n’est pas certain que ces augmentations auront lieu.) Ce qui m’intéresse ici, ce n’est pas l’aspect économique (les hausses prévues sont-elles nécessaires? vont-elles causer un exode d’entrepreneurs?), mais plutôt certaines questions relatives à la moralité politique tant de la hausse proposée que du départ possible de certains contribuables en réponse à cette hausse.

M. Dubuc suggère que les mesures proposées par le PQ sont illégitimes:

Ces taux élevés posent néanmoins un problème d’équité. Les deux tiers des contribuables visés ont un revenu entre 130 000 et 200 000$. Ce ne sont pas des Tony Accurso, mais des gens tout simplement à l’aise, souvent des salariés. Avant de leur taper dessus, il aurait fallu démontrer qu’ils ne paient pas leur juste part. Et démontrer que l’effort que l’on exige d’eux était essentiel. Ce n’est pas le cas. On leur demande d’absorber une contribution santé qui visait 4 millions de personnes et de payer le gel des droits de scolarité. Il y a là un choix idéologique qui entache la légitimité de la ponction.

Je pense que M. Dubuc a tort. Certes, augmenter les impôts sur certains contribuables pour redonner de l’argent à d’autres relève d’un choix idéologique. Et alors? Ne pas le faire, c’est un choix idéologique aussi, seulement animé par une idéologie différente. Comme le reconnaît M. Dubuc lui-même, “[l]a fiscalité, […] ce n’est jamais neutre.” Quels que soient les choix qu’on fait en la matière, l’idéologie y joue un rôle, et il faudrait éviter de prétendre que nos choix sont objectifs alors que ceux de nos adversaires, et seulement ceux-là, sont contaminés par l’idéologie.

Mais si on aurait tort d’essayer de délégitimer le choix politique d’augmenter les impôts sur les contribuables aisés sous prétexte qu’il s’agit d’un choix idéologique, on aurait tout aussi tort de vouloir délégitimer le choix de certains de ces contribuables de “se casser”, à la manière de Libération (dont le titre était, on s’en doute bien, une antiphrase). Car il n’y a rien d’immoral à ce que les membres d’une minorité―et il s’agit bien d’une petite minorité, 135 000 personnes sur les quelque 5 millions d’électeurs québécois―qui, par définition, ne peuvent prévaloir dans une compétition électorale démocratique, cherchent à échapper aux prétentions de la majorité. Ceux qui ne peuvent pas faire compter leur voix ont l’option de la sortie, et l’injuste consiste non pas à ce qu’ils l’exercent, mais à vouloir la leur enlever. Et cela ne dépend pas de la justice de sa revendication sous-jacente. Même si on pense que l’état est en droit d’interdire la consommation des drogues, on ne prétendrait pas, je pense, que la personne qui déménage dans le pays voisin qui, lui, la permet, afin de s’y adonner commet une injustice quelconque. (Pourtant, cette personne-là aussi nous prive de ses impôts et de sa contribution éventuelle à la chose publique.) Il y a quelque chose de pervers, de cruel, à dire à une personne que non seulement on n’a pas la même vision de ses droits qu’elle, mais qu’elle ne doit même pas aller rejoindre un groupe qui, lui, partage la sienne.

La fiscalité, comme bien d’autres enjeux de politique publique, divise les opinions. C’est normal, c’est tant mieux même. Cependant, lorsque nous débattons nos opinions contradictoires, il ne faut pas tomber dans le piège de diaboliser nos adversaires. C’est une chose que de les accuser de ne pas maîtriser les faits ou de ne pas comprendre les implications ou les conséquences de leurs positions. C’en est une autre que de les accuser de mauvaise foi. On peut les traiter de cons si on veut,  mais pas de salauds. Car c’est ça, justement, la vraie connerie.