This text is adapted from remarks delivered separately to the Edmonton Legal Forum and the Miller Thomson LLP Continuing Legal Education Series in May of 2019.
Introduction
I’ve been asked to speak about the carbon tax reference case from the Saskatchewan Court of Appeal. The case arose as a reference from the Saskatchewan Lieutenant Governor in Council as to the constitutionality of the federal Greenhouse Gas Pollution Pricing Act. The Court of Appeal issued a divided opinion, 3-2, upholding the Act.
Chief Justice Richards wrote the majority opinion, on behalf of himself and Justices Jackson and Schwann. Justices Ottenbreit and Caldwell issued a joint dissent.
This is a complex case involving the intersection of a number of challenging questions of constitutional law, including:
- The distinction between a tax and a regulatory charge;
- The constitutional requirements for a valid delegation of taxation authority;
- The potential role of structural principles of federalism in limiting otherwise valid exercises of legislative authority; and
- The requirements for recognizing a new matter under the federal Parliament’s peace, order, and good government (“POGG”) power.
It is a 150-page decision, including a 90-page dissent, so I will just give you the highlights.
Overview of the Legislation
There are essentially two main parts to the federal Act that was being challenged:
- Part 1 imposes a fuel levy on things like gasoline, propane, etc. These are remitted by distributors and passed on to consumers. The levy is based on the CO2 or CO2 equivalent associated with combustion of the fuel. The proceeds of the levy are then to be returned to the province in which they were received. The government has opted to do this in form of a rebate.
- Part 2 sets out a regime for large emitters. Rather than pay a flat levy like consumers, the regime for large emitters is based on emissions intensity. Facilities are assigned an annual limit based on average emissions intensity in their industry. Large emitters only have to pay if they exceed the limit assigned to them.
As an aside, the reason for the special regime for large emitters is probably that these tend to be in trade-exposed sectors of the economy. A straight-up carbon tax could affect their competitiveness, leading these facilities to simply move to a country without a carbon tax. The idea is to give these large facilities a nudge in the direction of efficiency without affecting their competitiveness too much. By contrast, consumers don’t really have the option of moving. You probably aren’t going to move to a different country just because it costs you more to fill up your F-150.
Anyway, that’s the regime in a nutshell. Part 1 is a fuel levy and Part 2 is an intensity-based regime for large emitters.
But now we get to the interesting part. The Act contains a sweeping delegation of authority to the Governor in Council to determine not just the details of how the regimes apply but also where and to whom they apply. A key feature of the regime is that the Governor in Council has the power to apply either Part 1 or Part 2, or both, to a jurisdiction if in the Governor in Council’s opinion, that jurisdiction fails to meet standards for price stringency of greenhouse gas (“GHG”) emissions.
So if you’re a province like British Columbia with a provincial carbon tax regime that meets federal standards, the federal regime doesn’t apply at all. By contrast, the federal regime does apply in Saskatchewan, Manitoba, Ontario, Prince Edward Island, Yukon, and Nunavut. And if you follow the news you’ll know that it will soon likely apply in Alberta. (Update: The regime now does apply in Alberta.)
Majority Opinion
The majority opinion begins by addressing an argument made by Saskatchewan that the non-uniform application of the regime violates principles of federalism. Essentially, Parliament is conditioning the application of the regime based on how a province has exercised its own jurisdiction – arguably trying to influence provinces in the process. The majority holds that the principle of federalism alone is not sufficient to render the regime unconstitutional.
Next, the majority addresses the argument that that the regime imposes a tax in violation of s. 53 of the Constitution Act, 1867. That provision requires that taxes originate as a bill in the House of Commons. Past decisions have allowed for taxation authority to be delegated as long as the Act that delegates the authority originates as a bill in the House of Commons, and as long as the delegation is sufficiently clear. So there are two aspects to this issue: 1) whether the measures in a question are “taxes” and 2) whether the delegation of authority to tax to the Governor in Council is sufficiently clear.
The majority concludes that the measures in question are regulatory charges and not taxes, and so s. 53 does not apply to these measures. In determining that the measures are regulatory charges rather than taxes, the majority relies on the criteria from the Westbank decision of the Supreme Court of Canada. Basically, a tax is a measure enacted for the purpose of raising revenue for general government purposes, whereas a regulatory charge is imposed primarily for regulatory purposes, or as necessarily incidental to a broader regulatory scheme. That case set out the following indicia for a regulatory charge: (1) a complete and detailed code of regulation; (2) a specific regulatory purpose which seeks to affect the behaviour of individuals; (3) actual or properly estimated costs of the regulation; and (4) a relationship between the regulation and the person being regulated, where the person being regulated either causes the need for the regulation, or benefits from it.
The third factor isn’t met in this case, but the majority, relying on Westbank, holds that it doesn’t necessarily have to be. Sometimes regulatory charges are about recouping the cost of the regulation, but in other cases, a regulatory charge can be about directly influencing behaviour – and there is language in the Westbank decision to this effect.
So the majority holds that this is not a tax. The decision goes on to hold that even if it were a tax, the delegation of taxation authority in the Act is sufficiently clear to satisfy s. 53.
The fact that the majority holds this to be a regulatory charge rather than a tax means that it can’t be upheld as an exercise of the federal taxation power under s. 91(3). The majority needs another head of federal power to fit this Act under if it is going to be upheld.
The majority upholds the act under the federal POGG power, specifically under the “national concern” branch of POGG. The federal POGG power is derived from the language at the beginning of s. 91, which states: “It shall be lawful for the Queen, by and with the Advice and Consent of the Senate and House of Commons, to make Laws for the Peace, Order, and good Government of Canada, in relation to all Matters not coming within the Classes of Subjects by this Act assigned exclusively to the Legislatures of the Provinces…”
POGG powers are essentially unenumerated federal powers. One branch of POGG is the “emergency” power, which allows for the federal Parliament to exercise time-limited jurisdiction during an emergency. Another branch, and the one we’re dealing with here, is “national concern”. These are unenumerated matters that have taken on a federal character.
The classic example of a “national concern” POGG power is aeronautics. The federal power over aeronautics was not enumerated in 1867, for obvious reasons, but it was recognized by the courts in the 20th century because the federal government was in a much better position to regulate this emerging area than the provinces. Canada argued that “GHG emissions” or “the cumulative impact of GHG emissions” ought to now be recognized as falling under the national concern branch of POGG.
The current approach to recognizing a new matter under the national concern branch of POGG was set out in Crown Zellerbach:
- The national concern doctrine is separate and distinct from the national emergency doctrine of the peace, order and good government power, which is chiefly distinguishable by the fact that it provides a constitutional basis for what is necessarily legislation of a temporary nature;
- The national concern doctrine applies to both new matters which did not exist at Confederation and to matters which, although originally matters of a local or private nature in a province, have since, in the absence of national emergency, become matters of national concern;
- For a matter to qualify as a matter of national concern in either sense it must have a singleness, distinctiveness and indivisibility that clearly distinguishes it from matters of provincial concern and a scale of impact on provincial jurisdiction that is reconcilable with the fundamental distribution of legislative power under the Constitution;
- In determining whether a matter has attained the required degree of singleness, distinctiveness and indivisibility that clearly distinguishes it from matters of provincial concern it is relevant to consider what would be the effect on extra‑provincial interests of a provincial failure to deal effectively with the control or regulation of the intra‑provincial aspects of the matter.
The majority rejects Canada’s argument that “GHG emissions” or the “cumulative impact of GHG emissions” ought to be recognized under POGG. The majority rejects this argument essentially because it would be too far-reaching. If the federal Parliament has exclusive jurisdiction over GHG emissions, it can regulate a large range of matters that would have previously been under provincial jurisdiction. Examples might include highway speeds, agricultural practices, natural resource extraction, etc.
But the majority does recognize an alternative, more restricted matter under POGG, namely “the establishment of minimum national standards of price stringency for GHG emissions”. This approach is based on British Columbia’s submissions before the Court. The Court holds that minimum pricing standards satisfied the Crown Zellerbach test, and would not unduly alter the division of powers because this matter is more restricted than the alternatives.
And so it is on that basis that the majority upolds the Act. The Act is in pith and substance about setting up minimum standards of price stringency, and that is now a power recognized under POGG. Finally, the majority considers some other heads of power but rejects all of them. Most significantly, the majority rejects the argument that the regime can be upheld under either the federal trade and commerce power or the criminal law power.
Dissenting Opinion
The dissent is interesting, and worth spending some time discussing because this case is most likely going to the Supreme Court eventually.
The dissent first disagrees with the majority’s characterization of the fuel levy as a regulatory charge rather than a tax. The dissent instead characterizes it as a tax. It holds that the primary purpose of the measure is revenue generation, even if the revenue is then returned through rebates. And the revenue generated is not connected to a broader regulatory scheme.
Because the dissent characterizes the levy as a tax, it goes on to consider the constitutional requirements for a valid tax. The dissent would have held that the levy does not satisfy the s. 53 requirement that taxes originate in the House of Commons. The delegation to the Governor in Council is said to be too sweeping to constitute a valid delegation of taxation authority. The dissent points out that the Governor in Council is essentially authorized to change any aspect of the Act by Order in Council, including tax rates, what the tax applies to, and where in Canada it applies. It is also not specific enough in authorizing the imposition of a tax, speaking generally of “charges” rather than specifically authorizing a tax.
Despite the fact that the levy is characterized as a tax, the dissent would not have held it to be a valid exercise of the federal taxation power under s. 91(3). Even though the federal Parliament seemingly has plenary taxation authority, the dissent holds that this does not give it carte blanche to devise taxes that intrude into areas of provincial jurisdiction. (I must say, at this point, the authority the dissent is relying on is pretty thin, but in principle it is an interesting argument.) And so, for the dissent, there has to be another head of federal power that this can fit under, besides the taxation power. The dissent rejects the main candidate, POGG, on the basis that recognizing federal authority in this way would unduly disrupt the division of powers. Accordingly, the dissent would have found the Act to be ultra vires.
Analysis & Commentary
I think there is something to be said for the reasoning of both the majority and the dissent. But the other side of the coin is that I think both the majority and the dissent get some things wrong.
The Majority Decision
First of all, I think there are some issues with the majority’s definition of the new POGG head of power as “the establishment of minimum national standards of price stringency for GHG emissions”. The majority shies away from accepting Canada’s submission that “GHG emissions” or “the cumulative impacts of GHG emissions” should be recognized as a power under POGG, because of the far-reaching effects of this in terms of the intrusion into traditional areas of provincial jurisdiction. The majority tries to settle on a more restricted middle ground: minimum national standards of price stringency for GHG emissions. The thinking is that this more restricted definition of the power will be less likely to significantly impact the balance between federal and provincial jurisdiction. But there are some problems with this way of restricting the ambit of the POGG power. On closer examination this looks like a bit of a shaky middle ground.
The majority defines the matter in question as setting “minimum national standards.” As the dissent points out, you can put “national standards” in front of anything and make it seem like something that only the federal government can address. Obviously, only the federal Parliament can enact laws with national reach. And so you could argue for “minimum national standards for securities regulation” or “minimum national standards for insurance regulation” or “maximum national standards for wage and price increases” and all of these would still be poor candidates for POGG jurisdiction. One might even go so far as to say that if you need to put the phrase “national standards” in front of something in order to make it sound like an appropriate matter under POGG, then maybe that matter on its own lacks the required “singleness, distinctiveness and indivisibility”.
I think there is also a more fundamental issue with how the majority defines the power in question. Limiting the federal power to minimum national standards of price stringency for GHG emissions and not including other means of addressing GHG emissions seems difficult to justify in a principled way. Setting standards of price stringency is really only one means among several options that a government could choose to address GHG emissions. And essentially all of the arguments that the majority makes in favour of federal jurisdiction over standards of price stringency for GHG emissions would also apply to other policy approaches to GHG emissions.
For instance, the majority tells us that GHG emissions are an international issue and that the federal government needs means of ensuring provincial cooperation in meeting targets agreed to in international agreements. If one province fails to take measures to address emissions, that could imperil the national objectives, and could lead to other problems like carbon leakage from provinces that do put a price on carbon to provinces that don’t. That’s a reasonably good explanation for why the federal government should have jurisdiction over the minimum price stringency of carbon emissions as a policy matter, but it’s also an explanation you could give for just about every other means of addressing carbon emissions.
Say governments, instead of using a price mechanism to reduce carbon emissions, chose instead to directly regulate conduct that leads to emissions, for example dictating what kinds of power plants can be built, what kind of cars can be driven, what kind of agricultural practices are allowed, what kinds of natural resources can be mined, etc. It would also be true that the failure to enact these policies nationally could imperil national emissions objectives, or could lead to leakage of carbon emitting activities from provinces that enact regulations to provinces that don’t enact regulations.
So we’re left asking what’s so special about minimum national standards of price stringency for carbon emissions? The majority clearly didn’t want to go all the way and recognize plenary federal jurisdiction over GHG emissions and so it found a way to restrict the scope of federal jurisdiction. But it seems to me that there isn’t actually a good justification for limiting federal jurisdiction in this way. There is instead a kind of arbitrary line drawn around one possible policy approach to the matter.
It would have been more coherent to say “Yes, the federal Parliament has jurisdiction over GHG emissions, or the cumulative impact of GHG emissions – including a range of policy tools it can use to address those emissions, beyond minimum standards of price stringency.” It also would have been more coherent to say, simply, “No, GHG emissions are not an appropriate POGG matter because of how far this would reach into provincial jurisdiction.” Instead we ended up with a kind of shaky middle ground where the federal government is restricted to using only the policy tool to address emissions that it happens to have adopted in this particular case.
The Dissent
The dissent, by Justices Ottenbreit and Caldwell, is interesting. In fact, I’m not sure I’ve ever read a judicial opinion quite like this. The dissenting opinion alone clocks in at around 90 pages. And I have to say there are some tangents that read more like an academic paper than a judicial decision. (And I mean that as a compliment!) The opinion at one point coins a new Latin phrase for the kind of tax at issue here. We are told this is best described as a carbo vestigium tax – AKA a carbon footprint tax. I’m not sure that coining new Latin terms is in keeping with the latest trends in plain-language legal writing, but it certainly made for an interesting read.
Where I think the dissent is weakest is on the question of determining whether the fuel levy is a tax or a regulatory charge. The dissent determines that it is a tax, and while I think that is a defensible outcome, the reasoning falls flat in a big way.
The dissent seems unable to accept the idea that one way of regulating conduct is to alter prices so as to indirectly alter behavior. In other words, if you make something more expensive, people will buy less of it. That is the essence of one of the core insights of economics, known as the law of demand. And it is, of course, the whole idea behind the fuel levy. The federal government isn’t generating any net revenue for general federal purposes; it’s giving the money back through rebates. The only reason the fuel levy exists is to attempt to alter people’s conduct in their purchasing decisions relating to carbon-emitting products.
For some reason, the dissenting judges seem unable to accept this. At paragraph 310, they write “it is manifest that the fuel levy adds nothing to any existing structure of regulation save revenue generation”. (emphasis added) I don’t think that’s a defensible statement. It flies in the face of the stated purpose of the levy, not to mention basic economic principles. At the very least, the dissenting judges are guilty of overstating their case here, saying the levy “adds nothing” except revenue generation.
My own view is that judges ought to be prepared to acknowledge that it is at least possible that by increasing the price of a product a government can seek to decrease the demand for that product. In their analysis, the dissenting judges give incredibly short shrift to the what is, to my mind, the only logical reason why the federal government would have implemented this policy in the first place – namely, to influence purchasing behaviour by increasing the price of carbon-emitting products.
I do think it’s still possible to get to the conclusion that this is a tax rather than a regulatory charge, even if one acknowledges that the levy could serve a valid regulatory purpose. So my issue here is more with the reasoning than with the result. The Westbank factors for distinguishing a tax from a regulatory charge are only indicia, after all, and the ultimate question is whether the primary purpose is generating revenue for general purposes rather than regulation. I think the better view is that the primary purpose here is regulatory, but I can see how reasonable people might take a different view, given that the fuel levy looks on its face an awful lot like an excise tax. What I don’t think is reasonable is saying that the levy serves no regulatory purpose at all.
Conclusion
So where does that leave us?
Tax Issues
Let’s start with tax issues. There is, I think, real ambiguity as to whether this is a tax, though as I have already said I think the majority got the better of this issue in characterizing it as a regulatory charge.
If it is a tax, the points made by the dissent about the scope of the delegation to the Governor in Council seem like they should carry some weight. The Act essentially delegates to the Governor in Council the power to change anything it wants about the fuel levy: the rates, what it applies to, who must pay it, and where in Canada it applies. A delegation of this nature comes close to making a mockery of the constitutional requirement that taxes originate in the House of Commons.
Moreover, if the levy is a tax, then it may raise interesting questions about Parliament’s authority to tax in a way that is designed to indirectly regulate matters under provincial jurisdiction. The dissenting opinion certainly saw that as a significant issue, though it didn’t have much authority to rely on.
POGG
If the measures are not valid exercises of Parliament’s taxation power, then we get to some really interesting POGG questions. The characterization of the POGG power in question by the majority seems difficult to justify in a principled way. If only the federal Parliament can adequately regulate the price stringency of GHG emissions, why wouldn’t the same arguments apply to other means of regulating GHG emissions? Shouldn’t there be a principled way of distinguishing the policy tools relating to GHG emissions that are under federal jurisdiction from those that are not?
The argument that there is a federal POGG power over GHG emissions generally, or their cumulative impacts, seems more logically coherent, but it has the risk of expanding federal power to a significant extent over matters traditionally under provincial jurisdiction.
I’m genuinely unsure what the SCC would do on the POGG question. I could see doctrinal traditionalists like Justices Brown and Côté rejecting a federal POGG power in this area outright, while other members of the court might be swayed by the emerging policy challenges of regulating GHG emissions.
Federalism
Finally, even if the regimes in question fall under federal taxation or POGG powers, there are also important arguments that can be made about how Parliament has chosen to exercise its authority. Questions arise particularly from the choice to make the regimes applicable in some provinces and not in others, depending on how those provinces have exercised their own legislative powers.
Conditional exercises of federal authority can be problematic in some cases, in that they can indirectly alter the formal division of powers through a kind of legislative coercion. Where Parliament effectively tells a province that it will have a burden imposed on it unless it conforms to federal policy, that can, at least in some cases, amount to a threat. In these cases, Parliament might be seen as using the threat to indirectly determine provincial policy, in a way that deprives provinces of meaningful autonomy.
This is an issue that is recognized in US federalism jurisprudence, most famously in the Obamacare litigation. The federal legislation at issue in that case threatened to take away all of a state’s federal Medicaid funding if it the state did not go along with the changes to the program in the Patient Protection and Affordable Care Act. In National Federation of Independent Business v. Sebelius, the US Supreme Court found that to be unconstitutional. While the opinion was split, Chief Justice Roberts’ lead opinion struck this part of the legislation down on the grounds that it was unduly coercive towards the states. The threat to take away all federal funding was said to deprive states of meaningful autonomy in administering their own programs.
I am not sure that the federal carbon tax regime rises of the level of coercion towards the provinces, but there are serious questions that can be asked about the limits of the federal taxation and spending powers, particularly in cases where they are exercised in a way seemingly designed to influence provincial action within areas of provincial jurisdiction.
This isn’t just about the unwritten principle of federalism seeking limit legislative action. At a certain point, there is a potential for creative uses of the federal taxation and spending powers to really fundamentally alter the structure of Canadian federalism and to undermine provincial autonomy by turning provinces into agents of federal policy. Again, I’m not sure we’re there yet in this case, but this is not a concern that should be easily dismissed.
Recent Comments