As someone who has been throwing around a lot of wild-eyed rhetoric about "godlike powers" recently, I should say a bit about an area where I think the Canadian courts should be a lot more active. Scott's post on Cuno provoked me to argue the courts should draw a harder line on excesses of both federal and state/provincial jurisdiction. I didn't call for "watertight compartments," but I came close.
Cuno involved a challenge to a special tax concession by Ohio to Daimler Chrysler to encourage it to manufacture jeeps there instead of neighbouring states. This kind of competitive corporate welfare has an obvious potential for a race to the bottom. The lower courts had found that this violated the "dormant commerce clause": in other words, it was constitutionally impermissible state protectionism. The SCOTUS punted the case on standing grounds, so we don't know whether the lower court's decision is right on the merits or not.
Scott, taking the traditional New Deal liberal view, likes the result of the lower court decision, but isn't really happy about federalism review of any kind. The problem seems to be line-drawing. Dormant commerce clause restrictions on the states are bad enough, but keeping the federal government out of areas that are not plausibly "interstate commerce" really drives him nuts. Scott's an expat Canadian of some kind, so we also exchanged views about the Judicial Committee of the Privy Council, and its generally pro-provincial jurisprudence (I'm pro, he's contra). He's probably got the bulk of legal academia on his side. Given the rarity with which the Canadian Supreme Court strikes anything down on federalism grounds (despite the absence of any clear doctrinal shift from the JCPC days), he probably speaks for the higher levels of the judiciary in this country too.
I think differently, but it's rude to set out controversial theories at length in other people's comment boxes, so I'll try to do it here. I don't claim to be original, but this exercise is valuable to me, and hardly anyone else will likely read it anyway.
One of the minimal reasons to enter a federal union is to have free trade and mobility of individuals among the sub-units. Section 121 of the British North America Act banned explicit tariffs, although the lack of an expansive reading of this clause or of the federal trade and commerce clause meant that we had the anomalous situation after 1988 of giving American (and then Mexican) companies better trade remedies for trade-restricting legislation by the government of Manitoba than Saskatchewan companies would have. Section 6 of the Charter provides for mobility, subject to a lot of exceptions and caveats. But for all the wimpiness of these provisions, they are pretty fundamental to the whole point of a federal union.
You are also going to want a common foreign/security policy. Optional is a thicker common citizenship -- basic civil, human and social rights that we want everyone in the federation to have.
The centralization of foreign policy will mostly take care of itself, as long as the sub-units are not recognized in foreign capitals and aren't allowed to have their own armies. Sure, provincial and state governments will likely have some cross-border functional relationships, but not being Andrew Coyne, I see nothing wrong with that.
But the common market and the common citizenship are going to require either judicial definition or a highly empowered central government to work. There will always be protectionist pressures within the sub-units and reasonably strong provincial/state governments at least potentially threaten meaningful citizenship rights, particularly of local minorities.
So far, we don't need judicial review, as long as we let the federal government keep protectionist and oppressive provincial/state governments in line.
But we chose a federal union precisely because we want to maintain a large amount of state/provincial sovereignty. Here we get into basic ideological commitments, but I have trouble arguing with the math. If 10 jurisdictions decide a controversial policy issue on the basis of majority vote, a lot more people are going to get the result they want than if it is decided at the higher level by majority vote. Moreover, minorities in a sub-unit have the option of exit, which minorities in the overall federation don't. In other words, if Quebec and Alberta both get to decide their own gun rules, then we have a lot fewer unhappy Quebecois/Albertans.
There are other benefits to powerful sub-units as well. Smaller jurisdictions are more accountable. It is easier to learn from other people's mistakes. And a division of sovereignty protects against tyranny. Having experienced both the federal and a provincial bureaucracy, I can tell you that the federal one is always worse. There is little real doubt that federal tax money is spent less effectively than provincial tax money.
These classic arguments for federalism point to the need to have powerful sub-units. But if the sub-units are doing something important, then they are doing something politically rewarding. And if it is politically rewarding, the politicians in the centre are going to want to have their hands on it. Paul Martin didn't want to talk about what he was doing for the Arctic or for migratory birds -- he wanted to talk about health care and education.
If no one polices the boundaries, then the tendency will be to have both levels of government legislating on everything and spending on everything. This can't be eliminated altogether, but it messes up accountability and reduces the ability of any jurisdiction to take a laissez-faire regulatory approach, since this will be cancelled out by the other.
So the courts need to take a role. They need to make sure that the federal power is not going where it shouldn't, and they need to make sure that the provinces respect the common market and common citizenship. The federal power should be limited to transactions where the costs are external to the province that benefits.
At this point, I'm just going to assert that the Privy Council, despite its formalism and aversion to evidence, filled this dual role better than the Supreme Court of Canada has. The SCC has sometimes had the clearer statements of doctrine: the "provincial inability" test is great, for example. But it hasn't really been willing to strike down significant laws. And since Hydro Quebec, the criminal law power has expanded to include almost everything. And the spending power seems immune from question.
On the other hand, the Court hasn't done much about provincial protectionism. Indeed, the provinces end up negotiating free trade deals with each other, as if they were completely sovereign entities, like the Agreement on Internal Trade, and the new BC-Alberta deal.
It isn't really anybody's fault. If you live in Ottawa, and are at the head of an important part of the federal government and are appointed by federal Prime Ministers, you can't help but see a lot of federal legislation as necessary, even though it isn't. And you don't really have the authority to keep the provinces in line either.