Yes, the Westminster system has a separation of powers
Our system connects the executive and the legislature, but it doesn’t fuse them into a single entity.
Canada’s Supreme Court recently reinforced that our system of government has a separation of powers doctrine. In a judgement last week, the Court found that the Ontario government could keep ministerial mandate letters confidential. The Court held that Cabinet confidence is essential for the functioning of the executive, just as parliamentary privilege is necessary for the legislature’s work. In an illuminating paragraph, Karakatsanis J. stated that:
in our Westminster system of government, the executive — like the judicial and legislative branches — also requires certain spheres of confidentiality to fulfill its constitutional role. Each of the executive, legislative branch, and judiciary play “critical and complementary roles in our constitutional democracy” and “each branch will be unable to fulfill its role if it is unduly interfered with by the others” (Ontario v. Criminal Lawyers’ Association of Ontario, 2013 SCC 43, [2013] 3 S.C.R. 3, at para. 29). Thus, constitutional conventions flow from the separation of powers and protect the spheres of confidentiality needed for a government institution “to perform its constitutionally-assigned functions” (British Columbia (Attorney General) v. Provincial Court Judges’ Association of British Columbia, 2020 SCC 20, [2020] 2 S.C.R. 506 (B.C. Judges), at para. 66). Just as legislative privilege protects the ability of elected representatives to act on the will of the people (Chagnon v. Syndicat de la fonction publique et parapublique du Québec, 2018 SCC 39, [2018] 2 S.C.R. 687), and deliberative secrecy preserves the independence of the judiciary (MacKeigan v. Hickman, [1989] 2 S.C.R. 796, at pp. 830-31), Cabinet confidentiality grants the executive the necessary latitude to govern in an effective, collectively responsible manner (Babcock v. Canada (Attorney General), 2002 SCC 57, [2002] 3 S.C.R. 3, at para. 15). Cabinet secrecy is “essential to good government” (ibid.), as it promotes deliberative candour, ministerial solidarity, and governmental efficiency by protecting Cabinet’s deliberations (B.C. Judges, at paras. 95-97; Carey v. Ontario, [1986] 2 S.C.R. 637, at pp. 658-59).
These findings align with earlier cases on parliamentary privilege, such as New Brunswick Broadcasting and Vaid, which held that the branches of the states must be able to fulfill their essential functions without improper interference from the other branches. Preventing the branches from autonomously performing their essential functions would undermine their “dignity” and “efficiency.” While the Canadian separation of powers doctrine isn’t akin to the stove-piped checking and balancing between the branches that we find in the United States, it does recognize that each branch has its own sphere of responsibility that the other shouldn’t muck around in.
What does this separation look like in practice? A good example are exercises of the Crown’s prerogative powers by the executive. When the executive deploys armed forces overseas, it does so under the authority of the prerogative - the powers that the Crown enjoys in its own right - as recognized by common law. Parliament could displace this prerogative with statute, but it hasn’t. Until the law is changed, the executive can deploy armed forces internationally without consulting Parliament. It also means that Parliament can’t prevent the executive from exercising the prerogative with through motions alone. If Parliament is determined to stop a military deployment, the House of Commons can withdraw confidence in the government, or the houses can try to legislate controls on the prerogative. Yet even these options have limits. The executive can still legally exercise the prerogative for military deployments after confidence has been withdrawn, and the military will be well underway by the time legislation to control the prerogative gets royal assent –if it ever does.
The same holds for executive discretion provided by statute. If Parliament has granted ministers discretionary powers, the legislature can’t suddenly revoke them through a simple vote, no matter how intensely the parliamentarians feel about a decision made under that statutory authority. If the executive has the discretionary authority to act under statute, then Parliament must change the law if it wishes to constrain the government.
Again, the fact that Parliament can legislate away the executive’s discretion shows that we don’t have an American-style separation of powers doctrine. But the process that Parliament must follow to bind the government demonstrates that the mere presence of the ministers in the legislature does not negate the boundaries that exist between the branches. While ministers may choose to back away from a discretionary decision if they face significant opposition in Parliament, it’s important to recognize that this involves a political choice, not a legal obligation.
There’s much more to say here, but I’ll stick to three concluding observations.
First, we haven’t really tested the boundaries of the executive’s constitutional responsibilities in Canada. As the Supreme Court hinted at in Khadr, there may be aspects of government that are so closely connected with the constitutional functions of the executive that they would be protected from undue statutory interferences thanks to sections 9 and 15 of the Constitution Act 1867. Following the logic of Khadr, for instance, it’s possible that the executive power to conduct foreign affairs is afforded a measure of constitutional protection from regular statutes owing to section 9. This would also align with the Court’s emphasis on the need for each branch to perform their core functions with dignity. Given the executive’s dominant role in the legislative process in Canada, we’ll likely never see a case that addresses whether sections 9 and 15 shield certain executive powers from statutory intrusions, but the Court’s language in cases involving the separation of powers suggests that it’s possible.
Secondly, as the Supreme Court’s majority judgement in Mikisew makes clear, the fact that ministers sit in Parliament does not erase the boundaries between the legislature and the executive in law. Bagehot’s observation that the legislature and the executive are fused in Westminster systems refers to the operation of the political constitution, not the legal one. Fusion-talk should be set aside when we’re dealing with legal questions.
Finally, I was motivated to write this post by recent events in Britain, where conservatives have insisted that there is no separation of powers in the United Kingdom. This insistence has come up in light of government legislation seeking to change the Supreme Court’s finding that Rwanda is not a safe country for refugee deportations. Critics of the government have argued that using legislation to make Rwanda a safe country violates the separation of powers between Parliament and the courts. It’s in this context that conservatives have denied that the United Kingdom has a separation of powers.
While I get what the conservatives are saying, it’s not that clear cut. As we saw in 2019, the British constitution doesn’t work well when the lines between the branches are blurred. In that instance, Parliament attempted to usurp executive decisions and the Fixed-Term Parliaments Act prevented the government from dissolving the legislature in response. This led to a complete shambles that required a circumvention of the Fixed-Term Parliaments Act to fix. Conservatives who deny that there is a separation of powers in Britain would do well to remember this episode. The United Kingdom isn’t the United States, but that doesn’t mean that the three branches don’t have core functions that they should be permitted to fulfill with a degree of autonomy and deference.