Blog of the International Journal of Constitutional Law

Taking Aim at Cooperative Federalism: The Long-Gun Registry Decision by the Supreme Court of Canada

Johanne Poirier[*], Université libre de Bruxelles 

On March 27, 2015, a highly divided Supreme Court of Canada rendered a momentous ruling which reveals a severe divergence on the nature of Canadian contemporary federalism.[1]  The tight 5 to 4 decision also underlines a different conception of the role of the judicial branch regarding the defence and promotion of cooperative behaviour by public authorities, principally the legislative branch, under Canadian constitutional law.

The majority judges ruled that pursuant to its criminal law power, the federal Parliament could not only decriminalise the possession of long-guns, but that it could also destroy all data relative to the pre-existing registration of those guns.  The decision raised the question whether a legislative assembly could unilaterally abrogate any provision it has previously – and constitutionally – adopted, when the programme at stake was designed partly in association with other orders of government.

1. Three dimensions of cooperative federalism

Over the last decade, “cooperative federalism” has entered the discourse of the Supreme Court, in a way that partly contradicts the original structure of the Canadian federation, which is largely “dualist”.   Hence, with very few exceptions, the 1867 Constitution Act grants each order of government exclusive competences over listed matters.  Moreover, each order is endowed with parallel and complete legislative, executive, and to a large extent judicial, powers and institutions, through which they are meant to act in an independent fashion. The evocation of “cooperative federalism” has had two distinct impacts on this traditional conception of federalism.

The first has been the recourse to a number of interpretative doctrines that have resulted in increasing legislative overlap between federal and provincial authorities. The dominant case law of the Supreme Court has described this gradual transformation as the recognition of “modern” and “flexible” federalism.[2]  One of the consequences of this lowering of  “picket fences” around exclusive spheres of competences has been the necessity for federal partners to negotiate “who should do what”, since both can often constitutionally intervene regarding interrelated policy domains.

The second impact of this jurisprudential endorsement of “cooperative federalism” is less visible, but just as significant.  It has led judges to refrain from finding “technical” (meaning legal) obstacles that could stand in the way of cooperation between orders of government. The latter seek to limit administrative duplication, to coordinate their actions or to delegate functions to one another to improve the efficiency of public management.  Cooperative arrangements to that effect are often grounded on complex combinations of federal and provincial laws and regulations, intergovernmental agreements, soft law instruments and simple cooperative practices without any legal grounding.

Unfortunately, these normative networks are difficult to reconcile with the official structures of parliamentary control and judicial review, which, in Canada, are still based on a “dualist” model of federalism.  In other words, parliaments and courts still control actions of “their” order, as if each was acting in a totally independent fashion.  The Supreme Court has been incredibly tolerant of these interwoven schemes, even when they have the effect of blurring lines of political or administrative accountability and transparency.[3]  In other words, judges have turned a blind eye to traditional rules of administrative law so as not to obstruct cooperation between orders of government.

In short, in recent years, the Supreme Court has largely moved away from “picket fence” federalism, and sought to facilitate intergovernmental cooperation.  When governments agree, courts applaud and sit back. Sometimes at the cost of legal clarity and judicial protection for citizens.

The “Long-Gun Registry Case” raised a third dimension of cooperative federalism. It concerns the rules of the federal game not when partners agree, but when they disagree, or no longer agree about an existing cooperative scheme, or when one order seeks to unilaterally put an end to a coordinated policy regime.

In a number of federal (or quasi-federal) systems, norms of “good federal conduct” enable courts to require the other branches to act according to some form of “federal loyalty” in their dealings with each other.  This is the case of Germany, Switzerland, Belgium, South Africa and to a certain extent the European Union.  In addition to their responsibility for tracing (or erasing) the borders between constitutional competences, such a obligation of good faith conveys upon judges a power of oversight over the manner in which competences are exercised.

Endorsing this third dimension of cooperative federalism would bring the Canadian federal system closer to those which are officially and structurally « cooperative » and in which a number of legal instruments and institutions provide a more explicit legal framework for intergovernmental cooperation. It is here that the majority and the dissenting judges of the Supreme Court part ways in the “Long-Gun Registry” case.  The former seeks to limit the reach of the Court’s previous appeal to “cooperative federalism”. The latter draw conclusions from previous case law, to deepen its meaning.

2. Cooperative Federalism and Gun Control

Gun control has a long history in Canada.  In 2000, the Supreme Court held that pursuant to its competence over criminal law, the federal legislator could require that all long-guns (rifles) in Canada be registered.[4]  The Court did so without excluding the possibility that provinces could also, pursuant to their own constitutional competences over private property, keep a registry of long-gun ownership.  In this context, federal authorities did legislate, but offered to provinces that wanted to, the option of collaborating in the gun-control scheme. Five provinces, including Québec – which had been at the forefront of the call for the extension of gun-control – chose to do so.  It is highly probable that had the federal authorities not acted, Québec would have, in the 1990s, legislated to regulate long-gun possession.

In the 2015 “Long-Gun Registry case”, the Supreme Court had to rule on the constitutionality of a federal legislative provision which ordered both federal and provincial civil servants to destroy all data related to the registration of long-guns (rifles) in Canada. This provision was (belatedly) included in a federal Act that abrogated the obligation to register the possession of long-guns, which had been in force since 1995.[5]  While Québec did not challenge the competence of the federal Parliament to abolish the registry, it sought to obtain the data in order to start its own registry.  No one denied that Québec had the constitutional competence to do so. The question was whether it could use the information found in the registry that the federal order was abolishing.  The federal government refused to transfer the data, arguing that as the possession of unregistered rifles was no longer criminal, the data concerning previous registration violated the right to privacy of lawful gun owners.

A major part of the divergence between the majority and minority opinions rests on the interpretation of the degree of legislative and administrative integration of the federal and provincial orders in this aspect of gun-control.

The majority held that the intricate scheme was designed under federal law, and that provincial agents were essentially “agents” of the federal order. It also held that although federal and provincial actors cooperated in the overall gun-control scheme, the actual collection of data regarding the registration of long-guns was a purely federal affair. Consequently, the dismantling of the registry – and the destruction of the data it contained – was a constitutionally valid exercise of the sovereignty of the federal Parliament.

By contrast, 4 dissenting judges – including all 3 judges from Québec – considered the scheme to be far more integrated.  They examined actual coordinated action and a number of statements made by federal and provincial public actors to conclude that the entire scheme was clearly inter-governmental.  If the federal Parliament could abolish the registry, it could only do so after having offered the information to provinces who may want it (to create their own registry, for instance).[6]

3. Cooperative Federalism = 0; Classical Parliamentary Sovereignty = 1

The principle of parliamentary sovereignty has, from the start, been a pillar of dualist federalism in Canada. To the extent that they respect fundamental rights and the formal distribution of competences, legislatures may basically adopt any law they see fit.  And abrogate any legislative provisions they have enacted. The majority decision reinforces a traditional conception of parliamentary sovereignty, which is difficult to reconcile with modern-day, intertwined federal practice.  Basically, what a legislative assembly can do, it can undo. Full stop.  And this – a priori – regardless of the impact of such action on the other orders of government.[7]

This Diceyan conception of parliamentary sovereignty was invoked in isolation, without any nuanced balancing with other unwritten constitutional principles, including the principle of federalism, as intimated by the Supreme Court in the 1998 Quebec Secession Reference.[8]  Instead, the majority revived, without much analysis, the reasoning a 1991 decision which had basically held that parliamentary sovereignty (both federal and provincial) was only limited by the protection of human rights and the formal distribution of competences.[9]

While not explicitly curbing parliamentary sovereignty, the minority opinion calls for an interpretation of the scope of legislative competences to take into consideration the interests of other orders. This is reminiscent of a “proportionality” approach to distribution of competences, which require judges to assess both the actual necessity of a challenged provision for the overall coherence of a legislative objective and the degree of a legislator’s intrusion into another’s sphere of competences. Here, the minority judges are of the opinion that the federal destruction of the data, without any offer to provinces that may want to use it, was not crucial to the abolition of the registry and actually prevented Québec from properly exercising its own competences.


So far, the dominant endorsement of “cooperative federalism” by the Supreme Court has led to increasing legislative overlap and to judicial restraint in the face of interlocking cooperative schemes. With the Long-Gun Registry decision, the future of “cooperative federalism” is uncertain.  When governments agree, courts will curb legal obstacles to their cooperation. Courts will even invite the other branches to cooperate.  But when governments no longer agree, the majority judges in the 2015 Long-gun Registry case revert back to a traditional “dualist” and positivist conception of public law in which individual governments can act in a unilateral fashion, at least to the extent that they “use” the legislative assemblies (which they generally control, under Canadian parliamentary and electoral law).

The dissenting judges, by contrast, have sought to deepen the meaning of “cooperative federalism”. Having promoted concerted action between orders, and having lowered the “picket fences” which defined the original Canadian federal system, they recognise that the judicial branch cannot logically slide back to a traditional dualist conception of federalism.  This recognition partly leads to unchartered territory.  The minority position is prudent, nuanced, and admittedly not always as legible as the majority, far more traditional, position. It is, however, in line both with contemporary federal practice, and with jurisprudential development.

Suggested Citation: Johanne Poirier, Taking Aim at Cooperative Federalism: The Long-Gun Registry Decision by the Supreme Court of Canada, Int’l J. Const. L. Blog, Apr. 15, 2015, at:

[*] B.A. (Queen’s) ; LL.B and D.C.L. (McGill) ; LLM (Université libre de Bruxelles); D.Phil (Cambridge).  Professor of comparative public law, Université libre de Bruxelles. As of August 2015, Johanne Poirier will occupy the MacKell Chair in Federalism at the Faculty of Law, McGill University.  Comments welcome at

[1] Quebec (Attorney General) v. Canada (Attorney General), 2015 SCC 14: available at

[2] Canadian Western Bank v. Alberta, [2007] 2 S.C.R. 3, par. 24-42 and 77-78 ; Reference re Assisted Human Reproduction Act, [2010] 3 S.C.R. 457, par. 61 ; Canada (Attorney General) v. PHS Community Services Society, [2011] 3 S.C.R. 134, par. 63 ; Tsilhqot’in Nation v. British Columbia, [2014] 2 S.C.R. 256, par. 148-149.

[3] On this, see POIRIER, Johanne, « Une source paradoxale du droit constitutionnel canadien: les ententes intergouvernementales », Revue québécoise de droit constitutionnel, 2009, vol.1, 20 pp. ( and POIRIER, Johanne, « Intergovernmental Agreements in Canada : at the Cross-roads Between Law and Politics », in MEEKISON, Peter J., TELFORD, Hamish, LAZAR, Harvey (eds), Reconsidering the Institutions of Canadian Federalism, Canada: the State of the Federation 2002, Institute of Intergovernmental Relations, Kingston (Canada), 2004, pp. 425-462, at pp. 452 ss.

[4] Reference re Firearms Act (Can.), [2000] 1 S.C.R. 783.

[5] The registry was created under a Liberal federal government. Its abrogation had been part of the Conservative Party electoral platform, which it implemented soon after its election: Firearms Act, S.C. 1995, c. 39.

[6] In a rather nuanced – but somewhat convoluted way – the minority concluded that if the federal order could not destroy the data without first offering them to the provinces, the latter did not actually have legal title to the data, and could not thus judicially request them … The actual transfer was subject to political negotiations and did not rest on legal rights.

[7] See the caveat by the majority that its reasoning may not apply to a « truly interlocking federal-provincial legislative framework” (par. 4). The dissenting judges were convinced that the scheme was sufficiently interlocking in this particular case to limit unilateral action by federal authorities.

[8] In the Quebec Secession Reference, [1998] 2 S.C.R. 217, the Supreme Court of Canada identified four underlying unwritten principles which ground the constitutional framework and provide life to the written constitutional text: democracy, federalism, constitutionalism and the rule of law, and the protection of minorities.

[9] Reference re Canada Assistance Plan (B.C.), [1991] 2 S.C.R. 525.


One response to “Taking Aim at Cooperative Federalism: The Long-Gun Registry Decision by the Supreme Court of Canada”

  1. Soraya Beukes Avatar
    Soraya Beukes

    Canada has a bicameral parliament consisting of the Senate and the house of Commons. Both have legislative competencies. The Senate although not elected represent the regions. It is noted that the Senate rarely opposes Bills originated from the House of Commons. This is regrettable because the interest of the regions are thus unprotected by those that need to protect their interest. So much so if the Senate played their part at looking after the interest of its regions they could have raised objection to the scrapping of the long gun laws and the distruction of the register. The Senate would have served it purpose by suggesting the offering of the information to its regions in their respective interest. This would have circumvented the whole issue of having the court decide on whether the register information on long guns be afforded to a region that poses interest in the matter as such. Moreover this positivist approach by the court does not bode well for cooperative governance. This approach prevails parliamentary sovereignty and by large is a backward leap from latter day federalism which seeks to transfer power closer to the electorate, that is the regions.

Leave a Reply

Your email address will not be published. Required fields are marked *