Quebec (Attorney General) v. Canada (Attorney General), 2015 SCC 14, [2015] 1 SCR 693

  

 

SUPREME COURT OF CANADA

 

Citation: Quebec (Attorney General) v. Canada (Attorney General), 2015 SCC 14, [2015] 1 S.C.R. 693

Date: 20150327 Docket: 35448

 

Between:

Attorney General of Quebec

Appellant

and

Attorney General of Canada,

Commissioner of Firearms and

Registrar of Firearms

Respondents

- and -

Chief Firearms Officer,

Coalition for Gun Control and

Canada’s National Firearms Association

Interveners

 

Official English Translation: Reasons of LeBel, Wagner and Gascon JJ.

 

Coram: McLachlin C.J. and LeBel, Abella, Rothstein, Cromwell, Moldaver, Karakatsanis, Wagner and Gascon JJ.

 

Joint Reasons for Judgment: (paras. 1 to 46)   Joint Dissenting Reasons: (paras. 47 to 203):

Cromwell and Karakatsanis JJ. (McLachlin C.J. and Rothstein and Moldaver JJ. concurring)   LeBel, Wagner and Gascon JJ. (Abella J. concurring)

 

 

 

Quebec (Attorney General) v. Canada (Attorney General), 2015 SCC 14, [2015] 1 S.C.R. 693

Attorney General of Quebec                                                                          Appellant

v.

Attorney General of Canada,

Commissioner of Firearms

and Registrar of Firearms                                                                         Respondents

 

and

Chief Firearms Officer,

Coalition for Gun Control and

Canada’s National Firearms Association                                                   Interveners

Indexed as: Quebec (Attorney General) v. Canada (Attorney General)

2015 SCC 14

File No.: 35448.

2014: October 8; 2015: March 27.

Present: McLachlin C.J. and LeBel, Abella, Rothstein, Cromwell, Moldaver, Karakatsanis, Wagner and Gascon JJ.

on appeal from the court of appeal for quebec

                    Constitutional law — Division of powers — Criminal law — Constitutional classification of repealing enactment — Firearms — Federal legislation abolishing long-gun registry also containing provision requiring destruction of long-gun registration data — Quebec challenging constitutionality of destruction provision and seeking transfer of data connected with province from federal government — Whether principle of cooperative federalism prevents Parliament from legislating to destroy data — Whether destruction provision ultra vires criminal law power of Parliament — Whether Quebec has right to receive long-gun registration data from federal government — Constitution Act, 1867, s. 91(27)  — Ending the Long-gun Registry Act, S.C. 2012, c. 6, s. 29 .

                    Adopted in 1995, the Firearms Act  created a comprehensive scheme requiring the holders of all firearms — including long guns — to obtain licences and register their guns. It also made it a criminal offence to possess an unregistered firearm. The Firearms Act  provided for the creation of two types of registries: the Canadian Firearms Registry (“CFR”), maintained by the Registrar of Firearms and containing records of the registration certificates for all prohibited firearms, restricted firearms, and long guns acquired, transferred, or possessed in Canada, and a registry kept by the Chief Firearms Officer (“CFO”) designated for each province and territory, containing records of every firearm’s licence and authorization issued or revoked. The Registrar and the CFOs could access all records through a single electronic database but the statutory authority of CFOs only permitted them to contribute and modify data in their specific licensing registry.

                    In 2012, Parliament enacted the Ending the Long-gun Registry Act (“ELRA”), which repealed the registration requirement for long guns and decriminalized the possession of an unregistered long gun. Section 29 of the ELRA requires the destruction of all records contained in the registries related to the registration of long guns. In reaction, Quebec expressed its intention to create its own long-gun registry and asked the federal authorities for the data connected to Quebec contained in the CFR. Canada refused and made clear that it intended to permanently destroy all long-gun registration data. In light of this refusal, Quebec sought a declaration that s. 29 of the ELRA is ultra vires and that Quebec has a right to obtain the data.

                    The Superior Court of Quebec declared s. 29 of the ELRA unconstitutional as it applies to data connected with Quebec and ordered Canada to transfer that data to the province. The Quebec Court of Appeal reversed that decision.

                    Held (LeBel, Abella, Wagner and Gascon JJ. dissenting): The appeal should be dismissed. Section 29 of the ELRA is constitutional, and Quebec has no legal right to the data.

                    Per McLachlin C.J. and Rothstein, Cromwell, Moldaver and Karakatsanis JJ.: The decision to dismantle the long-gun registry and destroy the data that it contains is a policy choice that Parliament was constitutionally entitled to make. The principle of cooperative federalism does not constrain federal legislative competence in this case, Quebec has no legal right to the data, and s. 29 of the ELRA is a lawful exercise of Parliament’s criminal law legislative power under the Constitution.

                    Quebec’s position that cooperative federalism prevents Canada and the provinces from acting or legislating in a way that would hinder cooperation between both orders of government has no foundation in our constitutional law and is contrary to the governing authorities from this Court. The principle of cooperative federalism does not prevent Parliament from exercising legislative authority that it otherwise possesses. The primacy of our written Constitution remains one of the fundamental tenets of our constitutional framework, and this is especially the case with regard to the division of powers. Neither this Court’s jurisprudence nor the text of the Constitution Act, 1867  supports using the principle of cooperative federalism to limit the scope of legislative authority or to impose a positive obligation to facilitate cooperation where the constitutional division of powers authorizes unilateral action. To hold otherwise would undermine parliamentary sovereignty and create legal uncertainty whenever one order of government adopted legislation having some impact on the policy objectives of another.

                    Although Quebec submits that it has a right to receive the long-gun registration data whether or not Parliament is constitutionally entitled to legislate with respect to the fate of that data, it has not established a legal basis for that right. As mentioned, the principle of cooperative federalism does not limit the scope of the legislative powers assigned by the Constitution. Furthermore, accepting Quebec’s position, which arises from its expectation of having continuing access to the data, would circumvent or effectively overturn this Court’s rejection of the “legitimate expectation” doctrine. The provinces’ reliance on the existence of the data cannot limit Parliament’s capacity to destroy a registry, which flows exclusively from its criminal law head of power. Lastly, even if the data accessible through the CFR was the result of a cooperative effort, any effort on Quebec’s part was statutorily limited to the licensing data held in the CFO’s licensing registry.

                    This Court has already been called upon, in the Reference re Firearms Act (Can.), 2000 SCC 31, [2000] 1 S.C.R. 783, to determine the pith and substance of the scheme enacted by the Firearms Act . In that case, the Court concluded that the “matter” of the registration and data retention provisions was public safety and should be classified as being in relation to the subject of criminal law. Legislation repealing that scheme, including a provision addressing what will happen to the data collected under the now repealed scheme, must be characterized in the same way. Section 29 , in essence, relates to public safety — as did the long-gun registration scheme being repealed by the balance of the ELRA. That provision does not limit Quebec’s legislative authority to create a provincial long-gun registry, it merely prevents Quebec from using the data obtained through the federal long-gun registry in establishing a provincial registry. The fact that it has the practical effect of making it more difficult financially for Quebec to create its own gun control regime is not indicative of a “colourable” purpose from a division of powers’ perspective and does not affect the pith and substance of s. 29 .

                    There is no significant legal distinction between repealing a criminal provision and providing for what will happen to the data collected under that provision where the data was collected exclusively through the exercise of the criminal law power. The power to repeal a criminal law provision must logically be wide enough to give Parliament jurisdiction to destroy the data collected for the purpose of that provision. Accordingly, s. 29 of the ELRA should be characterized as being in relation to criminal law. It therefore falls within the legislative competence of Parliament.

                    Per LeBel, Abella, Wagner and Gascon JJ. (dissenting): Section 29 of the ELRA is unconstitutional and should be declared to be invalid. The ELRA is the legislative measure chosen by Parliament to end its participation in long-gun regulation, but s. 29  goes beyond the scope of that purpose, as it requires that the data in question be destroyed without providing for a possibility of their first being transferred to the provincial partners, which prevents the latter from using them in the exercise of their powers. However, there is no legal basis for Quebec’s request for a compulsory transfer of the data. The conditions applicable to such a transfer are a matter for the governments concerned, not the courts.

                    When the constitutionality of a statutory provision is challenged on the basis of the division of powers, courts turn to the pith and substance doctrine. To apply this doctrine, they must review the extent to which the impugned provision intrudes on the powers of the other level of government. Where, because of its pith and substance, a provision found in an otherwise valid statute encroaches on the jurisdiction of the other level of government, it must be determined whether the encroachment is ancillary. The degree of integration of a provision that is needed for an encroachment to be considered ancillary varies with the seriousness, or extent, of the encroachment. If the encroachment of the impugned provision on the jurisdiction of the other level of government is merely marginal or limited, a functional relationship between the provision and the statutory scheme may suffice. If, on the other hand, the provision is highly intrusive vis-à-vis the powers of the other level of government, a stricter test of necessity will apply.

                    The unwritten principles that underlie our written Constitution, such as federalism, infuse the analysis and interpretation of the division of powers. The modern view of federalism favours a flexible conception of the division of powers and recognizes a significant overlap between the federal and provincial areas of jurisdiction, allowing governments at both levels to legislate for valid purposes in the areas of overlap. Such a conception facilitates intergovernmental co-operation. Both in law and in the political arena, the concept of co-operative federalism has been developed to adapt the principle of federalism to this modern reality; it reflects the realities of an increasingly complex society that requires the enactment of co-ordinated federal and provincial legislative schemes. From a legal perspective, it is by allowing for overlapping powers through the application of the pith and substance and ancillary powers doctrines that co-operative federalism is able to meet those needs.

                    In this case, the trial judge was right to find that there was a federal-provincial partnership with respect to firearms control. This partnership is consistent with the spirit of co-operative federalism. It enabled the federal and provincial governments to work together, rather than in isolation, to achieve both federal (criminal law) and provincial (public safety and administration of justice) purposes. In the novel circumstances of the dismantling of this partnership, the analysis must be guided by the Constitution’s unwritten principles so as to ensure that the principle of federalism and its modern form — co-operative federalism — are not placed in jeopardy. Parliament or a provincial legislature cannot pass legislation to terminate such a partnership without taking into account the reasonably foreseeable consequences of the decision to do so on its partner’s heads of power. The courts must, in considering whether legislation or a statutory provision having as its purpose to dismantle the partnership is constitutional, be aware of the impact of that legislation or provision on the other partner’s exercise of its powers, especially when the partner that terminates the relationship is intentionally bringing about that impact.

                    The Court of Appeal strayed from the analytical approach the courts must take. It is true that Parliament can repeal or amend legislation it has validly enacted under one of its heads of power. Nevertheless, the courts must consider the impugned provision or legislation to determine whether, in pith and substance, all that it does is in fact to repeal or amend legislation that was validly enacted. It is not enough to say that the legislative measure is merely repealing legislation.

                    The words of s. 29 of the ELRA pose no particular problems of interpretation. From a structural standpoint, s. 29  is distinct from other sections of the ELRA because it is a transitional provision. From a practical and legal standpoint, the principal effect of s. 29  is to delete the data in the CFR forever. The federal government’s decision to destroy the data without first transferring them to its partners, such as Quebec, has serious consequences that are relevant to the question whether s. 29  is constitutional. The extrinsic evidence shows that the purpose being pursued in enacting s. 29  was indeed to prevent the provinces from using the data. The trial judge was therefore right to find that Parliament’s intention in destroying the data was to hinder the provinces. In light of the purpose and the effect of s. 29 , therefore, the scope of the section is broader than the mere destruction of the data; it has harmful consequences for the federal government’s partners. The purpose of s. 29  does not relate to the repeal of part of the Firearms Act ; the abolition by the federal government of the requirement to register long guns and the destruction of the data are two distinct objectives.

                    Given that the data are to be destroyed with no possibility of their first being transferred to the partners, and therefore without the impact of this measure on the partners’ exercise of their powers being taken into account, the section’s true purpose is to ensure that the information on long guns can no longer be used for any provincial purposes. As a result, the pith and substance of s. 29  relates to the provinces’ power over property and civil rights.

                    To determine whether s. 29 of the ELRA is constitutional on the basis of the ancillary powers doctrine, the seriousness, or extent, of its encroachment on provincial powers must be considered, bearing in mind that the provincial power to make laws in relation to property and civil rights is a head that should not be intruded upon lightly. The seriousness of the encroachment of s. 29  must be analyzed on the basis of the specific factual and legal context of the case, which includes the existence of the partnership between the federal government and Quebec. In this case, in terms of both its nature and its effect, s. 29  causes a substantial encroachment on provincial jurisdiction. For its encroachment to be found to be ancillary to the ELRA, the degree to which s. 29  is integrated into the Act must therefore be high, that is, it must satisfy the necessity or “integral part” criterion. The destruction of the data in question in s. 29  cannot be considered necessary to the abolition of the requirement to register long guns, as these two purposes are distinct. Nor can s. 29  be linked to the ELRA on the basis of a test of rationality; it is hard to reconcile the manner in which the destruction of the data was provided for with the desire certain provinces might show to maintain a registry within the limits of their powers. Furthermore, Parliament declared that its intention was to cause harm to the other level of government.

                    Since, because of its pith and substance, s. 29 of the ELRA does not fall within the federal criminal law power and is not ancillary to the ELRA, it has not been shown to be constitutionally valid. A legislative measure cannot be found to be valid that (1) does not fall within the federal criminal law power and that (2) thwarts, by the substantial encroachment it causes, the corollary exercise of provincial powers that flowed from the partnership. To destroy the data without first offering to transfer them is unconstitutional. Section 29 of the ELRA must therefore be declared to be invalid under s. 52  of the Constitution Act, 1982 .

                    Nevertheless, Quebec has not established a legal basis for its claim to the data. The absence of a legal barrier to the transfer of the data does not necessarily mean that Quebec has proven that it is entitled to obtain them through the courts. It is up to the legislatures to fill legislative gaps that are incompatible with the Constitution, and not up to the courts to supply an exact description of the laws the legislatures must adopt to fulfill their constitutional obligations. In some cases, the source of the appropriate remedy must lie in the political process rather than in the courts. In this case, it was up to the members of the partnership to set out the conditions that were to apply upon termination of their joint venture in their agreements.

Cases Cited

By Cromwell and Karakatsanis JJ.

                    Considered: Reference re Firearms Act (Can.), 2000 SCC 31, [2000] 1 S.C.R. 783; referred to: Canadian Western Bank v. Alberta, 2007 SCC 22, [2007] 2 S.C.R. 3; Reference re Agricultural Products Marketing Act, [1978] 2 S.C.R. 1198; Fédération des producteurs de volailles du Québec v. Pelland, 2005 SCC 20, [2005] 1 S.C.R. 292; Quebec (Attorney General) v. Canadian Owners and Pilots Association, 2010 SCC 39, [2010] 2 S.C.R. 536; Canada (Attorney General) v. PHS Community Services Society, 2011 SCC 44, [2011] 3 S.C.R. 134; Reference re Secession of Quebec, [1998] 2 S.C.R. 217; Reference re Anti-Inflation Act, [1976] 2 S.C.R. 373; Reference re Securities Act, 2011 SCC 66, [2011] 3 S.C.R. 837; Reference re Canada Assistance Plan (B.C.), [1991] 2 S.C.R. 525; Wells v. Newfoundland, [1999] 3 S.C.R. 199; Authorson v. Canada (Attorney General), 2003 SCC 39, [2003] 2 S.C.R. 40; RJR-MacDonald Inc. v. Canada (Attorney General), [1995] 3 S.C.R. 199; Quebec (Attorney General) v. Lacombe, 2010 SCC 38, [2010] 2 S.C.R. 453; Kitkatla Band v. British Columbia (Minister of Small Business, Tourism and Culture), 2002 SCC 31, [2002] 2 S.C.R. 146; Ward v. Canada (Attorney General), 2002 SCC 17, [2002] 1 S.C.R. 569; Kirkbi AG v. Ritvik Holdings Inc., 2005 SCC 65, [2005] 3 S.C.R. 302; Reference re Upper Churchill Water Rights Reversion Act, [1984] 1 S.C.R. 297; Attorney-General for Ontario v. Attorney-General for Canada, [1912] A.C. 571; Attorney-General for Canada v. Attorney-General for Ontario, [1937] A.C. 326; Reference re Same-Sex Marriage, 2004 SCC 79, [2004] 3 S.C.R. 698; Hunt v. T&N plc, [1993] 4 S.C.R. 289.

By LeBel, Wagner and Gascon JJ. (dissenting)

                    Reference re Firearms Act (Can.), 2000 SCC 31, [2000] 1 S.C.R. 783; General Motors of Canada Ltd. v. City National Leasing, [1989] 1 S.C.R. 641; Reference re Upper Churchill Water Rights Reversion Act, [1984] 1 S.C.R. 297; Kirkbi AG v. Ritvik Holdings Inc., 2005 SCC 65, [2005] 3 S.C.R. 302; Reference re Securities Act, 2011 SCC 66, [2011] 3 S.C.R. 837; Quebec (Attorney General) v. Lacombe, 2010 SCC 38, [2010] 2 S.C.R. 453; Kitkatla Band v. British Columbia (Minister of Small Business, Tourism and Culture), 2002 SCC 31, [2002] 2 S.C.R. 146; Reference re Same-Sex Marriage, 2004 SCC 79, [2004] 3 S.C.R. 698; Reference re Assisted Human Reproduction Act, 2010 SCC 61, [2010] 3 S.C.R. 457; Canadian Western Bank v. Alberta, 2007 SCC 22, [2007] 2 S.C.R. 3; Attorney-General for Ontario v. Reciprocal Insurers, [1924] A.C. 328; R. v. Big M Drug Mart Ltd., [1985] 1 S.C.R. 295; R. v. Edwards Books and Art Ltd., [1986] 2 S.C.R. 713; Reference re Secession of Quebec, [1998] 2 S.C.R. 217; Canada (Attorney General) v. PHS Community Services Society, 2011 SCC 44, [2011] 3 S.C.R. 134; OPSEU v. Ontario (Attorney General), [1987] 2 S.C.R. 2; Tsilhqot’in Nation v. British Columbia, 2014 SCC 44, [2014] 2 S.C.R. 257; Reference re Canada Assistance Plan (B.C.), [1991] 2 S.C.R. 525; Northern Telecom Ltd. v. Communications Workers of Canada, [1980] 1 S.C.R. 115; Clark v. Canadian National Railway Co., [1988] 2 S.C.R. 680; R. v. Thomas Fuller Construction Co. (1958) Ltd., [1980] 1 S.C.R. 695; Doucet-Boudreau v. Nova Scotia (Minister of Education), 2003 SCC 62, [2003] 3 S.C.R. 3; R. v. Prosper, [1994] 3 S.C.R. 236; Hunter v. Southam Inc., [1984] 2 S.C.R. 145; Reference re Public Schools Act (Man.), s. 79(3), (4) and (7), [1993] 1 S.C.R. 839.

Statutes and Regulations Cited

Act respecting safety in sports, CQLR, c. S-3.1, s. 46.31.

Act to protect persons with regard to activities involving firearms, CQLR, c. P-38.0001, s. 11.

Code of Civil Procedure, CQLR, c. C-25, art. 778.

Constitution Act, 1867, ss. 91(27) , 92(13) .

Constitution Act, 1982, s. 52 .

Criminal Code, R.S.C. 1985, c. C-46, ss. 90 , 91 .

Criminal Law Amendment Act, 1968-69, S.C. 1968-69, c. 38, s. 6.

Ending the Long-gun Registry Act, S.C. 2012, c. 6, ss. 2  to 28 , 27 , 29 .

Firearms Act, S.C. 1995, c. 39, ss. 2  “chief firearms officer”, 4 to 14, 13, 56, 57, 60, 70, 83, 84, 87, 90, 95, 112 [rep. 2012, c. 6, s. 27].

Firearms Records Regulations, SOR/98-213, ss. 4(1), 7(1), (2).

Interpretation Act, R.S.C. 1985, c. I-21, s. 42 .

Library and Archives of Canada Act, S.C. 2004, c. 11 .

Order Declaring an Amnesty Period (2006), SOR/2006-95 [am. 2007-101, 2008-147, 2009-139, 2010-104, 2011-102, 2013-96, 2014-123].

Privacy Act, R.S.C. 1985, c. P-21 .

Regulation respecting the exclusion of certain premises and certain means of transportation and respecting the exemption of certain persons, CQLR, c. P-38.0001, r. 1.

Authors Cited

Abel, Albert S. “The Neglected Logic of 91 and 92” (1969), 19 U.T.L.J. 487.

Brown, R. Blake. Arming and Disarming: A History of Gun Control in Canada. Toronto: University of Toronto Press, 2012.

Brun, Henri, Guy Tremblay et Eugénie Brouillet. Droit constitutionnel, 6^e éd. Cowansville, Qué.: Yvon Blais, 2014.

Canada. House of Commons. Evidence of the Standing Committee on Public Safety and National Security, No. 012, 1st Sess., 41st Parl., November 17, 2011, pp. 1-2.

Canada. House of Commons. House of Commons Debates, vol. 133, No. 134, 1st Sess., 35th Parl., November 30, 1994, p. 8476.

Canada. House of Commons. House of Commons Debates, vol. 133, No. 154, 1st Sess., 35th Parl., February 16, 1995, p. 9709.

Canada. House of Commons. House of Commons Debates, vol. 133, No. 216, 1st Sess., 35th Parl., June 12, 1995, p. 13631.

Canada. House of Commons. House of Commons Debates, vol. 146, No. 036, 1st Sess., 41st Parl., October 25, 2011, p. 2437.

Canada. House of Commons. House of Commons Debates, vol. 146, No. 041, 1st Sess., 41st Parl., November 1, 2011, pp. 2779, 2780, 2799.

Canada. Library of Parliament. Parliamentary Information and Research Service. “Bill C-19: An Act to amend the Criminal Code and the Firearms Act”, Legislative Summary No. 41-1-C19-E, by Tanya Dupuis, Cynthia Kirkby and Robin MacKay, Legal and Legislative Affairs Division, November 1, 2011.

Canada. Office of the Privacy Commissioner. Review of the Personal Information Handling Practices of the Canadian Firearms Program, Final Report, August 29, 2001 (online: http://publications.gc.ca/collections/Collection/IP34-8-2001E.pdf).

Canada/Québec. Accord financier Canada-Québec relatif à l’administration de la Loi sur les armes à feu, 1^er avril 2006 au 31 mars 2010, entente n^o 2012-004.

Daly, Paul. “Dismantling Regulatory Structures: Canada’s Long-Gun Registry as Case Study” (2014), 33 N.J.C.L. 169.

Hogg, Peter W. Constitutional Law of Canada, 5th ed. Supp. Toronto: Carswell, 2007 (updated 2014, release 1).

Lederman, W. R. “Some Forms and Limitations of Co-Operative Federalism”, in Continuing Canadian Constitutional Dilemmas. Toronto: Butterworths, 1981, 314.

Québec. Assemblée nationale. Journal des débats, vol. 34, n^o 49, 1^re sess., 35^e lég., 23 mai 1995 (en ligne), à 14:40.

Quebec. National Assembly. Votes and Proceedings, No. 57, 2nd Sess., 39th Leg., November 2, 2011, pp. 693-94.

Quebec. National Assembly. Votes and Proceedings, No. 72, 2nd Sess., 39th Leg., December 6, 2011, p. 848.

Royal Canadian Mounted Police. National Program Evaluation Services. RCMP Canadian Firearms Program: Program Evaluation, Final Approved Report, February 2010 (online: http://www.rcmp-grc.gc.ca/pubs/fire-feu-eval/index-eng.htm).

Royal Canadian Mounted Police. Report on the Administration of the Firearms Act to the Solicitor General by the Registrar: Canadian Firearms Registry. Ottawa: RCMP, 1999.

                    APPEAL from a judgment of the Quebec Court of Appeal (Duval Hesler C.J. and Chamberland, Kasirer, St-Pierre and Lévesque JJ.A.), 2013 QCCA 1138, [2013] R.J.Q. 1023, [2013] AZ-50965380, [2013] Q.J. No. 6676 (QL), 2013 CarswellQue 7597 (WL Can.), setting aside a decision of Blanchard J., 2012 QCCS 4202, [2012] R.J.Q. 1895, [2012] AZ-50892414, [2012] Q.J. No. 8319 (QL), 2012 CarswellQue 10074 (WL Can.). Appeal dismissed, LeBel, Abella, Wagner and Gascon JJ. dissenting.

                    Éric Dufour, Hugo Jean and Suzanne-L. Gauthier, for the appellant.

                    Claude Joyal, Q.C., and Ian Demers, for the respondents.

                    No one appeared for the intervener the Chief Firearms Officer.

                    Frédérick Langlois and Alain M. Gaulin, for the intervener the Coalition for Gun Control.

                    Guy Lavergne, for the intervener Canada’s National Firearms Association.

                    The judgment of McLachlin C.J. and Rothstein, Cromwell, Moldaver and Karakatsanis JJ. was delivered by

                    Cromwell and Karakatsanis JJ. —

I.              Introduction

[1]                              Fifteen years ago, this Court determined that Parliament, acting under its power to legislate in relation to criminal law, had the constitutional authority to establish a nationwide gun control scheme.  Parliament’s decision to do so was a contentious policy choice that was contested on constitutional grounds.  Three years ago, Parliament reversed in part that earlier policy choice: it repealed the legislation that had established the long-gun registry within the gun control scheme, and provided for the destruction of the data it contained.  This too was a contentious policy choice which is now contested on constitutional grounds.

[2]                              When the government tabled the bill abolishing the long-gun registry and providing for the destruction of the data it contained, Quebec expressed its intention to create its own provincial gun control scheme and asked Canada to give it the data on long guns connected with the province.  The federal government refused.  As a result, Quebec challenged the constitutionality of the federal law providing for destruction of the data and sought an order requiring the federal government to turn it over.  The Superior Court of Quebec declared that Parliament’s legislative jurisdiction with respect to criminal law does not allow it to legislate for the destruction of the long-gun registration records without first making this data available to provinces seeking to establish their own registries, and ordered the federal government to transfer the relevant data to Quebec.  A five-member panel of the Quebec Court of Appeal disagreed.  Finding the law a valid exercise of the federal criminal law power, it set aside those declarations and orders.  Quebec now appeals to this Court, raising three issues:

1.                  Does the principle of cooperative federalism prevent Parliament from legislating to dispose of the data?

2.                  Does Quebec have the right to obtain the data?

3.                  Is s. 29 of An Act to amend the Criminal Code and the Firearms Act, S.C. 2012, c. 6, ultra vires Parliament’s criminal law power?

[3]                              We agree with the conclusions of the Quebec Court of Appeal and would dismiss the appeal.  The principle of cooperative federalism does not constrain federal legislative competence in this case, Quebec has no legal right to the data, and s. 29  of the Act to amend the Criminal Code and the Firearms Act  (short title Ending the Long-gun Registry Act (“ELRA”)) is a lawful exercise of Parliament’s criminal law legislative power under the Constitution.  We add this; to some, Parliament’s choice to destroy this data will undermine public safety and waste enormous amounts of public money.   To others, it will seem to be the dismantling of an ill-advised regime and the overdue restoration of the privacy rights of law-abiding gun owners.  But these competing views about the merits of Parliament’s policy choice are not at issue here.  As has been said many times, the courts are not to question the wisdom of legislation but only to rule on its legality.  In our view, the decision to dismantle the long-gun registry and destroy the data that it contains is a policy choice that Parliament was constitutionally entitled to make. 

[4]                              We note that our conclusion in this case partly rests on the fact that the Canadian Firearms Registry (“CFR”) flows directly from federal legislation and is not dependent on any provincial statutes.  Different considerations might arise in a case involving a truly interlocking federal-provincial legislative framework.  However, the CFR is not, in our respectful view, such a scheme.  Therefore we need not consider what might follow if it were.

II.           Facts and Judicial History

A.           Overview of the Facts

[5]                              In 1995, Parliament enacted the Firearms Act, S.C. 1995, c. 39 , which created a comprehensive scheme requiring the holders of all firearms — including long guns — to obtain licences and register their guns.  It also made it a criminal offence to possess an unregistered firearm.  The initial creation of the registry was challenged by several provinces, including Alberta who proceeded with a reference in the Alberta Court of Appeal which led ultimately to an appeal to this Court.  In the Reference re Firearms Act (Can.), 2000 SCC 31, [2000] 1 S.C.R. 783, this Court concluded that the Firearms Act  came within the criminal law authority of the federal Parliament, and that “the provinces ha[d] not established that the effects of the law on provincial matters [were] more than incidental”: para. 50. 

[6]                              The Firearms Act  provided for the creation of two types of registries.  The Registrar of Firearms (“Registrar”) was to maintain a single registry, the CFR, containing records of the registration certificates for all prohibited firearms, restricted firearms, and unrestricted firearms (long guns) acquired, transferred, or possessed in Canada: s. 83 .  In addition, the Firearms Act  requires that a Chief Firearms Officer (“CFO”) be appointed in each province and territory — either by a provincial minister or by a federal minister: s. 2 .  This officer must license individuals to possess firearms and must keep a registry containing records of every licence and authorization that she issues or revokes: s. 87 .  By law, the Registrar and the CFOs can access both types of registries: s. 90 .  In practice, that access was ensured by the Canadian Firearms Information System (“CFIS”), an electronic database managed by the Royal Canadian Mounted Police (“RCMP”).  Although the provincial government could access all registration and licensing information in all of the registries through the CFIS database, the statutory authority of CFOs only permitted them to contribute and modify data in their specific licensing registry: ss. 87  and 90 .

[7]                              In April 2012, Parliament enacted the ELRA, which repeals the registration requirement for long guns and decriminalizes the possession of an unregistered long gun.  The registration requirements for prohibited and restricted firearms remain in force, and the registries continue to collect and maintain that data.[1]  Section 29 of the ELRA requires the destruction of all records contained in the registries related to the registration of firearms that are neither prohibited firearms nor restricted firearms:

29. (1) The Commissioner of Firearms shall ensure the destruction as soon as feasible of all records in the Canadian Firearms Registry related to the registration of firearms that are neither prohibited firearms nor restricted firearms and all copies of those records under the Commissioner’s control.

 

(2) Each chief firearms officer shall ensure the destruction as soon as feasible of all records under their control related to the registration of firearms that are neither prohibited firearms nor restricted firearms and all copies of those records under their control.

 

(3) Sections 12  and 13  of the Library and Archives of Canada Act  and subsections 6(1)  and (3)  of the Privacy Act  do not apply with respect to the destruction of the records and copies referred to in subsections (1) and (2).

[8]                              When it became clear that Canada was going forward with the repeal of the registration requirement for long guns, Quebec expressed its intention to create its own long-gun registry and asked the federal authorities for the data connected to Quebec contained in the CFR.  Canada refused and made clear that it intended to permanently destroy all data related to the registration of long guns.  In light of this refusal, Quebec sought a declaration that s. 29 of the ELRA is ultra vires (that is, beyond the powers of Parliament), and that Quebec has a right to obtain the data.  To date, the Quebec National Assembly has not enacted legislation creating a provincial long-gun registry.

B.            Superior Court of Quebec, 2012 QCCS 4202

[9]                              In the Superior Court, Blanchard J. held that the CFR is the result of concerted efforts between the different levels of government (federal, provincial and municipal) and is therefore a result of cooperative federalism.  Finding that the pith and substance of s. 29  is to prevent provincial governments from exercising their legislative competence by precluding them from using the product of this partnership, he held that s. 29  amounts to a violation of the principle of cooperative federalism and is ultra vires the powers of Parliament to legislate in relation to criminal law. 

[10]                          In addition to declaring s. 29 of the ELRA unconstitutional as it applies to data connected to Quebec, Blanchard J. ordered Canada to transfer that data to the province.

C.            Quebec Court of Appeal, 2013 QCCA 1138

[11]                          A five-judge panel of the Court of Appeal unanimously reversed the decision of Blanchard J. 

[12]                          Writing for the Court of Appeal, Duval Hesler C.J.Q. held that Blanchard J. confounded some aspects of the Canadian Firearms Registration System.  This confusion resulted in an erroneous finding of fact that the CFR is the result of a partnership between both orders of government.  That being said, Duval Hesler C.J.Q. concluded that this error was of little significance since, in her view, Parliament clearly has the power to destroy the data independent of the existence of any federal-provincial partnership.

[13]                          The Court of Appeal reasoned that, since this Court held in the Reference re Firearms Act that the creation of the CFR was intra vires (that is, within the authority of) the federal Parliament, Parliament can also legislate to destroy it.  It also held that the principle of cooperative federalism cannot be used to supersede the formal division of powers provided in the Constitution Act, 1867 .

[14]                          Finally, the Court of Appeal determined that Quebec has no right to obtain the data.  The facts that Quebec had chosen to appoint a CFO and that both governments had entered into a financial agreement did not have the effect of granting Quebec any property right in the data.

III.        Analysis

A.           Does the Principle of Cooperative Federalism Prevent Parliament From Legislating to Dispose of the Data?

[15]                          Quebec invokes the principle of cooperative federalism in support of both its argument that s. 29 of the ELRA is ultra vires and its claim that Quebec has the right to receive the data contained in the CFR related to long guns connected to Quebec.  In essence, Quebec is asking us to recognize that the principle of cooperative federalism prevents Canada and the provinces from acting or legislating in a way that would hinder cooperation between both orders of government, especially in spheres of concurrent jurisdiction. 

[16]                          In our respectful view, Quebec’s position has no foundation in our constitutional law and is contrary to the governing authorities from this Court.

[17]                          Cooperative federalism is a concept used to describe the “network of relationships between the executives of the central and regional governments [through which] mechanisms are developed, especially fiscal mechanisms, which allow a continuous redistribution of powers and resources without recourse to the courts or the amending process”: P. W. Hogg, Constitutional Law of Canada (5th ed. Supp.), at p. 5-46; see also W. R. Lederman, “Some Forms and Limitations of Co-Operative Federalism”, in Continuing Canadian Constitutional Dilemmas (1981), 314.  From this descriptive concept of cooperative federalism, courts have developed a legal principle that has been invoked to provide flexibility in separation of powers doctrines, such as federal paramountcy and interjurisdictional immunity.  It is used to facilitate interlocking federal and provincial legislative schemes and to avoid unnecessary constraints on provincial legislative action: see, e.g., Canadian Western Bank v. Alberta, 2007 SCC 22, [2007] 2 S.C.R. 3, at paras. 24 and 43; Reference re Agricultural Products Marketing Act, [1978] 2 S.C.R. 1198; Fédération des producteurs de volailles du Québec v. Pelland, 2005 SCC 20, [2005] 1 S.C.R. 292, at para. 15; Quebec (Attorney General) v. Canadian Owners and Pilots Association, 2010 SCC 39, [2010] 2 S.C.R. 536, at paras. 44-45; Canada (Attorney General) v. PHS Community Services Society, 2011 SCC 44, [2011] 3 S.C.R. 134, at para. 63.  With respect to interjurisdictional immunity, for example, the principle of cooperative federalism has been relied on to explain and justify relaxing a rigid, watertight compartments approach to the division of legislative power that unnecessarily constrains legislative action by the other order of government: “In the absence of conflicting enactments of the other level of government, the Court should avoid blocking the application of measures which are taken to be enacted in furtherance of the public interest” (Canadian Western Bank, at para. 37). 

[18]                          However, we must also recognize the limits of the principle of cooperative federalism.  The primacy of our written Constitution remains one of the fundamental tenets of our constitutional framework: Reference re Secession of Quebec, [1998] 2 S.C.R. 217, at para. 53.  This is especially the case with regard to the division of powers:

. . . the text of the federal constitution as authoritatively interpreted in the courts remains very important.  It tells us who can act in any event.  In other words, constitutionally it must always be possible in a federal country to ask and answer the question — What happens if the federal and provincial governments do not agree about a particular measure of co-operative action?  Then which government and legislative body has power to do what? [Emphasis added; footnote omitted.]

 

(Lederman, at p. 315)

[19]                          The principle of cooperative federalism, therefore, cannot be seen as imposing limits on the otherwise valid exercise of legislative competence: Reference re Anti-Inflation Act, [1976] 2 S.C.R. 373, at p. 421.  This was recently reiterated by this Court in its unanimous opinion in Reference re Securities Act, 2011 SCC 66, [2011] 3 S.C.R. 837, at paras. 61-62:

While flexibility and cooperation are important to federalism, they cannot override or modify the separation of powers.  The Secession Reference affirmed federalism as an underlying constitutional principle that demands respect for the constitutional division of powers and the maintenance of a constitutional balance between federal and provincial powers.

 

In summary, notwithstanding the Court’s promotion of cooperative and flexible federalism, the constitutional boundaries that underlie the division of powers must be respected.  The “dominant tide” of flexible federalism, however strong its pull may be, cannot sweep designated powers out to sea, nor erode the constitutional balance inherent in the Canadian federal state. [Emphasis added.]

 

[20]                          In our respectful view, the principle of cooperative federalism does not assist Quebec in this case.  Neither this Court’s jurisprudence nor the text of the Constitution Act, 1867  supports using that principle to limit the scope of legislative authority or to impose a positive obligation to facilitate cooperation where the constitutional division of powers authorizes unilateral action.  To hold otherwise would undermine parliamentary sovereignty and create legal uncertainty whenever one order of government adopted legislation having some impact on the policy objectives of another.  Paradoxically, such an approach could discourage the practice of cooperative federalism for fear that cooperative measures could risk diminishing a government’s legislative authority to act alone. 

[21]                          We conclude that the principle of cooperative federalism does not prevent Parliament from exercising legislative authority that it otherwise possesses to dispose of the data.

B.            Does Quebec Have the Right to Obtain the Data?

[22]                          Quebec submits that it has a right to receive the long-gun registration data from the federal government whether or not Parliament is constitutionally entitled to legislate with respect to the fate of the data.  In addition to the principle of cooperative federalism considered above, Quebec relies on the findings of the trial judge that Quebec’s contributions to the data contained in the CFR render the data the result of a federal-provincial partnership.  We note here that there is a factual disagreement with regard to whether or not the data contained in the CFR is in fact the result of a cooperative effort.  The Court of Appeal was of the opinion that the conclusions of Blanchard J. on this matter were based on a palpable error of fact. In our view, this disagreement has no impact on the outcome of this case.

[23]                          We agree with the conclusion of our colleagues LeBel, Wagner and Gascon JJ. that Quebec “has not established a legal basis for its claim to the data”: para. 198.  We do so for three reasons.

[24]                          First, as mentioned above, the principle of cooperative federalism does not limit the scope of the legislative powers assigned by the Constitution. 

[25]                          Second, Quebec’s alleged right to obtain the data arises from its expectation of having continuing access to that data regardless of changes to federal legislation.  We cannot agree that the provinces’ reliance on the existence of the data can limit Parliament’s capacity to destroy a registry, which — as explained below — flows exclusively from its criminal law head of power.  Accepting this position would circumvent or effectively overturn this Court’s rejection of the “legitimate expectation” doctrine.  This Court has made it clear that a province’s legitimate expectation of action that the federal government would or would not undertake, even an expectation relating to financial considerations, cannot bind Parliament’s legislative action: Reference re Canada Assistance Plan (B.C.), [1991] 2 S.C.R. 525, at pp. 557-59; see also Wells v. Newfoundland, [1999] 3 S.C.R. 199, at para. 59, and Authorson v. Canada (Attorney General), 2003 SCC 39, [2003] 2 S.C.R. 40, at para. 39.  Quebec’s position would also mean by extrapolation that the adoption of the Firearms Act  had a binding effect — the continued existence of the CFR data — which Parliament cannot undo alone.  That conclusion must also be rejected; it is contrary to the established constitutional principle that Parliament cannot, through ordinary legislation, fetter itself by inhibiting future legislative action: Reference re Securities Act, at para. 119; Hogg, at pp. 12-8 and 12-9. 

[26]                          Third, even if the data accessible through the CFR was the result of a cooperative effort, as asserted by Quebec, any effort on Quebec’s part was statutorily limited to the licensing data held in the CFO’s licensing registry.  The Firearms Act  did not empower this officer to modify or contribute to the registration certificate data compiled and maintained by the Registrar, nor did the CFO act in her capacity as a provincial official in maintaining the licensing registry.  As s. 29 of the ELRA contemplates the registration data, not the licensing data, we agree with the Quebec Court of Appeal that [translation] “this claim fails to demonstrate that there is a right to the data in the CFR or that Quebec participated in the creation of the CFR”: para. 57 (CanLII).  In addition, we note that Quebec’s argument relies on agreements between the federal and provincial governments, and on other pieces of evidence which include RCMP documents, federal-provincial funding agreements, and statements from politicians and other officials involved.  These legal instruments and pieces of evidence are unquestionably subordinate to parliamentary sovereignty and can therefore be displaced by valid federal legislation.  Moreover, even if, as Quebec submits, there is no legal obstacle to the federal government transferring the data to Quebec, the absence of a legal obstacle to undertaking a given action does not, alone, create a legal obligation to do so.

C.            Is Section 29 Ultra Vires Parliament’s Criminal Law Power?

(1)           Applicable Legal Principles

[27]                          The Constitution confers on Parliament the exclusive authority to legislate with respect to, among others, “all [m]atters” coming within “[t]he Criminal Law, except the Constitution of Courts of Criminal Jurisdiction, but including the Procedure in Criminal Matters”: Constitution Act, 1867, s. 91(27) .  In that regard, the Chief Justice stated the following constitutional question in this case:

Is s. 29 of the Ending the Long-gun Registry Act, S.C. 2012, c. 6 , ultra vires Parliament having regard to its criminal law power under s. 91(27)  of the Constitution Act, 1867 ?

[28]                          To answer this question, we first characterize the law — that is, determine its subject matter or “pith and substance” — and then classify it as to whether it is in relation to the subject of criminal law and procedure. 

[29]                          The “pith and substance” analysis involves determining the law’s “dominant purpose or true character”: RJR-MacDonald Inc. v. Canada (Attorney General), [1995] 3 S.C.R. 199, at para. 29.  As Binnie and LeBel JJ. put it in Canadian Western Bank, at para. 26: “This initial analysis consists of an inquiry into the true nature of the law in question for the purpose of identifying the ‘matter’ to which it essentially relates” (emphasis added).  The object of the exercise is to determine whether that “matter” comes within a particular class of subjects for the purpose of determining which order of government can legislate.  Both the law’s purpose and its legal and practical effects are considered as part of this analysis: Reference re Securities Act, at paras. 63-64; Quebec (Attorney General) v. Lacombe, 2010 SCC 38, [2010] 2 S.C.R. 453, at para. 20; Kitkatla Band v. British Columbia (Minister of Small Business, Tourism and Culture), 2002 SCC 31, [2002] 2 S.C.R. 146, at paras. 53-54.  Care must be taken, however, not to confuse the law’s purpose with “the means chosen to achieve it”: Ward v. Canada (Attorney General), 2002 SCC 17, [2002] 1 S.C.R. 569, at para. 25.

[30]                          Where the challenge concerns a particular provision which forms part of a larger scheme, the pith and substance analysis begins with the challenged provision: Kitkatla Band, at para. 56.  However, the “matter” of the provision must be considered in the context of the larger scheme, as its relationship to that scheme may be an important consideration in determining its pith and substance: Kirkbi AG v. Ritvik Holdings Inc., 2005 SCC 65, [2005] 3 S.C.R. 302, at paras. 20-21.

[31]                          Courts must be careful not to endorse a “colourable” statute, that is, one that in form appears to relate to a matter within the legislative competence of the enacting order of government, but in substance addresses a matter falling outside its competence: see Hogg, at p. 15-19; Reference re Upper Churchill Water Rights Reversion Act, [1984] 1 S.C.R. 297. The colourability doctrine simply means that “form is not controlling in the determination of essential character”: A. S. Abel, “The Neglected Logic of 91 and 92” (1969), 19 U.T.L.J. 487, at p. 494; Hogg, at p. 15-20.  Courts are, for good reasons, reluctant to find legislation to be colourable: H. Brun, G. Tremblay and E. Brouillet, Droit constitutionnel (6th ed. 2014), at p. 464.  There is a danger that any broader application of the colourability doctrine may lead the courts to exceed their role of determining the constitutionality of legislation and, instead, express disapproval of either the policy of the statute or the means by which the legislation seeks to carry it out: Hogg, at p. 15-20; see also Ward, at para. 26.

[32]                          Once the subject matter has been identified, the next step of the pith and substance analysis consists of classifying the legislation or provision in relation to the division of legislative power in the Constitution.  Here, we must determine whether the “matter” of s. 29 of the ELRA comes within the “subject” of criminal law, a head of power attributed to Parliament at s. 91(27)  of the Constitution Act, 1867 .  This classification exercise may — and often does — involve defining the scope of the relevant head or heads of power: Reference re Securities Act, at para. 65.  If the “matter” of the legislation comes within the “subject” of the head of power of the enacting order of government, the legislation is intra vires even if it has incidental effects on the other jurisdiction’s legislative competence (subject to federal paramountcy in the case of provincial legislation): Reference re Firearms Act, at para. 49. 

[33]                          This point takes on special importance when considering the classification of a law, like this one, that undoes a previously enacted scheme.  In classifying legislation that undoes an existing legislative scheme, due regard must be paid to the proper classification of that scheme.  Consider the example of the repeal of a criminal offence.  The scope of the criminal law power extends to laws that create a prohibition backed by a penalty for a criminal law purpose: Reference re Firearms Act, at para. 27.  A law creating a true criminal offence clearly has these features, but a law repealing that same offence does not; the latter does not create a prohibition backed by a penalty for a criminal law purpose.  The repealing enactment, however, is clearly criminal law because the “matter” of that law comes within the same criminal law subject as did the provision it seeks to repeal.  This is why Quebec is clearly right in this case not to question Parliament’s authority to repeal the long-gun registry scheme, and to restrict its challenge to s. 29 of the ELRA, which provides for the destruction of the data.  The fatal flaw in Quebec’s analysis, however, is that in characterizing the pith and substance of s. 29, it ignores the place of the data gathering and retention provisions in the overall scheme.

(2)           Application

(a)           The “Matter” of Section 29

[34]                          In essence, Quebec submits that the pith and substance of s. 29 of the ELRA is to prevent the long-gun registry from being continued through provincial legislation.  Although Quebec does not suggest that this “matter” comes within any of the enumerated heads of provincial legislative power, it submits nonetheless that the federal legislation is not a valid exercise of the criminal law power because it [translation] “encroaches massively on the ability of a provincial legislature to exercise as it sees fit its powers with respect to the administration of justice, public safety, the prevention of crime and the social costs associated with crime”: A.F., at para. 122; see also para. 115.  Quebec relies on several statements made by the Prime Minister and other federal cabinet ministers to the effect that Canada’s objective is to put a definite end to the long-gun registry by destroying the data contained in the CFR connected to long guns, and that they will do nothing to support any province or territory who would like to create a local long-gun registry.  Quebec also submits that the effect of the destruction of the data will be to make it prohibitively expensive and complicated for the province to create its own long-gun registry.  Section 29 of the ELRA therefore constitutes an encroachment on the province’s powers.

[35]                          Canada submits that the subject matter of s. 29 is the same as the rest of the ELRA: to abolish the long-gun registry and put an end to what it considers an unwarranted intrusion into the private lives of Canadian gun owners.  Canada also asserts that the practical effects of the destruction of the data on provinces who want to create their own gun control scheme do not change the pith and substance of the impugned provision.

[36]                          In our respectful view, the proper characterization of the “matter” of s. 29 of the ELRA derives from that of the scheme that it is undoing.  In the Reference re Firearms Act, this Court held that both the criminal prohibition on possessing an unregistered firearm and the registration scheme provided for in the Firearms Act  were valid public safety measures.  It determined that the purpose of the scheme was to address “the problem of the misuse of firearms and the threat it poses to public safety”: para. 21.  It also held that the effects of the scheme supported the conclusion that the Firearms Act  is a public safety measure: para. 24.  The Court rejected arguments from Ontario and Saskatchewan to the effect that the “matter” of the registration and data retention provisions, which s. 29  seeks to dismantle, was the regulation of property within the province.  The Court concluded that the “matter” of these provisions was the same as the rest of the scheme — public safety — and that they should accordingly be classified as being in relation to the subject of criminal law:

We are not persuaded that the registration provisions can be severed from the rest of the Act, nor that they fail to serve Parliament’s purpose in promoting public safety.  The licensing provisions require everyone who possesses a gun to be licensed.  The registration provisions require all guns to be registered.  The combination of the two parts of the scheme is intended to ensure that when a firearm is transferred from one person to another, the recipient is licensed.  Absent a registration system, this would be impossible to ascertain.  If a gun is found in the possession of an unlicensed person, the registration system permits the government to determine where the gun originated.  With a registration scheme in place, licensed owners can be held responsible for the transfer of their weapons. . . .  These interconnections demonstrate that the registration and licensing portions of the Firearms Act  are both tightly linked to Parliament’s goal in promoting safety by reducing the misuse of any and all firearms.  Both portions are integral and necessary to the operation of the scheme. [Emphasis added.]

(Reference re Firearms Act, at para. 47)

[37]                          Viewed in this light, the “matter” of s. 29 is simply to determine what will happen with the data collected under a now repealed legislative scheme.  Given that the “matter” of the registration scheme, including data collection, was found to be public safety in the Reference re Firearms Act, dealing with the collected data following the repeal of the scheme must share that same characterization.  It is logical that the ELRA provides for what is to happen with the remains of the scheme that it repeals, and destroying the data is the means Parliament adopted to accomplish that purpose.  This is not to say that every measure that is part of or consequent to the repeal of legislation invariably shares the pith and substance of the repealed scheme.  However, in this case, the repealing enactment’s purpose and effects are to determine what will happen to the data collected under the now repealed scheme, a scheme that the Court previously characterized as being in relation to public safety.

[38]                          Quebec’s submissions, in our respectful view, confuse the subject matter of s. 29 of the ELRA with Canada’s motives and the means employed by Parliament.  In determining the true character of s. 29, we are not concerned with whether destroying the data is good policy, whether Canada’s motives were sound, or whether the destruction of that data conflicts with the policy objectives of Quebec.  We recognize that the federal government’s ultimate goal may well have been to prevent Quebec from creating its own long-gun registry.  We also accept that the destruction of the data is the means chosen by Canada because of its irremediable nature.  That being said, these considerations are not indicative of a “colourable” purpose from a division of powers’ perspective.  An intention on the part of one level of government to prevent another from realizing a policy objective it disagrees with does not, on its own, lead to the conclusion that there is an encroachment on the other level of government’s sphere of exclusive jurisdiction.  The fact that s. 29 of the ELRA has the practical effect of making it more difficult financially for Quebec to create its own gun control regime does not, in light of the rejection of the legitimate expectations doctrine, affect the pith and substance of the provision.  As Sopinka J. said in Reference re Canada Assistance Plan, “‘[i]mpact’ [upon a constitutional interest outside Parliament’s jurisdiction] with nothing more is clearly not enough to find that a statute encroaches upon the jurisdiction of the other level of government”: p. 567.  In our view, the practical effects of s. 29 do not exceed that threshold.

[39]                          Quebec also submits that an analogy can be drawn between the facts of this case and those in Upper Churchill, in which this Court held that Newfoundland legislation repealing a statutory lease granted to Hydro-Quebec was colourable because it was in fact expropriating rights situated outside the province: p. 335.  This argument implies that s. 29 exceeds Parliament’s criminal law jurisdiction and is a colourable attempt to legislate in relation to property and civil rights in the province: see P. Daly, “Dismantling Regulatory Structures: Canada’s Long-Gun Registry as Case Study” (2014), 33 N.J.C.L. 169, at pp. 178-80. 

[40]                          We reject this argument.  Like the Quebec Court of Appeal, we do not think that the principles in Upper Churchill are applicable to this case.  As we have discussed earlier, s. 29 of the ELRA does not seek either in substance or in form to limit Quebec’s legislative authority to create a provincial long-gun registry, whereas the impugned legislation in Upper Churchill did seek to interfere with Quebec’s legislative authority.  Rather, s. 29 simply prevents Quebec from using the data obtained through the federal long-gun registry in establishing a provincial registry.  Quebec does not (and in our view could not) take the position that it could validly enact legislation that deals with what will happen with the data of the repealed scheme.  That being the case, Parliament’s enactment of s. 29 cannot be a colourable attempt to legislate in relation to a provincial head of power.  In any event, there is no difference between the form and substance of the provisions at issue; both deal with what will happen to the data collected under the now repealed scheme.  In our view, the doctrine of colourability is not engaged here. 

[41]                          In summary, this Court has already been called upon in the Reference re Firearms Act to determine the pith and substance of the scheme enacted by the Firearms Act .  In our view, legislation repealing the part of that scheme relating to long guns, including a provision addressing what will happen to the data collected thereunder, must be characterized in the same way.  We conclude that s. 29 , in essence, relates to public safety — as did the long-gun registration scheme being repealed by the balance of the ELRA.

(b)           Classification: Does Section 29 Come Within the Criminal Law Power?

[42]                          Quebec submits that s. 29 cannot be a valid exercise of Parliament’s criminal law power because it is not aimed at preventing crime or at decriminalizing any conduct.  We disagree and would hold that s. 29 of the ELRA falls under the criminal law head of power. 

[43]                          It is not contested that the repeal of criminal provisions constitutes a valid exercise of the criminal law power: Hogg, at p. 18-21.  We accept that there is a factual difference between repealing a criminal provision and providing for what will happen to the data collected under that provision.  That being said, we are of the view that there is no significant legal distinction between these two actions in the current case because the data at issue here was collected exclusively through the exercise of the criminal law power.  The power to repeal a criminal law provision must logically be wide enough to give Parliament jurisdiction to destroy the data collected for the purpose of a criminal law provision.  If a law establishing a scheme requiring collection of data is legislation “in relation to” criminal law, then legislation providing for the destruction of that data on the repeal of the scheme must also be legislation “in relation to” criminal law.  This is the case here. 

[44]                          In addition, the prospect that Parliament would not have the power to destroy records under its control and created under a scheme validly enacted by it is concerning.  Since we do not see how provinces could have the power to destroy the data contained in the federally enacted long-gun registry, Quebec’s position — if it were accepted — would suggest that neither Parliament nor provincial legislatures could legislate to destroy the data of the repealed scheme.  We reject this proposition as it would run afoul of the principle that the Constitution Act, 1867  provides for a complete division of powers between both orders of governments: Attorney-General for Ontario v. Attorney-General for Canada, [1912] A.C. 571 (P.C.), at p. 581; Attorney-General for Canada v. Attorney-General for Ontario, [1937] A.C. 326 (P.C.). There is no topic that cannot be legislated upon: Reference re Same-Sex Marriage, 2004 SCC 79, [2004] 3 S.C.R. 698, at para. 34.  Quebec’s proposition would also be inconsistent with the notion that Parliament possesses the residual power to legislate on subject matters that do not come within a provincial head of power: see, e.g., Hunt v. T&N plc, [1993] 4 S.C.R. 289, at p. 326, and Hogg, at pp. 17-1 and 17-2.

[45]                          Therefore, we agree with the Quebec Court of Appeal that s. 29 of the ELRA should be characterized as being in relation to criminal law.  It therefore falls within the legislative competence of Parliament.

IV.        Disposition of the Appeal

[46]                          We would dismiss the appeal with costs throughout and answer the constitutional question as follows:

Is s. 29 of the Ending the Long-gun Registry Act, S.C. 2012, c. 6 , ultra vires Parliament having regard to its criminal law power under s. 91(27)  of the Constitution Act, 1867 ?

Answer: No.

                    English version of the reasons of LeBel, Abella, Wagner and Gascon JJ. delivered by

                    LeBel, Wagner and Gascon JJ. (dissenting) —

I.          Introduction

[47]                          In 2012, Parliament enacted the Act to amend the Criminal Code and the Firearms Act, S.C. 2012, c. 6  (short title Ending the Long-gun Registry Act (“ELRA”)). This Act repeals the provisions of the Firearms Act, S.C. 1995, c. 39  (“FA ”), relating to the mandatory registration of long guns. Section 29 of the ELRA provides for the destruction, as soon as feasible, of all data compiled since 1998 with respect to the registration of such guns in Canada.

[48]                          The Quebec government, which considered the data to be the product of a federal-provincial partnership and wanted to continue registering long guns within its territory, then asked the federal government to transfer to it all data from Quebec concerning registration certificates for long guns. The federal government refused to do so, and the Attorney General of Quebec (“AGQ”) then instituted legal proceedings. According to the AGQ, the unilateral destruction of these data under s. 29 of the ELRA would be unconstitutional unless they were first transferred to Quebec.

[49]                          The Quebec Superior Court declared that s. 29 of the ELRA is unconstitutional and that Quebec is entitled to a transfer of these data. The Quebec Court of Appeal unanimously reversed that judgment. The AGQ now appeals to this Court. Our colleagues would dismiss the appeal.

[50]                          We are of the opinion that the appeal should be allowed, but only in part. This conclusion is dictated by the exceptional circumstances in which long-gun regulation was implemented in Canada. In our opinion, and this is where we diverge from our colleagues’ view, both the collection of the data with respect to long guns and the broader initiatives aimed at regulating the use of such guns were the result of a partnership with the provinces, including Quebec. Where an integrated scheme such as this requires the exercise of both federal and provincial legislative powers, the analytical framework for questions related to the division of powers must be adapted and applied accordingly. Whether the means the federal government adopted to terminate this partnership were constitutional can be measured, in particular, in terms of the effect they will have on its partners’ powers.

[51]                          The ELRA is the measure chosen by Parliament to end its participation in long-gun regulation. But s. 29 goes beyond the scope of that purpose, as it requires that the data in question be destroyed without providing for a possibility of their first being transferred to the provincial partners, which prevents the latter from using them in the exercise of their powers. This section has significant effects on Quebec’s legislative powers and is not necessary to the achievement of the ELRA’s purpose. Section 29 is therefore unconstitutional and should be declared to be invalid.

[52]                          Although we find that s. 29 is unconstitutional, the AGQ has nonetheless failed to establish a legal basis for his request for a compulsory transfer of the data. The conditions applicable to such a transfer are a matter for the governments concerned, not the courts.

II.       Background

[53]                          Firearms have been subject to regulation in Canada since early in the country’s history: “Canada has had laws restricting the possession and use of firearms since 1877, and a nationwide permit system for the carrying of small arms has been in effect since 1892” (Library of Parliament, “Bill C-19: An Act to amend the Criminal Code and the Firearms Act”, Legislative Summary No. 41-1-C19-E, November 1, 2011 (“Legislative Summary”), at p. 1). Since 1969, firearms have been divided into three classes for purposes of regulation: prohibited firearms, restricted firearms and non-restricted firearms (also called “long guns”) (Criminal Law Amendment Act, 1968-69, S.C. 1968-69, c. 38, s. 6). The regulation of firearms was originally focused on prohibited and restricted firearms. During the 1970s and 1980s, there was growing discontent over the disparity in treatment between long guns and the other two classes of firearms. Various groups promoting stricter regulation of long guns started a serious debate about public safety: see R. B. Brown, Arming and Disarming: A History of Gun Control in Canada (2012), at pp. 167-203. Nothing came of their efforts.

[54]                          The tragedy at the École Polytechnique in Montréal on December 6, 1989 shook the Canadian public. In its wake, was formed in the early 1990s and the intervener Coalition for Gun Control began lobbying the Canadian government and other organizations for stricter gun control: Brown, at pp. 204-7. After a long debate and despite several failures, the Coalition’s efforts finally came to fruition a few years later.

[55]                          In 1994, the federal Minister of Justice announced his intention to introduce stronger firearms control measures. In tabling Bill C-68 (which would become the FA ) in the House of Commons, the Minister focused on one key element: the need for and usefulness of a long-gun registration system. He stressed the important role that the provinces would have in designing this system, making the following comments, among others:

                        During the course of the last several months, I have been in continuous touch with senior officials and indeed with provincial counterparts, attorneys general and ministers of justice, exchanging views about the proposals that will be decided upon and announced this afternoon.

(House of Commons Debates, vol. 133, No. 134, 1st Sess., 35th Parl., November 30, 1994, at p. 8476)

                    We can take the opportunity of designing and implementing such a system in collaboration with provincial authorities, with the input of the firearms’ groups to eliminate irritants, to overcome paperwork burden, to simplify and streamline the system so that all of our objectives can be achieved at the same time.

(House of Commons Debates, vol. 133, No. 154, 1st Sess., 35th Parl., February 16, 1995, at p. 9709)

[56]                          At the time of the enactment of the FA , there was no dispute that it would be impossible to set up a registration system without provincial collaboration. Member of Parliament Stephen Harper (who would subsequently become Prime Minister) said the following in the House:

                    We know that the Department of Justice retains overall responsibility for gun control but the program is predominantly administered by provincial and territorial governments, through chief firearms officers and local police agencies.

(House of Commons Debates, vol. 133, No. 216, 1st Sess., 35th Parl., June 12, 1995, at p. 13631)

[57]                          In Quebec, there was a consensus that the program would be useful, although some questioned whether the compensation for the costs of implementing and administering the program that the federal government proposed to pay the provinces that chose to take part would be sufficient: see, e.g., Quebec, National Assembly, Journal des débats, vol. 34, No. 49, 1st Sess., 35th Leg., May 23, 1995 (online), at 14:40, Mr. Brassard (Minister of the Environment and Wildlife). Other provinces shared this concern:

                        I have been told the funds that the provinces received are not adequate to cover their costs. . . .

                        The federal government’s latest offer amounts to only about two-thirds of the actual cost incurred for these programs.

(House of Commons Debates, June 12, 1995, at p. 13631)

[58]                          Upon receiving Royal Assent the day before the sixth anniversary of the École Polytechnique tragedy, the FA  created controversy. Alberta challenged its constitutionality in 1996, arguing that it was ultra vires Parliament. In a unanimous opinion, the Court rejected this argument and held that the FA  falls within the federal criminal law power: Reference re Firearms Act (Can.), 2000 SCC 31, [2000] 1 S.C.R. 783 (“FA Reference”). We will return to this question below.

[59]                          The coming into force of the FA  on December 1, 1998, marked the beginning of the registration of all firearms — including long guns — in Canada. After that date, any individual, unless exception applied, who acquired or possessed a firearm of any type in Canada was required to hold a licence to possess firearms as well as a specific registration certificate for the firearm in question: ss. 4  to 14  and 112  (rep. s. 27  ELRA) FA ; ss. 90  and 91  of the Criminal Code, R.S.C. 1985, c. C-46 . Moreover, it should be specified that holding a licence was a precondition for obtaining a registration certificate: s. 13  FA .

[60]                          In this regard, the FA  provides for the establishment of two types of registries: the Canadian Firearms Registry (“CFR”), which is to be maintained by the Registrar of Firearms (“Registrar”), and records of chief firearms officers (“CFOs”) (ss. 83  and 87  FA ). Whereas there is only one Registrar for all of Canada, each province or territory has its own CFO. The CFOs for the territories are appointed by the federal Minister of Public Safety and Emergency Preparedness: s. 2(1)  “chief firearms officer” FA . With respect to the provinces, on the other hand, it is only if a province has not designated a CFO that the federal minister will do so: ibid. At the time the FA  was passed, CFOs were designated by five provinces, namely Ontario, Quebec, New Brunswick, Prince Edward Island and Nova Scotia (Royal Canadian Mounted Police (“RCMP”), National Program Evaluation Services, RCMP Canadian Firearms Program: Program Evaluation, Final Approved Report, February 2010 (online) (“CFP Evaluation”), at p. 9; affidavit of Pierre Perron (Assistant Commissioner of the RCMP and Director General of the CFP), April 5, 2012, joint record (“J.R.”), vol. 3, at para. 10).

[61]                          Under the FA , the Registrar is responsible for issuing a registration certificate for each firearm: s. 60 . The CFOs, for their part, are responsible for issuing possession licences as well as authorizations to carry and authorizations to transport for firearms belonging to individuals or businesses in their respective provinces: ss. 56  and 57  FA . The CFR contains, inter alia, information on registration certificates, and the records of the CFOs contain information on licences and authorizations to carry. Section 90  of the FA  requires that the Registrar and the CFOs have access to one another’s registries.

[62]                          The Canadian Firearms Information System (“CFIS”) (formerly known as the Canadian Firearms Registration System (“CFRS”)) allows for the pooling of information in the CFR maintained by the Registrar and the registries kept by the CFOs. The CFIS is now administered by the RCMP, and it provides access to all information that must be kept pursuant to the FA . However, only the Registrar can modify information recorded in the CFR, and only the CFO can modify records he or she maintains: Firearms Records Regulations, SOR/98-213, s. 7(1) and (2).

[63]                          The keeping of the CFR and of the CFOs’ records is one aspect of the broader framework of the Canadian Firearms Program (“CFP”). The CFP has multiple components, including firearms regulation, training and support for police forces and those who work in the criminal justice system, education for firearms owners, and public awareness: CFP Evaluation, at pp. 7-8. Several aspects of the CFP, such as the mandatory firearms safety course for owners, were developed with provincial co-operation and support: affidavit of Pierre Perron, at para. 7 and Exhibit A.

[64]                          In 2006, a new federal government came to power. At the outset of its mandate, it announced an amnesty under which owners of long guns would not be liable to sanctions for having failed to register their weapons or to renew their licences to possess firearms: Order Declaring an Amnesty Period (2006), SOR/2006-95. The amnesty period was extended year after year, effectively decriminalizing the failure to obtain a licence and to register long guns: SOR/2007-101, SOR/2008-147, SOR/2009-139, SOR/2010-104, SOR/2011-102, SOR/2013-96 and SOR/2014-123. Nonetheless, the RCMP and its partners continued to administer the CFP, including its long-gun component.

[65]                          On October 25, 2011, after several bills had died on the Order Paper and following a spirited public debate, Bill C-19 was tabled in the House of Commons: House of Commons Debates, vol. 146, No. 036, 1st Sess., 41st Parl., at p. 2437. Its purpose was to abolish the requirement to register long guns and to ensure the destruction of all data related to the registration of such guns. It was only the portion of the CFR concerning long guns that was affected by this bill and is at issue in this appeal. Clause 29 of Bill C-19, which would become s. 29 of the ELRA, read as follows:

                        29. (1)  The Commissioner of Firearms shall ensure the destruction as soon as feasible of all records in the Canadian Firearms Registry related to the registration of firearms that are neither prohibited firearms nor restricted firearms and all copies of those records under the Commissioner’s control.

                        (2) Each chief firearms officer shall ensure the destruction as soon as feasible of all records under their control related to the registration of firearms that are neither prohibited firearms nor restricted firearms and all copies of those records under their control.

                        (3) Sections 12  and 13  of the Library and Archives of Canada Act  and subsections 6(1)  and (3)  of the Privacy Act  do not apply with respect to the destruction of the records and copies referred to in subsections (1) and (2).

[66]                          The Quebec government, with the support of its police officers and much of its population, had made its opposition to the federal government’s attempts to abolish the registration of long guns known. Between 2006 and 2011, the National Assembly had adopted at least six motions denouncing the various federal government bills, calling on members of Parliament to vote against these measures and demanding that the CFR be maintained. On November 2, 2011, it unanimously adopted a new motion, this time demanding that Quebec’s CFO “take all of the necessary measures to preserve the completeness of data from Québec entered in the Firearms Registry”: Votes and Proceedings, No. 57, 2nd Sess., 39th Leg., November 2, 2011, at pp. 693-94.

[67]                          Two weeks later, Quebec’s Minister of Public Security appeared before the House of Commons Standing Committee on Public Safety and National Security to ask that the CFR be maintained or, failing that, that Bill C-19 be amended by removing the provision on the destruction of the data, namely clause 29: House of Commons, Evidence of the Standing Committee on Public Safety and National Security, No. 012, 1st Sess., 41st Parl., November 17, 2011, at pp. 1-2. Quebec’s minister also wrote to his federal counterpart to ask him to agree on terms for transferring CFIS data concerning Quebec citizens. He noted that this would be [translation] “an excellent opportunity for the federal government to deal with this matter in a spirit of co-operative federalism that would be respectful of, rather than frustrate, Quebec’s collective goals”: letter dated December 2, 2011, J.R., vol. 10, at p. 161.

[68]                          Finally, on December 6, 2011, the National Assembly voted to adopt another motion in which it asked the Quebec government “to seek legal recourse aiming to maintain the federal long gun registry and to ensure that its data is preserved, before the abolition of this registry causes the destruction of data when Bill C-19 shall have been passed”: Votes and Proceedings, No. 72, 2nd Sess., 39th Leg., December 6, 2011, at p. 848. On December 13, having received no response from the federal government, Quebec’s Minister of Public Security announced that his government would commence legal proceedings to recover the data concerning registration certificates for long guns held by Quebec citizens. He stated that should the courts rule in Quebec’s favour, a bill would be tabled in the National Assembly to create a Quebec long-gun registry using the data obtained from the federal government: Press release from the Minister of Public Security, December 13, 2011, J.R., vol. 10, at pp. 48-49.

III.    Judicial History

[69]                          On April 2, 2012, three days before the ELRA came into force, the AGQ filed a motion in the Quebec Superior Court for a declaratory judgment, an interim interlocutory injunction, a permanent injunction and safeguard orders. The AGQ asked that s. 29 of the ELRA be declared unconstitutional. He also sought an order enjoining the federal government to transfer to him all Quebec data with respect to long guns.

A.       Quebec Superior Court, 2012 QCCS 4202

[70]                          Blanchard J. of the Superior Court granted the AGQ’s motion. His judgment can be broken down into three elements: (1) the facts; (2) the constitutionality of s. 29 of the ELRA; and (3) Quebec’s right to obtain the data.

(1)      Facts

[71]                          Blanchard J. first pointed out that the FA  created a distinction between the registries kept by the Registrar (concerning the registration of firearms) and those kept by the CFOs (concerning the issuance of licences) (para. 27).

[72]                          Blanchard J. next noted that the [translation] “collected information [from the two types of records] is stored in the CFIS, [a] complex central computer registry” (para. 28 (CanLII)). He also stressed that, “[a]ccording to the structure of the FA , the Registrar and the Chief Firearms Officers have the same level of access to the data” (para. 72).

[73]                          Blanchard J. mentioned that Quebec had contributed to [translation] “gathering, analyzing, organizing, and modifying” the data in the registry (para. 151), which derived from the combined efforts of governments at the municipal, provincial and federal levels (para. 102). The establishment of the registry had therefore created a partnership, particularly as regards the data it contained (para. 192).

[74]                          Blanchard J. found that the data in question could not be characterized as being solely [translation] “federal data”, given that they were subject to the applicable access to information and privacy legislation of both the federal government and Quebec (para. 82). He wrote the following:

                        [translation]  The Court must recognize an actual, concrete effect to this intent, clearly expressed by Canada and Quebec, to submit all of the information gathered to two jurisdictions concurrently, which must, in practice, have some meaning. [para. 83]

[75]                          According to Blanchard J., the partnership extended beyond the data; it also involved the implementation of the CFP (para. 73), the effective administration of the FA  (para. 83) and the joint and complementary exercise of relevant federal and provincial powers (para. 104).

[76]                          He concluded that [translation] “[t]he firearms registry is a complex interweaving of federal, provincial and municipal powers, such that it could not exist without the close and constant collaboration of each level of government” (para. 192).

(2)      Constitutionality of Section 29 of the ELRA

[77]                          Blanchard J. observed that a more flexible view of federalism than the one formerly adopted by the Judicial Committee of the Privy Council, a view based on co-operation rather than on exclusive powers, has underlain this Court’s decisions since 1949 (paras. 52-53). He added that since 1998, the Court had [translation] “set out an approach to the interpretation of constitutional powers based on cooperation rather than the strict enforcement of jurisdictions” (para. 71).

[78]                          Blanchard J. noted the importance of the FA Reference to the case at bar (paras. 63-66). In that case, this Court had found no indication of a colourable intrusion into provincial jurisdiction at the time of the enactment of the FA .  The Court also specified that an analysis of the FA ’s provisions showed that Parliament had not had an improper motive, nor was it taking over provincial powers under the guise of the criminal law.

[79]                          Blanchard J. explained that the AGQ did not dispute that Canada was acting within its sphere of jurisdiction in abolishing the long-gun registration system. Rather, he was asking the court to decide whether the destruction of the data, provided for in s. 29, could be linked to the federal criminal law power (para. 87). Blanchard J. found that s. 29, in its pith and substance, was not linked to that power. The provision did not suppress an evil, nor did it establish a prohibition or a penalty (para. 125). He concluded, on the basis of declarations of the Prime Minister of Canada, the Minister of Intergovernmental Affairs and the Parliamentary Secretary to the Minister of Justice, that Parliament instead had an intention to hinder the exercise of the powers of all the provincial legislatures (paras. 136-39).

[80]                          In light of his conclusion regarding the pith and substance of s. 29, Blanchard J. then considered the ancillary powers doctrine. He noted that an assessment of the seriousness of the encroachment must be grounded in the facts of each case, and that the factors discussed in General Motors of Canada Ltd. v. City National Leasing, [1989] 1 S.C.R. 641, can be used to guide the assessment (para. 129). These factors include the scope of the heads of power in play, the nature of the impugned provision, and the existence of prior legislation on the matter in question. Nonetheless, Blanchard J. found that [translation] “this analytical framework [seems] incongruous [in this case]. In reality, section 29 of C-19 is a legislative provision that does not establish jurisdiction in any strict sense; rather, in both intention and scope, it aims to prevent any other level of government from legitimately using the data recorded in the Registry” (para. 130). In his view, “[t]his situation, which is without precedent in Canadian constitutional history, justifies an analytical approach that takes this context into account” (para. 131).

[81]                          Blanchard J. found that s. 29 was intended to prevent the provinces from exercising their powers in their own spheres of jurisdiction. This meant that it encroached substantially, even [translation] “to excess”, on provincial powers and that it was neither necessary nor rationally connected to the purpose of the ELRA (para. 135). On this point, he noted that this Court had determined in the FA Reference that the firearms registration system was tightly linked to the administration of the records of licences — which remain in the system — and to the purpose of promoting safety by reducing the misuse of firearms (ibid.). He added that it would be possible to decriminalize the failure to register long guns without destroying the data (para. 140). Moreover, none of the earlier bills had provided for their destruction (para. 141).

(3)      Quebec’s Right to Obtain the Data

[82]                          Having found s. 29 to be unconstitutional, Blanchard J. then considered the transfer of the data requested by the AGQ. He found that, because the data to which s. 29 applied were the product of a partnership between Canada and the provinces, they had no [translation] “owner” and did not fall within the jurisdiction of a single level of government. As a result, Quebec was entitled to obtain them (paras. 148-49). He said the following in this regard: “It goes against common sense — not to say the common good — to prevent Quebec from using data that it took part in gathering, analyzing, organizing, and modifying” (para. 151).

[83]                          Blanchard J. pointed out that the data had been collected using forms that clearly disclosed both the interaction between federal and provincial governments and the intended use of the data, namely the enforcement of firearms legislation (para. 154). The fact that it is Quebec, rather than Canada, that would from now on be gathering these data for an identical purpose [translation] “[cannot] give rise to a legitimate objection” (para. 155).

B.       Quebec Court of Appeal, 2013 QCCA 1138

[84]                          On June 27, 2013, the Quebec Court of Appeal unanimously allowed the appeal of the Attorney General of Canada (“AGC”). It reversed the trial judge’s decision and dismissed the AGQ’s motion for a declaration of constitutional invalidity.

(1)      Facts

[85]                          The Court of Appeal first found that the Superior Court’s judgment was confused and erroneous as regards the nature of the registry. It stated that Blanchard J. had confused the CFIS with the CFR (para. 28): it was of the opinion that [translation] “the CFIS is nothing more than a system providing electronic access to the registries kept by the Registrar and the Chief Firearms Officers” (para. 31 (CanLII)). The Court of Appeal also rejected certain other findings of fact made by the trial judge, particularly those to the effect that the CFR resulted from a partnership between the provinces and the federal government, and that Quebec’s CFO had a certain control over the CFR and had contributed to it (paras. 29 and 56). The Court of Appeal found that the CFR was a registry that was maintained by the Registrar and remained independent of and distinct from the registries maintained by the CFOs. The CFIS merely enabled CFOs, like various federal, provincial and municipal public safety organizations, to access data in the CFR (para. 28).

(2)      Constitutionality of Section 29 of the ELRA

[86]                          Next, the Court of Appeal found that s. 29 falls within Parliament’s jurisdiction. In its view, it makes little sense that Parliament would not have the power to amend a statute it had enacted or to destroy data in a registry it had created (paras. 37 and 45).

[87]                          The Court of Appeal reversed the trial judge’s conclusion that s. 29 had been enacted to prevent Quebec from using the data in the registry. Moreover, it rejected an argument by the AGQ based on the Reference re Upper Churchill Water Rights Reversion Act, [1984] 1 S.C.R. 297 (“Upper Churchill”). In that case, the province of Newfoundland had enacted a statute under which the right to use water from and rights related to power generated at Churchill Falls were to revest in the Crown (Court of Appeal’s reasons, at para. 41). A previous provincial statute had assigned those rights to the Churchill Falls (Labrador) Corp., which had then signed a contract for the sale of power with Hydro-Québec. The province argued, among other things, that the assignment was inoperative on the basis of its power to repeal an existing statute. This Court concluded that the new statute was colourable legislation whose sole purpose was to amend the power contract, thereby depriving Hydro-Québec of its rights, and that that purpose was ultra vires the province. In the Court of Appeal’s view, Upper Churchill can be distinguished from the instant case, in which the federal government’s intention is to abolish a registry it itself created. Parliament is not preventing the provinces from creating their own registries, but is simply refusing to collaborate in such an initiative (ibid., at para. 42).

[88]                          Turning to the ancillary powers doctrine, the Court of Appeal found that, since the ELRA abolished a scheme the creation of which had been held to be valid, it could not encroach on provincial jurisdiction any more than did the Act that had created the scheme (para. 49). The court stated that the trial judge had erred in using the principle of co-operative federalism [translation] “not as a mere interpretive tool but as the legal basis for a declaration that section 29 of the Act was ‘of no force or effect’” (para. 54).

(3)      Quebec’s Right to Obtain the Data

[89]                          Finally, the Court of Appeal declined to find that Quebec was entitled to receive the data. Its decision was based upon three grounds.

[90]                          First, the Court of Appeal found that Quebec did not have any property rights in the CFR’s data, because it had no control over them (para. 55). According to the court, the evidence revealed that Quebec’s CFO had never contributed data to the CFR, and that the CFIS merely made it possible for users to access data gathered independently by the Registrar and the CFOs (para. 56).

[91]                          Second, the Court of Appeal expressed the opinion that the CFO does not represent Quebec in exercising his or her duties. Rather, the CFO’s powers flow from the federal statute — the FA  — that created the position (paras. 58-59).

[92]                          Third, the Court of Appeal held that there was no partnership between Quebec and Canada with regard to the collection and preservation of the data targeted by s. 29. The financial agreements between Quebec and Canada did not create a partnership; they simply provided that Quebec was to be compensated for certain costs it incurred in administering the FA  within its territory (paras. 61-63). Consequently, there was nothing to justify ordering a transfer.

[93]                          According to the Court of Appeal, [translation] “[t]he remedy determined by the judge — that is, to compel the federal government to continue to compile the data — was clearly inappropriate here. So was the obligation to transfer this data to a future provincial registry” (para. 64).

IV.    Issues

[94]                          In this Court, the AGQ does not dispute the federal government’s power to repeal the FA ’s provisions on the registration of long guns. As the Court of Appeal correctly stated, it would make little sense, and would be contrary to the principle of parliamentary sovereignty, if Parliament could not repeal a statute it itself had enacted. Nor is the provinces’ power to have a long-gun registry at issue here. Instead, the AGQ argues that s. 29 , which provides for the unilateral destruction of the data with respect to long guns, is unconstitutional. The AGQ also submits that the federal government cannot refuse to transfer these data to Quebec before destroying them.

[95]                          In light of the issues raised by this appeal, the Chief Justice of Canada stated the following constitutional question:

                    Is s. 29 of the Ending the Long-gun Registry Act, S.C. 2012, c. 6 , ultra vires Parliament having regard to its criminal law power under s. 91(27)  of the Constitution Act, 1867 ?

[96]                          To answer this question, we must consider four issues:

(1)               What are the constitutional basis and the nature of the registry at issue, particularly in light of the FA Reference?

(2)               Is there a federal-provincial partnership with respect to the organization and management of the CFP and, if so, what are the consequences of that partnership from the standpoint of federalism?

(3)               In light of the applicable constitutional analytical framework, is s. 29 of the ELRA invalid?

(4)               If so, what is the appropriate remedy in the case at bar?

V.       Analysis

A.       The Registry

(1)      Constitutional Basis of the Registry

[97]                          The analysis of the constitutional basis of the CFR must begin with this Court’s opinion in the FA Reference. In that case, Alberta challenged the validity of the FA  and, more specifically, of the provisions on licences and on the registration of long guns. It asked the Court to rule on the constitutional basis for the creation of a firearms registry by the federal government, and for the requirement that long guns be registered. Note that at the time of the FA Reference, the CFR had not yet been set up and the practical effects of the FA  therefore had yet to materialize.

[98]                          In a unanimous opinion, this Court held that the FA , as a whole, fell within Parliament’s jurisdiction over criminal law:

                    The [FA ] in “pith and substance” is directed to enhancing public safety by controlling access to firearms through prohibitions and penalties. This brings it under the federal criminal law power. While the law has regulatory aspects, they are secondary to its primary criminal law purpose. The intrusion of the law into the provincial jurisdiction over property and civil rights is not so excessive as to upset the balance of federalism. [Emphasis added.]

(FA Reference, at para. 4)

[99]                          The regulatory aspects mentioned by the Court in this passage include the FA ’s provisions on licensing, on the registration of long guns and on maintaining a registry. The maintaining of a long-gun registry is, at first blush, a matter within provincial jurisdiction. In principle, the regulation of property within a province stems from that province’s jurisdiction over property and civil rights, except insofar as the property also falls within a listed federal head of power: Constitution Act, 1867, ss. 91  and 92(13) . Thus, the constitutionality of these provisions is based upon the fact that they are ancillary to the FA ’s primary purpose, which relates to criminal law.

[100]                      Moreover, although the Court did not accept the existence of a parallel between the regulation of firearms in the FA  and the regulation by the provinces of property such as land titles and automobiles (FA Reference, at paras. 41-43), that parallel is nonetheless relevant as regards the provisions — considered individually — that provide for the collection of information on long guns. On the ancillary nature of those provisions, three findings of this Court are of particular importance here.

[101]                      First, the Court found that the FA  “[did] not significantly hinder the ability of the provinces to regulate the property and civil rights aspects of guns”: FA Reference, at para. 51. Thus, the provisions on the registration of firearms did not encroach unduly on provincial powers: see also para. 58.

[102]                      Second, there was no indication that Parliament had an improper motive or that it was taking over a provincial power under the guise of the criminal law: FA Reference, at para. 53.

[103]                      Third, in the Court’s opinion, the system for monitoring the transfer of firearms, as well as the misuse of such weapons, included two categories of provisions (registration and licensing) that were “integral and necessary to the operation of the scheme”: FA Reference, at para. 47.

[104]                      In short, this Court found that the FA , including the long-gun registration provisions, was valid because it did not upset the balance of federalism: see FA Reference, at paras. 4, 53 and 58. The conclusion that the CFR is constitutionally valid flows directly from that finding. The CFR can be understood in this light, but it is just as important to define its exact nature, that is, to properly define the limits of what is to be destroyed under s. 29 .

(2)      Nature of the Registry

[105]                      The trial judge and the Court of Appeal reached opposite conclusions regarding the nature of the registry at issue. Blanchard J. focused on the CFIS and found that the data it contained, including those to which s. 29 applied, came from both the Registrar and the CFOs. As for the Court of Appeal, it found that Blanchard J. had erred as regards the nature of the different types of registries provided for in the FA . In its view, the CFIS is merely a system for accessing data and is not a registry in itself.  Moreover, the CFOs did not contribute data to the CFR.

[106]                      It is true that the trial judge was at times unclear in his reasons in referring to the various types of registries established under the FA . However, this confusion did not affect his conclusion regarding the nature of the registry that is actually at issue here or the data it contains. By contrast, the Court of Appeal limited its analysis to the registries mentioned in the FA  but did not consider how they were set up and kept. Certain facts must be reviewed in order to correctly grasp the nature of the registry that is at the heart of this case.

[107]                      The FA and the regulations made under it certainly provide — and the AGC emphasizes this point — that the Registrar and the CFOs are to maintain distinct “records”: ss. 83(1) and 87(1). In practice, however, the government chose for the implementation of this legislation to create one central registry — the CFIS — containing information entered by both the Registrar and the CFOs. To suggest that the information contributed by the Registrar is independent of the information contributed by the CFOs is to disregard this reality, which is supported by the evidence Blanchard J. accepted. In our opinion, Blanchard J. was right to find that the CFIS is primarily a data bank, and not a mere system for accessing information.

[108]                      Extensive evidence was presented at trial demonstrating that — to borrow the words of the Privacy Commissioner of Canada — the CFIS is “a fully integrated, automated information system that is used to enter, analyze, maintain and store” all firearms-related information required under the FA : Review of the Personal Information Handling Practices of the Canadian Firearms Program, Final Report, August 29, 2001 (online), at p. 14 (referring to the CFRS, the predecessor of the CFIS). Thus, insofar as the CFIS contains data, it is reasonable to refer to it, too, as a “registry”.

[109]                      Furthermore, the Assistant Commissioner of the RCMP and Director General of the CFP described the CFIS as a [translation] “complex computerized central registry”: affidavit of Pierre Perron, at para. 5. The AGC himself acknowledged that the CFIS [translation] “contains all the information required by the FA  to be entered in one of the two distinct ‘records’ that must be kept”, and that this information comes from both the Registrar and the CFOs: Written contestation and arguments of the defendant Attorney General of Canada and the mis en cause Commissioner of Firearms and Registrar of Firearms, June 4, 2012, J.R., vol. 1, at para. 160 (emphasis in original). The information entered by the Registrar or by the CFOs can be analyzed by the latter through an interface integrated into the CFIS. Thus, the data in the CFIS are interrelated: affidavit of Pierre Perron, at paras. 27(b) and (d); cross-examination of Isabelle Boudreault (Quebec’s Chief Firearms Officer), May 8, 2012, J.R., vol. 4, at pp. 63-64.

[110]                      In our view, this finding is reinforced by two additional factors. First, the evidence shows that the RCMP expected it to take several months to delete the data contemplated in s. 29 of the ELRA from the CFIS without affecting the other data in that registry: affidavit of Pierre Perron, at pp. 70 et seq. (“Bill C-19 — Implementation Plan Summary”). This is indicative of the extent to which the data accumulated in the CFIS are interrelated and integrated, regardless of where they come from. Second, s. 29(2) of the ELRA requires that each CFO ensure “the destruction . . . of all records under their control related to the registration of [long guns]”. This statutory requirement presupposes that like the Registrar, the CFOs possess certain information with respect to the registration of firearms that they must destroy. The need to have the CFOs intervene in this way in the destruction process is still further evidence of the extent to which all the data relating to long guns are integrated.

[111]                      Furthermore, as Quebec’s CFO explained, the CFR and the CFOs’ records do not exist independently of the CFIS: [translation] “. . . it’s a shared registry, since my records, I keep them in the CFIS. I don’t have a registry at my level. It’s really a registry I keep in the CFIS” (cross-examination of Isabelle Boudreault, at p. 53). In fact, even though the FA  provides that the Registrar and the CFOs can modify only their own records, this restriction does not preclude them from co-operating to ensure that the data for which they are responsible are accurate. For example, Quebec’s CFO stated that when inspecting businesses, she sometimes finds errors in the entry of firearm registration numbers. Although she cannot herself correct the data relating to registration certificates in the CFIS, she notifies the Registrar, who then makes the necessary changes: ibid., at p. 66. The Registrar, the CFOs and their employees can also add comments in the CFIS regarding their own work: ibid., at pp. 82 et seq.; affidavit of Pierre Perron, at para. 27 (c).

[112]                      We should add that the information with respect to the registration of long guns becomes fully relevant and useful only once it is integrated with and linked to the data on individuals and their licences. Section 13  of the FA  provides that the Registrar cannot register a gun unless a CFO has already issued a licence. The issuance of registration certificates therefore depends directly on the work done by Quebec’s CFO and firearms officers in the issuance of licences.

[113]                      This leads us to conclude that Blanchard J. was right to state that the information entered into the CFR by the Registrar was [translation] “compiled” in the CFIS with the information entered by the CFOs in their records. He did not make a palpable and overriding error in his factual characterization of the CFIS. To this extent, the Court of Appeal was wrong to find that he had failed to distinguish the CFR from the CFIS. In fact, Blanchard J. had based his decision first and foremost on the existence of the CFIS, which he saw as a centralized registry that contains all the information collected by both the Registrar and the CFOs (paras. 28 and 82). As the evidence shows, even though the FA  provides for distinct registries, their implementation depends upon the existence of one common registry, the CFIS. Although s. 29 provides for the destruction of data in the CFR, in light of how the FA  was implemented, the data in question are in fact the data in the CFIS related to the registration of long guns. The data that the province of Quebec wishes to obtain and preserve are therefore the data on long guns in the CFIS that have a connection with Quebec.

[114]                      In the end, Blanchard J.’s analysis of the factual background to this appeal is valid on the essential points. The Court of Appeal’s approach, on the other hand, is based upon a compartmentalized view of the FA . The different aspects of the scheme established by the FA  do not operate in isolation. There is an undeniable interaction between the Registrar and the CFOs, which is reflected in particular in the fact that their respective records are interrelated. This interaction is one aspect of the federal-provincial partnership upon which Blanchard J. focused, and which we will now discuss.

B.       Federal-Provincial Partnership

[115]                      The AGQ relies, as did Blanchard J., on the existence of a partnership in the broad sense in relation to the application of the FA , one that goes beyond the simple collection of the data in the CFR. He refers to numerous documents from federal government bodies that include expressions of this view of relations between the two levels of government in respect of the implementation of the FA . The AGC agrees instead with the Court of Appeal, arguing that Quebec did not participate in gathering the data in the CFR. He adds that the agreements entered into between Quebec and Canada in relation to the administration of the FA  were strictly financial in nature. Thus, the partnership upon which the appellant relies never existed.

[116]                      In our opinion, Blanchard J. was right to find that there was a partnership between Canada and Quebec in the case at bar. Although it is true that Quebec did not enter the information related to the registration of long guns in the CFIS database, this fact is not incompatible with the existence of a partnership between Quebec and Canada, given that the partnership was not based solely on the entry of the information. We must consider the nature and consequences of the partnership, and the importance to it of the data to which s. 29 applies. This partnership has been at the heart of the evolving implementation of the FA  since the very beginning.

[117]                      First of all, even before the FA  came into force, federal government ministers had observed that it would be necessary to work together with the provinces to administer the legislation. Then, in the first year after it was implemented, the Registrar stated that he was “working closely with [his] provincial . . . partners”: RCMP, Report on the Administration of the Firearms Act to the Solicitor General by the Registrar: Canadian Firearms Registry (1999), at p. iv.

[118]                      This co-operative view continued to ground relations between the various interested parties in subsequent years. For example, the RCMP commented as follows in 2010, when the provisions of the FA  concerning the registration of long guns were still in force: “Effective delivery of the CFP depends upon partnerships involving the federal and provincial governments and law enforcement agencies” (CFP Evaluation, at p. 8).

[119]                      The same view continued to be expressed even after the tabling of Bill C-19, which would become the ELRA:

                    Delivery of the CFP depends on the existing partnership between the federal government, the provincial governments and law enforcement agencies. The federal government contributes funds to the provinces that are responsible for the administrative work relating to various aspects of the Act, such as processing licences and registration certificates.

(Legislative Summary, at p. 4)

[120]                      As can be seen from these passages as well as from the declarations to the House of Commons quoted above, the implementation of the FA  and the CFP required provincial collaboration. In this regard, the evidence in the case at bar concerned primarily the specific relationship that developed between the federal government and Quebec. Accordingly, we will concentrate our analysis on the partnership at issue here.

[121]                      This partnership is reflected first in the construction of the CFIS database. As we mentioned above, Quebec, through its CFO, collaborated in the CFIS by gathering, analyzing, organizing and modifying certain data that it contained. The validity of this finding is not weakened by the fact that Quebec did not “enter” certain information in the CFIS, since it was a system that contained interrelated data: affidavit of Pierre Perron, at paras. 27(b) to (e); cross-examination of Isabelle Boudreault, at pp. 63-64.

[122]                      Moreover, Quebec’s CFO supported the Registrar’s efforts to create a reliable and comprehensive database that would be updated daily. He or she participated in each stage of firearms registration and worked to ensure that when the Registrar issued a registration certificate, it was done in conformity with the FA . Furthermore, one CFO explained that when she revoked a licence to possess firearms (under s. 70  FA ), she alerted the Registrar, who then had to revoke the registration certificate: cross-examination of Isabelle Boudreault, at pp. 83-84; re-examination of Isabelle Boudreault, May 8, 2012, J.R., vol. 4, at p. 153. We should also point out that s. 29(2) of the ELRA provides that each CFO must ensure “the destruction . . . of all records under their control related to the registration of [long guns]”. Thus, according to the ELRA, the collaboration of the CFOs is needed up to the time of the abolition of the requirement to destroy the data related to the registration of long guns.

[123]                      The AGC argues that the CFOs are federal agents who exercise powers conferred on them by federal legislation, even where the provinces choose to designate their own CFOs. He also points out that this position is financed by the federal government. In the AGC’s view, this leads to the conclusion that the provincial component of the partnership to which the AGQ refers is in reality federal.

[124]                      This argument must fail. The AGC is distorting the CFOs’ role. Although it is true that, strictly speaking, the CFOs exercise powers conferred on them by federal legislation, the fact remains that several of them were designated by the provinces. In Quebec’s case, the province set up, managed and supervised the office, and also added to its incumbent’s responsibilities.

[125]                      For example, Quebec’s CFO performs functions delegated to him or her by Quebec legislation: An Act to protect persons with regard to activities involving firearms, CQLR, c. P-38.0001 (“Anastasia’s Law”); An Act respecting safety in sports, CQLR, c. S-3.1, s. 46.31; Regulation respecting the exclusion of certain premises and certain means of transportation and respecting the exemption of certain persons, CQLR, c. P-38.0001, r. 1; Delegation of powers of the provincial minister to the Chief Firearms Officer, J.R., vol. 11, at p. 160. Thus, Anastasia’s Law provides that Quebec’s CFO must be informed of any application to a court to obtain that a person refusing to undergo a psychiatric assessment whose mental state is considered to present a danger to that person or to other persons be submitted to such assessment, or that the person be confined against his or her will in a health institution: s. 11; see also art. 778 of the Code of Civil Procedure, CQLR, c. C-25. The CFO must then verify “whether the person is in possession of a firearm, has access to a firearm or holds a licence to acquire a firearm”: s. 11 of Anastasia’s Law. This verification with respect to possession is carried out on the basis of the data on registration certificates: affidavit of Isabelle Boudreault, March 8, 2012, J.R., vol. 3, at paras. 59-64. The federal government has never disputed Quebec’s use of CFIS data for purposes of the administration of its provincial legislation.

[126]                      Moreover, Quebec police officers, particularly officers of the Sûreté du Québec, also play an important role in the CFP. In fact, Quebec’s CFO is a Sûreté du Québec officer. One example of this role is that they use a computer system to identify individuals involved in certain incidents to which they have responded. In such a case, they register the incident in their system, and an alert is forwarded electronically via the CFIS to the CFO, who decides whether the incident will have an impact on the individual’s eligibility to hold a licence and, consequently, a registration certificate: affidavit of Isabelle Boudreault, at paras. 13-15; cross-examination of Pierre Perron, May 1, 2012, J.R., vol. 3, at pp. 102 et seq.; re-examination of Pierre Perron, ibid., at pp. 126-27. The effectiveness of the entire registration system thus depends on the work of the CFO and, more generally, of the police.

[127]                      In addition, the Quebec police, like all police forces in Canada, have access to a component of the CFIS that assists their officers in responding to calls and conducting investigations: CFP Evaluation, at p. 11. It enables officers, before answering a call in a residence, to confirm the possible presence of firearms and determine whether they are possessed legally: ibid., at p. 17. In 2011, Quebec police officers consulted this component of the CFIS more than 700 times a day on average: affidavit of Isabelle Boudreault, at para. 65. Thus, the application of certain Quebec legislation and the implementation of Quebec police procedures depend to a large extent upon access to CFIS data.

[128]                      It is thus clear that Quebec uses data from the CFIS for the purposes of its provincial legislation. This goes without saying, given that, as Blanchard J. mentioned several times, those data are the product of collaboration. They are also at the heart of a partnership that is grounded in the exercise of both federal and provincial powers. It is in this sense that Blanchard J.’s conclusion that the data are not purely “federal” must be understood. We should mention that Blanchard J. concluded that Quebec most likely participated in this partnership on the assumption that the federal government would not be able to unilaterally destroy its fruits (para. 35).

[129]                      In short, the CFO’s powers under the FA  overlap with his or her powers under provincial legislation. To argue that the position of CFO is of purely federal origin is to disregard this reality.

[130]                      Moreover, s. 95 of the FA  provides that the federal government may enter into agreements with provincial governments to compensate them for administrative costs they have incurred in relation, inter alia, to processing licences and registration certificates. The federal government did in fact enter into such agreements with Quebec. It can be seen from the Canada-Quebec financial agreement with respect to the administration of the FA  (“Accord financier Canada-Québec relatif à l’administration de la Loi sur les armes à feu”) for the period from April 1, 2006 to March 31, 2010, agreement No. 2012-004 between the two governments, that although the agreement’s primary purpose was to set out financial terms, it also contained clauses with a broader scope. For example, art. 5 provided that Quebec agreed to deliver all necessary services, described in the agreement, relating to the application of the FA  within its jurisdiction. Under art. 7, Canada was required to provide Quebec with a functioning and operational CFIS together with its related components. Finally, art. 23 provided that [translation] “[a]ny information collected by Canada or by Quebec under this Agreement is subject to the rights and protections set out in federal and Quebec legislation concerning access to information and the protection of personal information.”

[131]                      It should also be noted that notwithstanding the possibility of compensation from the federal government, the provinces that participated in the CFP, including Quebec, had to invest considerable amounts in relation to the implementation of the FA , as can be seen from an RCMP report:

                    It appears as if the opt-out provinces/territories, where the RCMP is the provincial/territorial police service, benefit substantially from the current cost arrangement, in terms of dollar value. The other opt-in provinces, where there are other provincial police services, have made substantial investments to benefit from the weight and services of the program . . . .

(CFP Evaluation, at p. 68)

It would therefore seem incorrect to say that the province of Quebec made no financial contribution whatsoever to firearms control and to the administration of the FA . By focusing on the financing of the CFOs, the AGC disregards the broader implications and consequences of the CFP. We cannot accept this approach.

[132]                      Indeed, it is our opinion that the AGC is mistaken as regards the nature of the partnership relied on by the AGQ. This partnership involves more than just the entry of data in the CFIS and the functions performed by Quebec’s CFO under the FA . It relates to the whole of the CFP and also includes the role played by Quebec and its police officers in regulating firearms in collaboration with Canada.

[133]                      Through the role played by the CFO and by Quebec’s police forces, and aided by the CFIS (whose construction it contributed toward), the province of Quebec has for years implemented its own view of firearms control from the perspective of the administration of justice and health. Moreover, the RCMP recognized the possibility of the provinces adapting the program to their needs, as can be seen from the following passage from a report we cited above:

                    The main benefit cited for creating the opt-in program was the promotion of a program delivery model that met an identified need for the provinces to be able to adapt the delivery of the program to local circumstances. In this area, it was believed that the opt-in program has been successful; service delivery has been aligned to meet the service delivery priorities of the provincial governments, and be supported to work with other areas of provincial jurisdiction, such as justice and policing, and health. In Québec, this culminates in an integrated service delivery model, and a profound interest to improve the program to meet heightened provincial public safety and health expectations. [Emphasis added.]

(CFP Evaluation, at p. 67)

[134]                      We find from the foregoing that a partnership has developed between the federal government and Quebec that is structured around the following main elements:

-     the design, implementation and administration of the CFP;

-     the compiling and pooling of data by the Registrar and the CFO together with a mutually guaranteed right of access to the data in question;

-     the use of CFIS data for federal and provincial purposes; and

-     the joint and complementary exercise of provincial and federal powers in the context of firearms control.

[135]                      In our opinion, this factual reality unique to the CFIS and the partnership of which it forms a part is essential to the determination of whether s. 29 of the ELRA is constitutional. Before turning to the analysis on that subject, we will briefly discuss the framework applicable to it.

C.       Analytical Framework Applicable to the Division of Powers

(1)      Pith and Substance

[136]                      When the constitutionality of a statutory provision is challenged on the basis of the division of powers, courts turn to the “pith and substance” doctrine. To apply this doctrine, they must review the extent to which the impugned provision intrudes on the powers of the other level of government: Kirkbi AG v. Ritvik Holdings Inc., 2005 SCC 65, [2005] 3 S.C.R. 302, at paras. 20-21. According to the principles developed by this Court, it is necessary first to identify the impugned provision’s “pith and substance”, and then to determine whether the provision falls within a head of power of the level of government at which it was enacted: see Reference re Securities Act, 2011 SCC 66, [2011] 3 S.C.R. 837, at para. 63; Quebec (Attorney General) v. Lacombe, 2010 SCC 38, [2010] 2 S.C.R. 453, at para. 20; FA Reference, at para. 16; Kitkatla Band v. British Columbia (Minister of Small Business, Tourism and Culture), 2002 SCC 31, [2002] 2 S.C.R. 146, at paras. 52-58; see also H. Brun, G. Tremblay and E. Brouillet, Droit constitutionnel (6th ed. 2014), at p. 463; P. W. Hogg, Constitutional Law of Canada (5th ed. Supp.), at pp. 15-5 to 15-7.

[137]                      A provision’s pith and substance is its “dominant characteristic”: see Reference re Same-Sex Marriage, 2004 SCC 79, [2004] 3 S.C.R. 698, at para. 13. In this Court’s words, “[a] pith and substance analysis looks at both (1) the purpose of the legislation as well as (2) its effect”: Kitkatla Band, at para. 53; see also Reference re Securities Act, at paras. 63-64; Reference re Assisted Human Reproduction Act, 2010 SCC 61, [2010] 3 S.C.R. 457, at para. 199.

[138]                      To establish the purpose of legislation or of a statutory provision, “the Court may look at both intrinsic evidence, such as purpose clauses, or extrinsic evidence, such as Hansard or the minutes of parliamentary committees”: Kitkatla Band, at para. 53. In this regard, the Court has held that it is necessary to “seek to ascertain the true purpose of the legislation, as opposed to its mere stated or apparent purpose”: Canadian Western Bank v. Alberta, 2007 SCC 22, [2007] 2 S.C.R. 3, at para. 27 (emphasis in original), citing Attorney-General for Ontario v. Reciprocal Insurers, [1924] A.C. 328 (P.C.), at p. 337. Thus, in any division of powers case, the intention behind a statute, that is, the purpose actually being pursued by the legislature, can cause the entire statute to be invalid: see, e.g., R. v. Big M Drug Mart Ltd., [1985] 1 S.C.R. 295; R. v. Edwards Books and Art Ltd., [1986] 2 S.C.R. 713.

[139]                      As for the review of the provision’s effects, it concerns both those legal and practical, which include direct effects of the provision and the “side” effects flowing from its application: Kitkatla Band, at para. 54; Reference re Securities Act, at para. 64; Lacombe, at para. 20. The foregoing analysis as a whole makes it possible to determine the constitutional nature of the provision, and to assess whether it encroaches on the powers of the other level of government.

(2)      Ancillary Powers

[140]                      Where, because of its pith and substance, a provision found in an otherwise valid statute encroaches on the jurisdiction of the other level of government, the Court must determine whether the encroachment is ancillary: see Reference re Assisted Human Reproduction Act, at paras. 187-89. If the encroachment is in fact ancillary, the provision will be considered valid provided that it is sufficiently integrated into the scheme of the statute: Lacombe, at paras. 34-36.

[141]                      The degree of integration of a provision that is needed for an encroachment to be considered ancillary varies with the seriousness, or extent, of the encroachment. If the encroachment of the impugned provision on the jurisdiction of the other level of government is merely “marginal” or “limited”, a functional relationship between the provision and the statutory scheme may suffice. If, on the other hand, the federal provision is “highly” intrusive vis-à-vis provincial powers, a stricter test of necessity will apply: General Motors, at pp. 668-70.

[142]                      More generally, we wish to repeat the following comment about the ancillary powers doctrine made by the Court in Lacombe:

                    . . . the availability of ancillary powers is limited to situations in which the intrusion on the powers of the other level of government is justified by the important role that the extrajurisdictional provision plays in a valid legislative scheme. The relation cannot be insubstantial . . . . [Emphasis added; para. 35.]

[143]                      Thus, the Court recognized that a government at one level can validly pass legislation on a matter within its jurisdiction even though the legislation intrudes upon the other level’s jurisdiction. Yet the constitutional doctrines that make such an intrusion permissible — namely the pith and substance and ancillary powers doctrines — cannot be fully understood without considering the constitutional principles upon which they are based: Reference re Assisted Human Reproduction Act, at para. 196. We will now turn to these principles, in particular that of federalism.

(3)      Unwritten Principles: Federalism

[144]                      The Court has, while stressing that our written Constitution is paramount, recognized the importance of the unwritten principles that underlie it. These principles infuse the analysis and interpretation of the division of powers. They “assist in the interpretation of the text and the delineation of spheres of jurisdiction, the scope of rights and obligations, and the role of our political institutions”: Reference re Secession of Quebec, [1998] 2 S.C.R. 217, at para. 52. They also reflect our Constitution’s historical context and have facilitated its application throughout its history; thus, they are to the Constitution what sap is to a tree. In this sense, they are “a necessary part of our Constitution because problems or situations may arise which are not expressly dealt with by the text of the Constitution”: ibid., at para. 32.

[145]                      As this Court explained in Reference re Secession of Quebec, federalism “was a legal response to the underlying political and cultural realities that existed at Confederation and continue to exist today”, and “political leaders [had] told their respective communities that the Canadian union would be able to reconcile diversity with unity”: para. 43. The principle of federalism requires that the constitutional division of powers be respected and that a balance be maintained between federal and provincial powers. One “power may not be used in a manner that effectively eviscerates another”: Reference re Securities Act, at para. 7; Reference re Secession of Quebec, at paras. 57-58.

[146]                      According to the “classical” approach favoured by the Judicial Committee of the Privy Council until 1949, the heads of power constituted “watertight compartments”, and overlaps between them were to be avoided to the extent possible: Reference re Securities Act, at para. 56.

[147]                      The modern view of federalism rejects this approach and replaces it with a more flexible conception of the division of powers that is dominant in this Court’s recent jurisprudence. This conception “recognizes that in practice there is significant overlap between the federal and provincial areas of jurisdiction, and provides that both governments should be permitted to legislate for their own valid purposes in these areas of overlap”: Canada (Attorney General) v. PHS Community Services Society, 2011 SCC 44, [2011] 3 S.C.R. 134, at para. 62; Canadian Western Bank, at paras. 36-37; Reference re Securities Act, at para. 57. Such a conception thus facilitates intergovernmental co-operation: see OPSEU v. Ontario (Attorney General), [1987] 2 S.C.R. 2, at pp. 19-20, per Dickson C.J.; Reference re Securities Act, at paras. 57-58. Both in law and in the political arena, the concept of “co-operative federalism” has been developed to adapt the principle of federalism to this modern reality.

(4)           Recognition of Co-operative Federalism

[148]                      Co-operative federalism reflects the realities of an increasingly complex society that requires the enactment of co-ordinated federal and provincial legislative schemes to better deal with the local needs of unity and diversity: PHS, at para. 63; Tsilhqot’in Nation v. British Columbia, 2014 SCC 44, [2014] 2 S.C.R. 257, at paras. 148-49. From a legal perspective, it is by allowing for overlapping powers through the application of the pith and substance and ancillary powers doctrines that co-operative federalism is able to meet those needs and, in this sense, to enable the goals of federalism to be realized.

[149]                      In our opinion, the federal-provincial partnership with respect to firearms control is consistent with the spirit of co-operative federalism. This partnership has enabled the federal and provincial governments to work together, rather than in isolation, to achieve both federal (criminal law) and provincial (public safety and administration of justice) purposes.

[150]                      In the case at bar, this Court is not being asked to hold that a co-operative scheme based on the powers of both levels of government is valid. Rather, the AGQ is challenging the validity of a provision that contributes to the dismantling of a partnership that was created in the spirit of co-operative federalism. This situation is unique and novel.

(5)           The End of a Partnership Based on “Co-operative Federalism”

[151]                      In the novel circumstances of this case, our analysis must be guided by the Constitution’s unwritten principles. In particular, we must be careful not to place the principle of federalism and its modern form — co-operative federalism — in jeopardy. As the Court has noted, “the very functioning of Canada’s federal system must continually be reassessed in light of the fundamental values it was designed to serve”: Canadian Western Bank, at para. 23.

[152]                      The dominant tide with respect to the division of powers admits of overlapping powers and favours co-operation between the different levels of governments. It also supports the validity of schemes established jointly through partnerships developed between members of our federation. In our opinion, our courts must protect such schemes both when they are implemented and when they are dismantled. It would hardly make sense to encourage co-operation and find that schemes established in the context of a partnership are valid while at the same time refusing to take this particular context into account when those schemes are terminated.

[153]                      In our opinion, the dismantling of a partnership like the one established with respect to gun control must be carried out in a manner that is compatible with the principle of federalism that underlies our Constitution. Thus, Parliament or a provincial legislature cannot adopt legislation to terminate such a partnership without taking into account the reasonably foreseeable consequences of the decision to do so for the other partner. The courts must, in considering whether legislation or a statutory provision having as its purpose to dismantle the partnership is constitutional, be aware of the impact of that legislation or provision on the other partner’s exercise of its powers, especially when the partner that terminates the relationship is intentionally bringing about that impact.

[154]                      In other words, a co-operative scheme from which both the federal and provincial governments benefit cannot be dismantled unilaterally by one of the parties without taking the impact of such a decision on its partner’s heads of power into account. To conclude otherwise would be to accept a one-way form of co-operative federalism. That would upset the balance between, on the one hand, the principle of co-operative federalism — which permits a government at one level to pass laws that have an impact on the powers of the other level — and, on the other hand, the doctrine of interjurisdictional immunity — which is inherent in the principle of federalism. The requirement that a government respect the other level’s powers is all the more important when the parties have chosen to form a partnership. In a co-operative context, actions of a government at one level can have serious consequences for the other level. It is therefore necessary to show vigilance for the increased risk of disrupting the constitutional balance that is protected by the principle of federalism. The concern here is not to alter the separation of powers in our Constitution through the application of co-operative federalism, but to ensure that it is respected. The Court explained this as follows in Reference re Securities Act:

                    Yet we may appropriately note the growing practice of resolving the complex governance problems that arise in federations, not by the bare logic of either/or, but by seeking cooperative solutions that meet the needs of the country as a whole as well as its constituent parts.

                          Such an approach is supported by the Canadian constitutional principles and by the practice adopted by the federal and provincial governments in other fields of activities. The backbone of these schemes is the respect that each level of government has for each other’s own sphere of jurisdiction. Cooperation is the animating force. The federalism principle upon which Canada’s constitutional framework rests demands nothing less. [Emphasis added; paras. 132-33.]

[155]                      We could not state these principles more clearly than the Court has already done in Reference re Assisted Human Reproduction Act:

                          In short, care must be taken to maintain the constitutional balance of powers at all stages of the constitutional analysis. Be it in identifying the pith and substance of a statute or a provision or in reviewing the limits of an assigned power or of the exercise of an ancillary power, the courts must bear the importance of the unwritten constitutional principles in mind and must adhere to them. [para. 196]

[156]                      Thus, subject to adherence to the above-mentioned principles, the courts nonetheless must not stray from the analytical framework applicable to the division of powers. To determine whether s. 29 of the ELRA is valid, it is therefore necessary to consider that framework and these principles.

D.       Constitutionality of Section 29 of the ELRA

[157]                      In the instant case, the Court of Appeal strayed from the analytical approach the courts must take. It is true that Parliament can repeal or amend legislation it has validly enacted under one of its heads of power: Reference re Canada Assistance Plan (B.C.), [1991] 2 S.C.R. 525, at pp. 548-49. Nevertheless, the courts must consider the impugned provision or legislation to determine whether, in pith and substance, all that it does in fact is repeal or amend legislation that was validly enacted.

[158]                      On the one hand, it is not enough in this regard to say, as the Court of Appeal did, that the legislative measure is merely repealing legislation. That does not resolve the issue of whether it is constitutional. In Upper Churchill, this Court held that repealing legislation can, as a result of its purpose and its effect, be ultra vires the government that enacted it: see also, on this subject, P. Daly, “Dismantling Regulatory Structures: Canada’s Long-Gun Registry as Case Study” (2014), 33 N.J.C.L. 169.

[159]                      On the other hand, a statutory provision might be enacted to abolish a scheme that was incidental to the jurisdiction of the legislature that initially created it. In such a case, the legislature in question might in theory wish to eliminate an encroachment on a head of power that does not belong to it. But it is quite possible that what the legislature intends to do to eliminate that initial encroachment in fact encroaches to an even greater extent on that head of power. This is all the more likely to occur in the context of a partnership.

[160]                      As a result, whether s. 29 of the ELRA is constitutional must be determined in the same way as for any legislation enacted by Parliament. To do this, we must identify the pith and substance of that section and link it to a head of legislative power.

(1)      Analysis Regarding the Pith and Substance of Section 29

[161]                      At first glance, the words of s. 29 of the ELRA pose no particular problems of interpretation. The provision orders the Registrar and the CFOs to destroy, as soon as feasible, all records in the CFR that relate to the registration of long guns or are under their control.

[162]                      From a structural standpoint, s. 29 of the ELRA is distinct from ss. 2  to 28 . The latter sections amend the Criminal Code  and the FA  to abolish the requirement to register long guns and decriminalize the failure to comply with that requirement. Section 29 , on the other hand, is a transitional provision. It is also relevant that the destruction of the data under s. 29  is exempted from the application of certain provisions of the Library and Archives of Canada Act, S.C. 2004, c. 11 , and the Privacy Act, R.S.C. 1985, c. P-21 : s. 29(3)  ELRA.

[163]                      From a practical and legal standpoint, the principal effect of s. 29 of the ELRA is to delete the data in the CFR, and therefore in the CFIS, forever. Although it is true that s. 84  of the FA  provides that “[t]he Registrar may destroy records kept in the [CFR] at such times and in such circumstances as may be prescribed”, s. 29  cannot, as the AGC argues, be compared to s. 84 . First, the applicable regulations provide that a record of a registration certificate that is issued or revoked may not be destroyed: s. 4(1) of the Firearms Records Regulations. Second, the destruction of data on a massive scale would be contrary to the very purpose of the FA .

[164]                      Moreover, the analysis cannot be limited to a finding that s. 29 provides for the destruction of data. The federal government’s decision to destroy the data without first transferring them to its partners, such as Quebec, will have serious consequences that are relevant to the question whether s. 29 is constitutional.

[165]                      Considered prospectively, the destruction of the data without first transferring them will ensure that the information contained in the records cannot be used by any subsequent federal government, or by the provinces or the police organizations that have, until now, used it for various purposes. In particular, the evidence in the record shows that the destruction of data required by s. 29 will prevent the provinces from using the data to create their own registries.

[166]                      Moreover, this destruction will retrospectively obliterate Quebec’s collaboration in the construction of the long-gun database. It will also interfere with the current use of the data by Quebec that we discussed above.

[167]                      The extrinsic evidence shows that the purpose being pursued in enacting s. 29 of the ELRA was indeed to prevent the provinces from using the data. This intention is confirmed by some passages from the parliamentary debate that were quoted by Blanchard J.:

                    Mr. Speaker, this government committed to eliminating the ineffective long gun registry and we do not intend to help other levels of government create registries.

(House of Commons Debates, vol. 146, No. 041, 1st Sess., 41st Parl., November 1, 2011, at p. 2799, The Rt. Hon. Stephen Harper, Prime Minister of Canada)

                    Our government has no intention of transferring the information that it has in its offices to the provinces, nor will it make available that same information to be used by future governments to be re-enacted or brought back in the future.

(Ibid., at p. 2779, The Hon. Peter Penashue, Minister of Intergovernmental Affairs)

                    We will fulfill the promise that we made and that includes doing the right thing and ensuring that no other government could use the information to resurrect the failed long gun registry.

(Ibid., at p. 2780, Kerry-Lynne D. Findlay, Parliamentary Secretary to the Minister of Justice)

[168]                      These declarations by the Prime Minister and by federal ministers show that the purpose of s. 29 was in fact to impose these prospective and retrospective effects of the destruction of the data on the provinces. Blanchard J. was therefore right to find that Parliament’s intention in destroying the data was to hinder the provinces. This intention can be seen in how, and how quickly, Parliament wished to act. Allow us to specify that it is not so much the destruction of the data itself that is problematic, but the fact that the data are to be destroyed unilaterally combined with the effect of doing so, that is, inter alia, to prevent the provinces from using them.

[169]                      In light of the purpose and the effect of s. 29, therefore, the scope of the section is broader than the mere destruction of the data. Section 29 has harmful consequences for the federal government’s partners. It is these consequences that require, having regard for the principle of federalism, a close consideration of the provision’s pith and substance and the powers upon which the AGC relies in support of its being valid.

[170]                      The AGC submits that s. 29 falls within the federal criminal law power because its purpose relates to the repeal of part of the FA , that is, to the abolition of the long-gun registry. A legislative measure must meet three criteria to fall within that head of power. First, it must have a valid public purpose. Second, it must pursue that purpose by means of prohibitions. Third, the prohibitions must be supported by criminal sanctions: see Reference re Assisted Human Reproduction Act, at para. 233. It is impossible to conclude that these three criteria are met in the case of s. 29, however, since what is at issue in the case at bar is the power to repeal a statute that was enacted under that head of power.

[171]                      Yet we cannot find that the purpose of s. 29 relates to the repeal of the FA . First, as can be seen from the structure of the ELRA, the abolition by the federal government of the requirement to register long guns and the destruction of the data are two distinct objectives. Second, the purpose of s. 29  is not simply to destroy the data. Rather, it seeks to abolish a scheme in which those data are used and to terminate the partnership with Quebec.

[172]                      Nor, in this regard, can we conclude, as the Court of Appeal did and as our colleagues do, that s. 29 merely abolishes a “federal” scheme. We believe that this conclusion is based on the mistaken premise that the scheme abolished by s. 29 was created solely by the FA . Although it is true that the FA  constitutes the statutory framework for the CFR, it must not be forgotten that that framework encroached on provincial jurisdiction at the time of its creation. Even more importantly, the CFR is only one component of the federal-provincial partnership for the control of long guns. As we showed above, this partnership, considered in its broad sense, required Quebec’s administrative, financial and legislative participation.

[173]                      We should also mention that unlike the situation in Reference re Canada Assistance Plan, which concerned the federal spending power, both the setting up and the maintenance of the CFR at issue in the case at bar gave it a provincial aspect. As the trial judge found, the data to which s. 29 applies are not only “federal”. The creation of the database, the CFIS, was dependent on co-operation between the provinces and the federal government. The interrelated information it contains was accumulated in the context of the close federal-provincial partnership described above. In this context, s. 29 of the ELRA was, in requiring that the data be destroyed without first being transferred to Quebec, also intended to abolish the “provincial” component of the scheme, both in the broad sense and in respect of the data.

[174]                      Finally, the AGC asserts that the purpose of s. 29 is to eliminate an invasion by the state of the privacy of Canadians. If that were really the case, it is hard to understand why that section was inserted only in the final version of the bill, and only after Quebec had announced that it wanted to create its own registry. Quite obviously, in light of the extrinsic evidence, we do not believe the elimination of an invasion of privacy to have been the federal government’s purpose.

[175]                      What, then, is the pith and substance of s. 29 of the ELRA?

[176]                      In our opinion, in light of its purpose and its effect, s. 29 goes much farther than simply requiring that the data be destroyed. Given that they are to be destroyed with no possibility of their first being transferred to the partners, and therefore without the impact of this measure on the partners’ exercise of their powers being taken into account, the section’s true purpose is to ensure that the information on long guns can no longer be used for any provincial purposes. As we mentioned above, however, the regulation of long guns and the use of information about them fall primarily within the provinces’ power to make laws in relation to property and civil rights. This is all the more true now that the failure to comply with the requirement to register has been decriminalized, because the use of the information has lost its federal aspect. As a result, the pith and substance of s. 29 relates to the provinces’ power over property and civil rights.

[177]                      To be valid, therefore, s. 29 must be integrated into the ELRA as a whole. This conclusion makes sense to us, given that whether the CFR is constitutional in fact depends on its being ancillary to a criminal law purpose, as the Court found in the FA Reference.

(2)      Ancillary Powers Analysis

[178]                      To determine whether s. 29 of the ELRA is constitutional on the basis of the ancillary powers doctrine, we must consider the seriousness, or extent, of its encroachment on provincial powers, bearing in mind that the provincial power to make laws in relation to property and civil rights is a head that should not be intruded upon lightly: General Motors, at pp. 672-73. We should mention that s. 29 may also encroach on the provinces’ powers in relation to the administration of justice, as can be seen from our analysis below.

[179]                      The seriousness of the encroachment of s. 29 must be analyzed on the basis of the specific factual and legal context of the instant case, which includes the existence of the partnership between the federal government and Quebec.

[180]                      By enacting the FA , the federal government effectively took over the field of firearms regulation. The decision to do so had two major consequences. On the one hand, the provinces had to deal with a national registry. As a result, the creation of a provincial long-gun registry became unnecessary, especially given that the existence of the federal legislation made it possible for the provinces to play a major role in the CFP and in the creation of registries and of databases such as the CFIS through the joint and complementary exercise of their powers in relation to property and civil rights and the administration of justice. As well, Quebec passed legislation based upon this partnership with the federal government. The creation and management of a federal firearms registry thus led to the development of a provincial aspect. To destroy the registry unilaterally without taking this reality into account is to negate Quebec’s financial, administrative and legislative participation.

[181]                      On the other hand, although the FA  did not have a significant effect on the provinces’ powers with respect to firearms at the time of its enactment, that is no longer true today in a context in which the data are to be destroyed without first being transferred to Quebec. The consequences of the loss of the information collected since 1998 will not be only financial in nature. The quality and usefulness of a restored registry created by Quebec would be diminished, since, to give one example, the chain of ownership of each weapon will have been lost. We note that in 2011, long guns represented more than 90 percent of registered firearms in Quebec: affidavit of Isabelle Boudreault, at paras. 18 and 31. Quebec maintains that, if it does not obtain and cannot retain the data concerning the registrations of interest to it, it will lose track of more than one and one half million long guns: Evidence of the Standing Committee on Public Safety and National Security, at p. 2. Thus, the destruction of the data would compromise the creation and the usefulness of a future Quebec firearms registry.

[182]                      Furthermore, the CFO’s role under the FA  and the partnership discussed above led to the enactment by Quebec of legislation that falls within its powers in relation to the administration of justice. Destruction of the data would therefore interfere with the use Quebec makes of them in the administration of certain provincial legislative schemes, such as that of Anastasia’s Law, and with their use by the police in investigations in particular.

[183]                      Finally, the federal government could have implemented its vision of firearms control without involving the provinces in any significant way, but it did not choose to take that approach. Doubtless for the sake of efficiency, it instead developed a legislative scheme that required active provincial participation and took the form of a partnership. This cannot be disregarded in analyzing the seriousness of the encroachment on the powers of the other level of government and, consequently, in determining the scope and validity of s. 29.

[184]                      This leads us to conclude that in terms of both its nature and its effects, s. 29 causes a substantial encroachment on provincial jurisdiction. In such a case, the Court’s past decisions tell us that for the encroachment of s. 29 to be found to be ancillary to the ELRA, the degree to which the section is integrated into the Act must be high, that is, it must satisfy the necessity or “integral part” criterion: General Motors, at p. 671; see also Northern Telecom Ltd. v. Communications Workers of Canada, [1980] 1 S.C.R. 115; Clark v. Canadian National Railway Co., [1988] 2 S.C.R. 680; R. v. Thomas Fuller Construction Co. (1958) Ltd., [1980] 1 S.C.R. 695.

[185]                      The purpose of the ELRA as a whole, with the exception of s. 29, is to abolish the requirement to register long guns. The fact that the “summary” of the ELRA refers to two purposes cannot serve on its own as proof of the true purpose of the ELRA. It goes without saying that the courts, in determining whether legislation is constitutionally valid, must look beyond the legislation’s declared or apparent purpose. Were they not to do so, it would be possible for a legislature to shield statutory provisions from constitutional challenge simply by drafting a preamble. Furthermore, it is only s. 29 of the ELRA that is related to the supposed second purpose of the Act.

[186]                      In our opinion, the destruction of the data in question in s. 29 cannot be considered necessary to the abolition of the requirement to register long guns. These two purposes are distinct. Moreover, as we mentioned above, the earlier bills did not provide for the destruction of these data.

[187]                      We also doubt that s. 29 can be linked to the ELRA on the basis of a test of rationality. Although the destruction of data in which the federal government is no longer interested might seem rational at first glance, we find it hard to reconcile the manner in which their destruction was provided for with the desire certain provinces might show to maintain a registry within the limits of their powers. The AGC in fact acknowledged at the hearing that what was problematic was not really the fact of transferring the data to the government of Quebec, but that of transmitting personal information about long-gun owners: transcript, at p. 55. We have already rejected this argument as an explanation of the refusal to transfer the data.

[188]                      Finally, we agree with Blanchard J. that Parliament’s declared intention to cause harm to the other level of government cannot be disregarded. This Court had written the following in the FA Reference:

                    This law does not allow the federal government to significantly expand its jurisdictional powers to the detriment of the provinces. There is no colourable intrusion into provincial jurisdiction, either in the sense that Parliament has an improper motive or that it is taking over provincial powers under the guise of the criminal law. [para. 53]

In this appeal, we cannot say the same thing about the ELRA, as the exact opposite is true.

[189]                      Although it is not necessary to take this into account for the purposes of our analysis, the intention to cause harm is all the more relevant in that it arises in a context in which Quebec and the federal government had agreed to act jointly, in the spirit of co-operative federalism, a principle whose source lies in the growing need for intergovernmental co-operation in order to enable the Canadian federation to act to address questions of a polycentric nature. In this context, it is hard to imagine how a provision whose purpose is to put an end to this co-operation and that is enacted with an intention to cause harm to a partner can be rational.

[190]                      The foregoing analysis leads to an inescapable conclusion: because of its pith and substance, s. 29 of the ELRA does not fall within the federal criminal law power and is not ancillary to the ELRA, which is otherwise valid. The section has not therefore been shown to be constitutionally valid. In our opinion, a legislative measure cannot be found to be valid that (1) does not fall within the federal criminal law power and that (2) thwarts, by the substantial encroachment it causes, the corollary exercise of provincial powers that flowed from the partnership. To conclude otherwise would be inconsistent with the principles of federalism.

[191]                      Our conclusion does not necessarily mean that Parliament cannot enact legislation to destroy the data. What we are saying is merely that to destroy the data, as provided for in s. 29, without first offering to transfer them is unconstitutional. This conclusion is therefore not incompatible with the principle of parliamentary sovereignty as set out in s. 42  of the Interpretation Act, R.S.C. 1985, c. I-21 , as parliamentary sovereignty is applicable only where Parliament has exercised one of its legislative powers validly, that is, in a manner consistent with the Constitution. That is not the case here.

E.        Appropriate Remedy

(1)      Declaration of Invalidity

[192]                      In light of this conclusion, s. 29 of the ELRA must be declared to be invalid under s. 52  of the Constitution Act, 1982: Canadian Western Bank, at para. 26.

[193]                      The Superior Court judge tailored the declaration of invalidity of s. 29 to limit it to data from Quebec with respect to long guns (that is, to data originating in Quebec or concerning citizens of Quebec, persons living there or persons who have committed acts involving firearms) (para. 194). Moreover, data from the other provinces relating to the registration of long guns have been destroyed since Blanchard J.’s judgment, on October 31, 2012: Court of Appeal’s reasons, at para. 9; transcript, at p. 87. Like Blanchard J., therefore, we would limit the declaration of invalidity to data with respect to long guns that have a connection with Quebec.

(2)      Order to Transfer the Data

[194]                      The trial judge also declared that Quebec was entitled to receive the data referred to in the declaration of invalidity within 30 days (para. 195). In support of this declaration, he cited Doucet-Boudreau v. Nova Scotia (Minister of Education), 2003 SCC 62, [2003] 3 S.C.R. 3, and found that, although that case concerned possible remedies under the Canadian Charter of Rights and Freedoms , the principles developed in it could serve as a guide in the case at bar (para. 163). Blanchard J. thus stated that [translation] “the remedy must be adapted to the circumstances of the case and the nature of the right infringed” (para. 164). He also noted “that an appropriate and just remedy should have the capacity to evolve to meet new challenges, [as a] traditional and historical practice must not create barriers to the requirements of a reasoned and compelling notion of the appropriate remedy” (para. 165). However, he expressed the following reservation:

                          [translation] With respect to constitutional remedies, the Court must show deference. It must correct the wrong by prescribing the appropriate solution, without going beyond what is required to remedy the violation of the constitutional right. [para. 167]

[195]                      Because of the [translation] “remarkable history of compliance with court decisions by . . . governmental institutions”, Blanchard J. found that it would be inappropriate to issue an injunction against the AGC (para. 190). He stated that a declaratory judgment “will certainly suffice as an efficient and effective determination of the rights and obligations of the parties” (ibid.).

[196]                      The Court of Appeal, after having found that s. 29 was constitutional, dealt briefly with the remedy granted at trial, simply stating that the declaration that the data had to be transferred to Quebec was [translation] “clearly inappropriate” and pointing out that “[t]he courts must not substitute their assessment of the appropriateness of a legislative measure for the intent of the legislature” (para. 64).

[197]                      In this Court, the AGQ argued that the federal government cannot refuse to transfer data that Quebec participated in compiling and enriching for provincial and federal purposes in the context of the partnership that existed between them. Such a refusal would be inconsistent with co-operative federalism. At the hearing, the AGQ instead stressed the principle of federalism. The AGQ added that a transfer of data would not be in breach of existing privacy legislation. The AGC responds that, regardless of “[t]he designation of a Chief Firearms Officer, the right to access the registration records, the financial Agreements as well as the notices concerning the protection of personal information”, Quebec has no right to the data, given that they are under the Registrar’s control: R.F., at para. 104.

[198]                      In our opinion, the AGQ has not established a legal basis for its claim to the data. The absence of a legal barrier to the transfer of the data does not necessarily mean that Quebec has proven that it is entitled to obtain them through the courts.

[199]                      Blanchard J.’s declaration that Quebec is entitled to receive the data essentially amounted to an injunction. Even though this Court has the power to make orders that are incidental to declarations of invalidity, it must refrain from intervening actively in legislative and governmental functions: see R. v. Prosper, [1994] 3 S.C.R. 236, at pp. 298-99. The Court has stated on several occasions that it is generally up to the legislatures to fill legislative gaps that are incompatible with the Constitution, and not up to the courts to supply an exact description of the laws the legislatures must adopt to fulfill their constitutional obligations: see Reference re Securities Act, at para. 132; Hunter v. Southam Inc., [1984] 2 S.C.R. 145, at p. 169; Edwards Books, at p. 783; Reference re Public Schools Act (Man.), s. 79(3), (4) and (7), [1993] 1 S.C.R. 839, at pp. 860-61. In some cases, the source of the appropriate remedy must lie in the political process rather than in the courts: Reference re Secession of Quebec, at para. 102.

[200]                      In the case at bar, if the Court were to make an order in the nature of an injunction or to declare that Quebec is entitled to receive the data, it would be overstepping the limits of its role in the Canadian constitutional order. Although it is true that the federal government cannot decide unilaterally to destroy the data accumulated within the framework of the partnership without considering the impact of such a measure on provincial heads of power, the other side of the coin is that Quebec cannot dictate to the federal government what it is “entitled” to receive when their relationship comes to an end. It was up to the members of the partnership to set out the conditions that were to apply upon termination of their joint venture in their agreements or, if they did not do so, it is up to them to agree now on the applicable terms in this regard, including the destruction of the data by the federal government. Thus, how their collaborative relationship is to be terminated is primarily dependent upon their will.

[201]                      However, these observations change nothing. Since s. 29 does not, in pith and substance, fall within a federal head of power and since it is not sufficiently integrated into the ELRA, it is invalid.

VI.    Conclusion

[202]                      We would allow this appeal in part and answer the constitutional question as follows:

                    Is s. 29 of the Ending the Long-gun Registry Act, S.C. 2012, c. 6 , ultra vires Parliament having regard to its criminal law power under s. 91(27)  of the Constitution Act, 1867 ?

Yes.

[203]                      The other orders being sought — an injunction or a safeguard order — are not justified, however, and should be rejected. In view of the qualified success on both sides, costs should be awarded to no one.

                    Appeal dismissed with costs throughout, LeBel, Abella, Wagner and Gascon JJ. dissenting.

                    Solicitors for the appellant: Bernard, Roy & Associés, Montréal.

                    Solicitor for the respondents: Attorney General of Canada, Montréal.

                    Solicitors for the intervener the Coalition for Gun Control: Juripop, Saint-Constant, Quebec; Deveau, Bourgeois, Gagné, Hébert & associés, Gatineau.

                    Solicitor for the intervener Canada’s National Firearms Association: Guy Lavergne, Saint-Lazare, Quebec.