Supreme Court of Canada

Appeal Heard: March 21, 2018 Judgement Rendered: January 11, 2019 Docket: 36645

Between:

Gillian Frank and Jamie Duong Appellants

and

Attorney General of Canada Respondent

— and —

Attorney General of Quebec, Canadian American Bar Association, Canadian Expat Association, David Asper Centre for Constitutional Rights, Canadian Civil Liberties Association, Chinese and Southeast Asian Legal Clinic and British Columbia Civil Liberties Association Interveners

Coram: Wagner C.J. and Moldaver, Karakatsanis, Gascon, Côté, Brown and Rowe JJ.

Reasons for Judgment: (paras. 1 to 83) Wagner C.J. (Moldaver, Karakatsanis and Gascon JJ. concurring)

Concurring Reasons: (paras. 84 to 110) Rowe J.

Joint Dissenting Reasons: (paras. 111 to 173) Côté and Brown JJ.

Frank v. Canada (Attorney General), 2019 SCC 1, [2019] 1 S.C.R. 3

Gillian Frank and Jamie Duong Appellants

Attorney General of Canada Respondent

and

Indexed as: Frank v. Canada (Attorney General)

2019 SCC 1

File No.: 36645.

2018: March 21; 2019: January 11.

Present: Wagner C.J. and Moldaver, Karakatsanis, Gascon, Côté, Brown and Rowe JJ.

on appeal from the court of appeal for ontario

Headnote

Constitutional law — Charter of Rights — Right to vote — Residence — Canada Elections Act denying right to vote in federal elections to Canadian citizens residing abroad for five consecutive years or more — Attorney General of Canada conceding infringement of right to vote — Whether infringement justified — Canadian Charter of Rights and Freedoms, ss. 1, 3 — Canada Elections Act, S.C. 2000, c. 9, ss. 11(d), 222.

The combined effect of ss. 11(d), 222 and other related provisions of the Canada Elections Act is to deny Canadian citizens who have resided abroad for five years or more the right to vote in a federal election unless and until they resume residence in Canada. The constitutionality of these provisions was challenged by two non-resident Canadian citizens, who applied for a declaration that their right to vote entrenched in s. 3 of the Charter was infringed, and that the impugned provisions were unconstitutional. The application judge agreed, found that the impugned provisions could not be saved under s. 1 of the Charter, and made an immediate declaration of invalidity. A majority of the Court of Appeal allowed the Attorney General of Canada's appeal. Although the Attorney General of Canada conceded that the impugned provisions breach s. 3 of the Charter, the violation of s. 3 was found to be justified.

Held (Côté and Brown JJ. dissenting): The appeal should be allowed. Sections 222(1)(b) and (c), 223(1)(f) and 226(f) of the Canada Elections Act are declared to be of no force or effect; the words "a person who has been absent from Canada for less than five consecutive years and who intends to return to Canada as a resident" are struck from s. 11(d) of the Act and are replaced with the words "an elector who resides outside Canada"; and the word "temporarily" is struck from ss. 220, 222(1) and 223(1)(e) of the Act.

Per Wagner C.J. and Moldaver, Karakatsanis and Gascon JJ.: As conceded by the Attorney General of Canada, the limit on the voting rights of long-term non-resident citizens breaches s. 3 of the Charter. This limit cannot be justified under s. 1. In particular, the Attorney General of Canada has failed to show that limiting the voting rights of non-resident citizens is minimally impairing.

Since voting is a fundamental political right, and the right to vote is a core tenet of Canadian democracy, any limit on the right to vote must be carefully scrutinized and cannot be tolerated without a compelling justification. Intrusions on this core democratic right are to be reviewed on a stringent justification standard. Reviewing courts must examine the proffered justification carefully and rigorously rather than adopting a deferential attitude. Two central criteria must be met for a limit on a Charter right to be justified under s. 1. First, the objective of the measure must be pressing and substantial. Second, the means by which the objective is furthered must be proportionate; this requires a rational connection to the objective, minimal impairment of the right, and proportionality between the effects of the measure and the objective. The proportionality inquiry is both normative and contextual, and requires that courts balance the interests of society with those of individuals and groups.

The integrity of the justification analysis requires that the legislative objective be properly stated. The relevant objective is that of the infringing measure, not, more broadly, that of the provision. In this case, the Attorney General of Canada has centrally and consistently asserted that the voting restrictions in question advance the objective of maintaining the fairness of the electoral system to resident Canadians. This is a sufficiently important legislative objective to ground the s. 1 analysis, and it can be a pressing and substantial concern even if the measures taken to achieve that objective impair the democratic rights of other citizens. This objective is sufficiently precise to continue the justification analysis.

The question at the first step of the proportionality inquiry is whether the measure that has been adopted is rationally connected to this objective. In cases where such a causal connection is not scientifically measurable, one can be made out on the basis of reason or logic. In this case, it must be shown that the infringement of non-residents' voting rights is rationally connected to the legislative objective of ensuring electoral fairness to resident voters. Here, there is no evidence of the harm that these voting restrictions are meant to address. No complaint has been identified with respect to voting by non-residents, and no evidence has been presented to show how voting by non-residents might compromise the fairness of the electoral system. Furthermore, it has not been definitively shown that a limit of any duration would be rationally connected to the electoral fairness objective. Overall, however, it is not necessary to come to a firm conclusion on this point in view of the result at the minimal impairment stage.

The second component of the proportionality test requires evidence that the measure at issue impairs the right as little as reasonably possible. In this case, the limit on voting by non-residents is not minimally impairing. There is little to justify the choice of five years as a threshold or to show how it is tailored to respond to a specific problem. As well, the five-year limit is overinclusive. It improperly applies to people to whom it is not intended to apply, and it does so in a manner that is far broader than necessary. While it seeks to bar people from voting who lack a sufficient connection to Canada, no correlation has been shown between, on the one hand, how long a Canadian citizen has lived abroad and, on the other hand, the extent of his or her subjective commitment to Canada. Many non-resident citizens maintain deep and abiding connections to Canada through family, online media and visits home, and by contributing taxes and collecting social benefits. Likewise, no correlation has been shown between residence and the extent to which citizens are affected by legislation. Non-resident citizens do live with the consequences of Canadian legislation: they are subject to Canadian legislation during visits home; Canadian laws affect the resident families of non-resident Canadians; some Canadian laws have extraterritorial application; government policies can have global consequences; and Parliament can alter the extent to which Canadian electoral legislation applies to non-resident citizens, which would make the constitutional right to vote subject to shifting policy choices.

At the final stage of the s. 1 analysis, it must be asked whether there is proportionality between the overall effects of the Charter-infringing measure and the legislative objective. In this case, any salutary effects of ensuring electoral fairness are clearly outweighed by the deleterious effects of disenfranchising non-resident Canadians who are abroad for five years or more. The benefits of the impugned legislation are illusory and speculative. It is unclear how the fairness of the electoral system is enhanced when long-term non-resident citizens are denied the right to vote. The deleterious effects on affected non-resident citizens, on the other hand, are serious. Denial of the right to vote, in and of itself, inflicts harm on affected citizens; proof of additional harm is not required. The disenfranchisement of long-term non-resident citizens not only denies them a fundamental democratic right, but also comes at the expense of their self-worth and their dignity.

Per Rowe J.: There is agreement that the appeal should be allowed. The limit on voting in federal elections for citizens who have not been resident in Canada for five years or more constitutes an unjustified infringement of s. 3 of the Charter. However, any evaluation of this kind of limit should acknowledge the significance and centrality of residence to Canada's system of representative democracy, and should not foreclose the possibility that residence requirements in another context might be constitutional.

Residence is significant because it establishes a connection to a particular electoral district and to the concerns of persons living there. While this aspect of Canada's representative democracy is not constitutionally entrenched, residence has been historically and remains today more than just an organizing mechanism. It is foundational to Canada's electoral system. However, its significance does not elevate residence to an inherent limit on the s. 3 right to vote.

Section 3 protects the right to vote, but it does not follow as a corollary that there is a right to vote in the constituency or province of one's choosing. The provinces and territories have each crafted residence requirements that reflect the concerns and circumstances that are particular to their jurisdiction. The concession by the Attorney General of Canada that the impugned residence requirements infringe s. 3 does not prejudice provincial or territorial governments in arguing that their legislation does not do so. Different considerations will apply at the s. 1 analysis of any established or conceded breach of s. 3 at the provincial or territorial level, and evidence of the circumstances relating to the various residence requirements in each of the provinces and territories may well affect the analysis.

In this case, promoting electoral fairness for resident Canadians is a pressing and substantial objective, and the impugned measures are rationally connected to this objective. If the law's legitimacy derives from the fact that those who are subject to it are the ones who indirectly create it, then it is unfair that individuals who are not subject to or affected by the law can decide for those who are. Long-term non-residents are likely to be less connected to any Canadian community. Individuals who have not lived in a constituency for over five years are less likely to be informed about the issues affecting that constituency, and long-term non-resident voters who have no intention to return will not feel the impacts of federal laws and policies as they manifest themselves at the local level. Similarly, at the national level, those who have not lived in Canada for a long time are likely to be less connected to Canada and are affected by Canadian law to a far lesser degree than are resident Canadians.

However, the impugned measures ultimately do not withstand s. 1 scrutiny, as the salutary effects of promoting fairness for resident Canadians are outweighed by the deleterious effects of denying long-term non-resident Canadians the right to vote in federal elections. The primary salutary effect of the impugned measures is that long-term non-residents will not cast a decisive vote in a constituency in which they are not resident, and therefore, the local representative will not be selected (in part) by individuals from outside the constituency. However, this salutary effect has not been shown to be consequential: there is almost no evidence of the impact that long-term non-residents would or could have had either locally or nationally if permitted to vote, and the evidence that exists suggests that the impact would likely be negligible, since a very small number of Canadians living abroad who are currently eligible to vote choose to exercise that right. By contrast, the deleterious effects of the provisions on long-term non-residents are clear: they cannot vote. While they may not feel the local consequences of particular federal policies in the constituencies in which their votes would be counted, they stand nonetheless to be affected by certain federal laws and policies. Furthermore, the right to vote is not merely instrumental. Denying long-term non-residents the right to vote denies them the opportunity to participate, through their vote, in the formation of policy and the functioning of public institutions.

Per Côté and Brown JJ. (dissenting): The appeal should be dismissed. The non-resident voting restriction represents a reasonable limit on the right to vote under s. 3 of the Charter.

The fundamental point of disagreement with the majority lies in the proper judicial approach to the limitations analysis under s. 1 of the Charter. No right is absolute, including Charter rights such as the s. 3 right to vote. However, to speak of an "infringement" based solely on the fact that an impugned measure imposes a limit on a Charter right distorts the s. 1 analysis. The Charter cannot and should not be read so as to ever allow for justified infringements. The text of s. 1 speaks not of reasonable and demonstrably justifiable infringements, but of reasonable and demonstrably justifiable limits. A reasonable limit is inherent in the right itself, shaping the right's outer boundaries. A right is infringed only where the right, as reasonably limited, is breached; as such, an "infringement" is a limit that is not justified. A conceptually sensible and textually faithful account of the s. 1 analysis thus properly focuses on whether a limit on a Charter right is justified. The issue presented by this appeal, then, is not whether the limit to the right to vote effected by the restriction on long-term non-resident voting justifies an infringement of s. 3, but whether that limit is unreasonable, such that s. 3 is infringed.

The first step in the limitations analysis is to identify the objective behind the impugned measure and determine whether it is sufficiently important to justify the limit on a Charter right. Identifying the objective of a rights-limiting measure enacted by a legislature raises the methodological difficulty that the objective may not be immediately apparent. The difficulty is augmented where the impugned limitation arises by the absence of legislative action. Context — being both the present and past state of the law — is essential to the proper characterization of the objective. However, there should be some circumspection in relying upon parliamentary debates to identify the legislative objective of a provision, as the intent of particular members of Parliament is not the same as the intent of Parliament as a whole.

Here, there is a clear and readily discernable purpose behind the legislative design as a whole: Parliament sought to privilege a relationship of some currency between electors and their communities. This objective is sufficiently pressing and substantial to survive scrutiny under s. 1. Parliament was quite properly striving to shape the boundaries of the right by enacting legislation governing the terms on which elections are conducted, by drawing a line at citizens who have a current relationship to the community in which they seek to cast a ballot. Parliament is permitted, within limits, to shape the scope of voting rights under s. 1. While most Charter rights are negative in the sense that they preclude the state from acting in ways that would impair them, the right to vote is a positive entitlement. It requires legislative specification in order for the right to be operative. Limits to the right to vote can be justified, because some specification of the right to vote (whether to account for age, or the currency of relationship between electors and the communities they represent) is necessary. As well, the Act contains a range of restrictions on voting — including withholding the vote from Canadian citizens who have never before lived in Canada and Canadian citizens under the age of 18 — which are no less the product of principled and unavoidably philosophical reasoning than is the long-term non-resident restriction. The limitations analysis must be flexible enough to account for Parliament's ability to legislate in pursuit of philosophical, moral or otherwise normative considerations. The appropriate inquiry in limitations analysis is, therefore, not whether Parliament's legislative objective rests on such considerations, but whether the objective that it pursues is pressing and substantial.

In this case, the restriction at issue is a residence requirement. Residence has been described as a fundamental requirement of the right to vote. While citizenship is a necessary requirement to vote, it is therefore not the only constitutionally permissible limit. Citizenship is a status. It does not itself indicate a relationship of any currency to a particular Canadian community. Parliament, not unreasonably, deemed residence or recent residence to be indicative of this relationship. The fact that the Act includes certain exceptions to the residence rule supports the notion that a relationship of currency is essential. Preserving a relationship of currency between electors and their communities by limiting long-term non-resident voting ensures reciprocity between exercising the right to vote and bearing the burden of Canadian laws. The reciprocity principle justifies limiting non-resident voting precisely because long-term non-residents are not generally subject to Canadian laws. It is unfair to Canadian residents for their lawmakers to be elected by long-term non-residents who have no connection of any currency to their electoral district. Preserving the relationship between electors and their communities through limits on long-term non-resident voting also protects the integrity of the Canadian electoral system, which is founded on geographical representation. The s. 3 voting right is premised upon electors voting for a representative of their community. This regional structure must therefore inform any consideration of the electoral system, and Canadians' participation therein. Limiting long-term non-resident voting ensures that the electors residing in a particular constituency, who share a community of interests that is typically derived at least in part from geographical proximity, retain the power to decide for themselves who would best advance those shared interests on their behalf in the House of Commons.

The second question in the limitations analysis asks whether the means that Parliament selected to pursue its objective are proportionate to the rights limitation — i.e., whether the measure is rationally connected to the objective; whether the measure minimally impairs a claimant's Charter rights; and whether there is proportionality between the effects of the measure and the objective. The inquiry is not to ask what the Court prefers, but whether the limit was one that Parliament could reasonably impose. This is particularly so in the case of a challenge to Canada's election laws, in respect of which the Court has previously held that a natural attitude of deference is required. In this case, the means by which Parliament chose to pursue its objective are proportionate. A five-year non-resident cut-off was a reasonable and constitutionally permissible demarcation. The measure is rationally connected to the objective of preserving a relationship of currency between electors and their communities because it logically distinguishes short-term from long-term non-residents. Five years corresponds to the maximum length of a Parliament, thereby ensuring that all non-residents can vote in at least one election after leaving Canada, and it is sufficiently long to permit students who travel abroad to study to complete their programs without foregoing the ability to vote. The limit is also minimally impairing because, on balance, a five-year time period falls within the range of reasonable options that were open to Parliament and within the range of limits adopted by other internationally respected liberal democracies sharing Canada's parliamentary framework. Opening the vote to long-term non-residents would not be an example of progressive enfranchisement; rather, it would be a regressive development, undermining the longstanding and entirely salutary practice in Westminster parliamentary democracies of privileging local connections in deciding who may elect local representatives.

In the final balancing, the salutary effects of preserving the integrity of Canada's geographically based electoral system and upholding a democratically enacted conception of the scope of the right to vote in Canada are significant. The deleterious effect of denying some citizens the right to vote is not insubstantial, but it is tempered by the fact that the restriction is reversible rather than permanent, as any adult Canadian citizen can still exercise the right to vote at any point, provided that he or she re-establishes residence in Canada. Thus, the restriction at issue is not a permanent denial of the right to vote. Just like the age requirement, it represents a distinction based on the experiential situation of all citizens in that category; it is not a distinction based on moral worth. The deleterious effects of the limit are therefore of less significance, and are outweighed by the salutary effects.

Cases Cited

Cited by Wagner C.J.

Cited by Rowe J.

Distinguished: Harper v. Canada (Attorney General), 2004 SCC 33, [2004] 1 S.C.R. 827; considered: Reference re Yukon Election Residency Requirements (1986), 1986 CanLII 3944 (YK CA), 1 Y.R. 23; referred to: Haig v. Canada (Chief Electoral Officer), 1993 CanLII 58 (SCC), [1993] 2 S.C.R. 995; RJR-MacDonald Inc. v. Canada (Attorney General), 1995 CanLII 64 (SCC), [1995] 3 S.C.R. 199; Lavigne v. Ontario Public Service Employees Union, 1991 CanLII 68 (SCC), [1991] 2 S.C.R. 211; Sauvé v. Canada (Chief Electoral Officer), 2002 SCC 68, [2002] 3 S.C.R. 519; Figueroa v. Canada (Attorney General), 2003 SCC 37, [2003] 1 S.C.R. 912.

Cited by Côté and Brown JJ. (dissenting)

Sauvé v. Canada (Chief Electoral Officer), 2002 SCC 68, [2002] 3 S.C.R. 519; R. v. Oakes, 1986 CanLII 46 (SCC), [1986] 1 S.C.R. 103; R. v. K.R.J., 2016 SCC 31, [2016] 1 S.C.R. 906; Law Society of British Columbia v. Trinity Western University, 2018 SCC 32, [2018] 2 S.C.R. 453; Bracken v. Niagara Parks Police, 2018 ONCA 261, 141 O.R. (3d) 168; R. v. Keegstra, 1990 CanLII 24 (SCC), [1990] 3 S.C.R. 697; R. v. Sharpe, 2001 SCC 2, [2001] 1 S.C.R. 45; Thomson Newspapers Co. v. Canada (Attorney General), 1998 CanLII 829 (SCC), [1998] 1 S.C.R. 877; R. v. Heywood, 1994 CanLII 34 (SCC), [1994] 3 S.C.R. 761; Ruparel v. Canada (Minister of Employment and Immigration), 1990 CanLII 13102 (FC), [1990] 3 F.C. 615; Re B.C. Motor Vehicle Act, 1985 CanLII 81 (SCC), [1985] 2 S.C.R. 486; Reference re Upper Churchill Water Rights Reversion Act, 1984 CanLII 17 (SCC), [1984] 1 S.C.R. 297; Rizzo & Rizzo Shoes Ltd. (Re), 1998 CanLII 837 (SCC), [1998] 1 S.C.R. 27; Placer Dome Canada Ltd. v. Ontario (Minister of Finance), 2006 SCC 20, [2006] 1 S.C.R. 715; Canadian National Railway Co. v. Canada (Attorney General), 2014 SCC 40, [2014] 2 S.C.R. 135; R. v. Morgentaler, 1993 CanLII 74 (SCC), [1993] 3 S.C.R. 463; R. v. Butler, 1992 CanLII 124 (SCC), [1992] 1 S.C.R. 452; Harper v. Canada (Attorney General), 2004 SCC 33, [2004] 1 S.C.R. 827; Reference re Prov. Electoral Boundaries (Sask.), 1991 CanLII 61 (SCC), [1991] 2 S.C.R. 158; Opitz v. Wrzesnewskyj, 2012 SCC 55, [2012] 3 S.C.R. 76; Haig v. Canada (Chief Electoral Officer), 1993 CanLII 58 (SCC), [1993] 2 S.C.R. 995; R. v. Edwards Books and Art Ltd., 1986 CanLII 12 (SCC), [1986] 2 S.C.R. 713; Stoffman v. Vancouver General Hospital, 1990 CanLII 62 (SCC), [1990] 3 S.C.R. 483; Irwin Toy Ltd. v. Quebec (Attorney General), 1989 CanLII 87 (SCC), [1989] 1 S.C.R. 927; R. v. Bryan, 2007 SCC 12, [2007] 1 S.C.R. 527.

Statutes and Regulations Cited

Canada Elections Act, R.S.C. 1985, c. E-2, s. 51(e).

Canada Elections Act, S.C. 2000, c. 9, ss. 3, 6, 8, 11(d), 127, 191(d), 220, 222, 223(1)(e), (f), 226(f), Part 11.

Canadian Charter of Rights and Freedoms, ss. 1, 3, 33.

Citizenship Act, R.S.C. 1985, c. C-29, s. 3.

Constitution Act, 1867, preamble, ss. 21, 22, 37 to 40, 52, 69 to 72.

Constitution Act, 1982, s. 52.

Dominion Elections Act, S.C. 1920, c. 46.

Election Act, R.S.O. 1990, c. E.6, s. 15(1.1).

Election Act, R.S.Q., c. E-3.3, s. 1.

Electoral Boundaries Readjustment Act, R.S.C. 1985, c. E-3.

Nunavut Elections Act, S. Nu. 2002, c. 17, ss. 4(12), 7(1).

Patent Act, R.S.C. 1985, c. P-4.

Authors Cited

Ajzenstat, Janet. The Canadian Founding: John Locke and Parliament. Montréal: McGill-Queen's University Press, 2007.

Atkinson, Neill. "Voting rights", in Te Ara — The Encyclopedia of New Zealand, 2015 (online: https://teara.govt.nz/en/voting-rights/print; archived version: https://www.scc-csc.ca/cso-dce/2019SCC-CSC1_1_eng.pdf).

Baker, John H. "Case and Comment: Statutory Interpretation and Parliamentary Intention" (1993), 52 Cambridge L.J. 353.

Blais, André, et al. "Deciding Who Has the Right to Vote: A Comparative Analysis of Election Laws" (2001), Electoral Studies 20.

Bogdanor, Vernon. "Introduction", in V. Bogdanor and David Butler, eds., Democracy and Elections: Electoral Systems and Their Political Consequences. Cambridge: Cambridge University Press, 1983.

Boyer, J. Patrick. Election Law in Canada: The Law and Procedure of Federal, Provincial and Territorial Elections, vol. 1. Toronto: Butterworths, 1987.

Canada. Department of Justice. Principles guiding the Attorney General of Canada in Charter litigation, 2017 (online: https://www.justice.gc.ca/eng/csj-sjc/principles2-eng.pdf; archived version: https://www.scc-csc.ca/cso-dce/2019SCC-CSC1_2_eng.pdf).

Canada. House of Commons. Minutes of Proceedings and Evidence of the Special Committee on Electoral Reform, No. 3, 3rd Sess., 34th Parl., March 25, 1992.

Canada. House of Commons. Minutes of Proceedings and Evidence of the Special Committee on Electoral Reform, No. 5, 3rd Sess., 34th Parl., November 24, 1992.

Canada. House of Commons. Minutes of Proceedings and Evidence of the Standing Committee on Privileges and Elections, No. 1, 2nd Sess., 32nd Parl., March 27, 1984.

Canada. House of Commons. Special Committee on Electoral Reform. Third report to the House of Commons, No. 7, 3rd Sess., 34th Parl., December 11, 1992.

Canada. Privy Council. White Paper on Election Law Reform. Ottawa, 1986.

Canada. Royal Commission on Electoral Reform and Party Financing. Reforming Electoral Democracy: Final Report, vol. 1. Ottawa: The Commission, 1991.

Dodek, Adam. The Canadian Constitution, 2nd ed. Toronto: Dundurn, 2016.

Ekins, Richard. "Legislation as Reasoned Action", in G. Webber et al., Legislated Rights: Securing Human Rights through Legislation. Cambridge: Cambridge University Press, 2018.

Ekins, Richard. The Nature of Legislative Intent. Oxford: Oxford University Press, 2012.

Eskridge, William N., Jr. Dynamic Statutory Interpretation. Cambridge, Mass.: Harvard University Press, 1994.

Hogg, Peter W. Constitutional Law of Canada, vol. 2, 5th ed. Scarborough, Ont.: Thomson/Carswell, 2017.

MacGregor Dawson, Robert. The Government of Canada, 4th ed. Toronto: University of Toronto Press, 1963.

Miller, Bradley W. "Justification and Rights Limitations", in G. Huscroft, ed. Expounding the Constitution: Essays in Constitutional Theory. Cambridge: Cambridge University Press, 2008.

Pound, Roscoe. "Mechanical Jurisprudence" (1908), 8 Colum. L. Rev. 605.

Scalia, Antonin and Bryan A. Garner. Reading Law: The Interpretation of Legal Texts. St. Paul, Minn.: Thomson/West, 2012.

Sikkema, John. "The 'Basic Bedford Rule' and Substantive Review of Criminal Law Prohibitions Under Section 7 of the Charter", in D. B. M. Ross, ed., Assisted Death: Legal, Social and Ethical Issues after Carter. Toronto: LexisNexis, 2018.

Sullivan, Ruth. Sullivan on the Construction of Statutes, 6th ed. Markham, Ont.: LexisNexis, 2014.

Sunstein, Cass R. "Interpreting Statutes in the Regulatory State" (1989), 103 Harv. L. Rev. 405.

Waldron, Jeremy. Law and Disagreement. New York: Oxford University Press, 2001.

Webber, Grégoire C. N. The Negotiable Constitution: On the Limitation of Rights. New York: Cambridge University Press, 2009.

APPEAL from a judgment of the Ontario Court of Appeal (Strathy C.J. and Laskin and Brown JJ.A.), 2015 ONCA 536, 126 O.R. (3d) 321, 338 O.A.C. 218, 388 D.L.R. (4th) 1, 340 C.R.R. (2d) 323, [2015] O.J. No. 3820 (QL), 2015 CarswellOnt 10870 (WL Can.), setting aside a decision of Penny J., 2014 ONSC 907, 119 O.R. (3d) 662, 372 D.L.R. (4th) 681, 310 C.R.R. (2d) 17, [2014] O.J. No. 2098 (QL), 2014 CarswellOnt 5850 (WL Can.). Appeal allowed, Côté and Brown JJ. dissenting.

Shaun O'Brien and Amanda Darrach, for the appellants.

Sean Gaudet and Gail Sinclair, for the respondent.

Jean-Vincent Lacroix and Dominique A. Jobin, for the intervener the Attorney General of Quebec.

Bradley E. Berg and Max Shapiro, for the intervener the Canadian American Bar Association.

Sean Sutherland and Colin Feasby, for the intervener the Canadian Expat Association.

Louis Century and Cheryl Milne, for the intervener the David Asper Centre for Constitutional Rights.

Mark J. Freiman and Jameel Madhany, for the intervener the Canadian Civil Liberties Association.

Avvy Yao-Yao Go, for the intervener the Chinese and Southeast Asian Legal Clinic.

Brendan van Niejenhuis, Michael Sobkin and Stephen Aylward, for the intervener the British Columbia Civil Liberties Association.

The judgment of Wagner C.J. and Moldaver, Karakatsanis and Gascon JJ. was delivered by

The Chief Justice —

I. Overview

[ 1 ] Voting is a fundamental political right, and the right to vote is a core tenet of our democracy. Its primacy is entrenched in s. 3 of the Canadian Charter of Rights and Freedoms, which states: "Every citizen of Canada has the right to vote in an election of members of the House of Commons or of a legislative assembly and to be qualified for membership therein." Any limit on the right to vote must be carefully scrutinized and cannot be tolerated without a compelling justification.

[ 2 ] Canada's history has been one of progressive enfranchisement. The right to vote in federal elections was originally restricted to property-owning men aged 21 or older, but the franchise has gradually been extended to include almost all citizens aged 18 or older. Women, racial minorities, individuals formerly described as having a "mental disease", penitentiary inmates, and Canadian residents living abroad in service of Canada's armed forces and public administration were once excluded but now have the right to vote.

[ 3 ] This case calls into question one of the last restrictions on the right to vote in federal elections: residence. Canadian citizens who reside abroad for less than five consecutive years and who intend to return to Canada maintain their right to vote in Canadian federal elections. But once non-resident citizens have lived abroad beyond that five-year period, they are disenfranchised.

[ 4 ] The respondent, the Attorney General of Canada ("AGC"), concedes that the limit on the voting rights of non-residents breaches s. 3 of the Charter. It follows that the central question in this appeal is whether this is a reasonable limit that can be demonstrably justified under s. 1. I conclude that it cannot. The vague and unsubstantiated electoral fairness objective that is purportedly served by denying voting rights to non-resident citizens simply because they have crossed an arbitrary five-year threshold does not withstand scrutiny.

[ 5 ] In particular, the AGC has failed to show that limiting the voting rights of non-resident citizens is minimally impairing. There is little to justify the choice of five years as a threshold or to show how it is tailored to respond to a specific problem. It is also clear that the measure improperly applies to many individuals with deep and abiding connections to Canada and to Canadian laws, and that it does so in a manner that is far broader than necessary to achieve the electoral fairness objective advanced by the AGC. The disenfranchisement of these citizens not only denies them a fundamental democratic right, but also comes at the expense of their sense of self-worth and their dignity. These deleterious effects far outweigh any speculative benefits that the measure might bring about.

[ 6 ] Accordingly, I would allow the appeal. The limit on the voting rights of non-residents violates s. 3 of the Charter and is not saved by s. 1.

II. Background

[ 7 ] This appeal stems from a challenge by two Canadian citizens, Dr. Gillian Frank and Jamie Duong, the appellants in this Court, of the denial of their right to vote in a Canadian federal election on the basis that they have lived abroad for longer than five years.

[ 8 ] Dr. Frank resides in the United States. He lived in Toronto until the age of 21. After having obtained a university degree in Canada, he was accepted on a full scholarship for graduate studies at Brown University in Providence, Rhode Island. He is currently completing post-doctoral studies and living in Princeton, New Jersey. Dr. Frank's wife is also a Canadian citizen, and the members of his immediate family live in Toronto. He travels on a Canadian passport and is not entitled to vote in the United States. Dr. Frank has applied, as yet unsuccessfully, for work in Canada. He states that he would return to Canada without hesitation if he were to find a suitable academic position in this country.

[ 9 ] Mr. Duong was born in Montréal. He attended Cornell University in Ithaca, New York, where as a student he worked part time on campus at a job that he subsequently converted into a permanent position. He currently lives in Fairfax, Virginia. The members of his immediate family live in Montréal, and he visits Canada regularly. He also expects partial ownership of several family-owned properties in Canada to be transferred to him in the near future. Mr. Duong has applied, without success, for a suitable position in Canada. Like Dr. Frank, Mr. Duong expresses a desire to return to Canada should he find appropriate employment.

[ 10 ] Dr. Frank and Mr. Duong both tried to vote in the Canadian federal election in May 2011. Both were notified that they were not entitled to receive a ballot, because they had been residing outside Canada for more than five years. In response, they challenged the provisions of the Canada Elections Act, S.C. 2000, c. 9 ("Act"), that deny them the right to vote — i.e., ss. 11(d), 222(1)(b) and (c), 223(1)(f), 226(f) and the word "temporarily" in ss. 220, 222(1) and 223(1)(e). Specifically, the appellants argue that the requirements of residing outside Canada for less than five consecutive years and of intending to resume residence in Canada in the future unjustifiably violate their rights under s. 3 of the Charter.

[ 11 ] The Act is a comprehensive statute which regulates federal elections in Canada. Its central purposes are to enfranchise all persons who are entitled to vote, and to protect the integrity of the democratic process (Opitz v. Wrzesnewskyj, 2012 SCC 55, [2012] 3 S.C.R. 76, at paras. 35 and 38). In contrast to s. 3 of the Charter, which enshrines the constitutional right of every Canadian citizen to vote, the Act establishes specific rules with respect to qualification as an elector and to voting entitlements, as well as voting mechanisms, in pursuit of its broad enfranchising purpose.

[ 12 ] Under the Act, everyone who is a Canadian citizen and is 18 or older is qualified as an elector (s. 3). A person who is qualified as an elector is entitled to vote at the polling station for the polling division in which he or she is ordinarily resident (s. 6). In other words, a qualified elector is entitled to vote at the place in Canada where he or she ordinarily resides. This is known as the "residence requirement".

[ 13 ] Despite the residence requirement, there is an existing legislative scheme which allows certain non-resident citizens to vote from abroad by means of a "special ballot" (s. 127). The special ballot procedure is the only mechanism by which citizens who do not reside in Canada can vote.

[ 14 ] Eligibility for voting by way of special ballot is set out in s. 11 of the Act, which reads:

Any of the following persons may vote in accordance with Part 11:

(a) a Canadian Forces elector;

(b) an elector who is an employee in the federal public administration or the public service of a province and who is posted outside Canada;

(c) a Canadian citizen who is employed by an international organization of which Canada is a member and to which Canada contributes and who is posted outside Canada;

(d) a person who has been absent from Canada for less than five consecutive years and who intends to return to Canada as a resident;

(e) an incarcerated elector within the meaning of that Part; and

(f) any other elector in Canada who wishes to vote in accordance with that Part.

[ 15 ] This appeal concerns the limit on non-residents' voting rights that is established in s. 11(d) of the Act. This limit is also mentioned in the "Special Voting Rules" set out in Part 11 of the Act, which govern the special ballot procedure. Part 11 requires that the Chief Electoral Officer maintain a register of electors who are temporarily resident outside Canada and who have applied for a special ballot. To be included in the register, these non-resident electors must have previously resided in Canada, must have been residing outside Canada for less than five consecutive years immediately before making the application, and must intend to resume residence in Canada in the future (s. 222(1)).

[ 16 ] The Act establishes exceptions to the five-year rule for members of the Canadian Forces, including individuals employed as teachers in or support staff for Canadian Forces schools, who are posted outside Canada; employees in the federal public service or that of a province who are posted outside Canada; employees of an international organization of which Canada is a member who are posted outside Canada; and any elector who lives with a Canadian citizen in one of these exempted groups (ss. 11, 191(d) and 222(2)). Any other citizen who has resided outside Canada for five consecutive years or more will have his or her name deleted from the register (s. 226(f)).

[ 17 ] The combined effect of ss. 11(d) and 222 of the Act is that — subject to the exceptions listed in s. 222(2) — a Canadian citizen who has resided abroad for five years or more is not entitled to vote in a federal election unless and until he or she resumes residence in Canada.

III. Judicial History

A. Ontario Superior Court of Justice, [2014 ONSC 907](https://www.canlii.org/en/on/onsc/doc/2014/2014onsc907/2014onsc907.html), 119 O.R. (3d) 662

[ 18 ] The appellants applied to the Ontario Superior Court of Justice for a declaration that the provisions of the Act which deny them their right to vote are unconstitutional and of no force or effect. Penny J. found that the impugned provisions infringe s. 3 of the Charter and cannot be saved under s. 1. The government had asserted that the objectives of the limit were generally to ensure fairness to resident voters and to maintain the proper functioning and the integrity of Canada's electoral system. Penny J. expressed concern over the rhetorical nature of those objectives, which were not substantiated by concrete evidence. Out of prudence, however, he declined to dismiss them outright and proceeded to the proportionality inquiry from R. v. Oakes, 1986 CanLII 46 (SCC), [1986] 1 S.C.R. 103. He found that the stated objectives failed at each step of the analysis, and made an immediate declaration of invalidity.

B. Court of Appeal for Ontario (Strathy C.J.O., Brown J.A. Concurring; Laskin J.A. Dissenting), [2015 ONCA 536](https://www.minicounsel.ca/oca/2015/536), 126 O.R. (3d) 321

[ 19 ] A majority of the Ontario Court of Appeal allowed the appeal. Because the AGC was now conceding that the impugned legislation breaches s. 3 of the Charter, the appeal turned entirely on whether the violation can be justified under s. 1.

[ 20 ] In the Court of Appeal, the AGC reframed the legislative objectives significantly, arguing that "the residency requirement fulfills the pressing and substantial objective of preserving the social contract at the heart of Canada's system of constitutional democracy" (para. 51). The AGC explained that the social contract ensures that citizens are subjectively connected to Canada through their knowledge and affiliation, and also objectively connected through citizenship responsibilities and the duty to obey domestic laws. Strathy C.J.O., writing for the majority, relied heavily on this Court's decision in Sauvé v. Canada (Chief Electoral Officer), 2002 SCC 68, [2002] 3 S.C.R. 519 ("Sauvé #2"), to find strong support for the social contract both in political theory and in this Court's jurisprudence, and to conclude that preserving it was a valid objective for the purposes of the s. 1 analysis.

[ 21 ] Turning to the proportionality inquiry of the Oakes test, Strathy C.J.O. found that the impugned provisions are proportionate to the objective. In his view, excluding non-resident Canadians from the franchise strengthens the social contract and enhances the legitimacy of laws; he therefore found that the impugned provisions are rationally connected to the social contract objective. He determined that the five-year limit falls within the range of reasonable policy choices that were available to Parliament and is therefore minimally impairing. Finally, he concluded that there is proportionality between the salutary and deleterious effects of the law, in large part because non-resident citizens can move back to Canada at any time and thus regain their right to vote. As the limit in question satisfies all three steps of the proportionality analysis from Oakes, the majority of the Court of Appeal concluded that the violation of s. 3 of the Charter is justified under s. 1.

[ 22 ] Laskin J.A., dissenting, would have dismissed the appeal. He took issue with the AGC's "social contract" objective, which he considered to amount to a new argument that had been improperly raised on appeal. Further, he was concerned that preserving the social contract did not in fact correspond to Parliament's intent at the time the law was enacted. Finally, even if that objective were assumed to have been validly raised, Laskin J.A. was of the view that it would not be a pressing and substantial objective and would fail to satisfy the proportionality requirements of the Oakes test.

IV. Issue

[ 23 ] This appeal raises the issue of the constitutionality of the provisions of the Act that limit the right to vote of non-residents. The AGC originally took the position that the residence requirement constitutes an internal limit on the right to vote. On that basis, she claimed there was no breach of s. 3 of the Charter. However, the AGC has now conceded that the impugned provisions do breach s. 3 of the Charter. The central question in the appeal is therefore whether this breach can be justified under s. 1.

V. Analysis

A. Right to Vote

[ 24 ] Since context is the key to understanding the scope of a limit on a Charter right, I will begin by discussing the nature of the right to vote and the role of residence in our electoral system in order to lay a proper foundation for the justification analysis (Toronto Star Newspapers Ltd. v. Canada, 2010 SCC 21, [2010] 1 S.C.R. 721, at para. 3).

[ 25 ] The right of every citizen to vote lies at the heart of Canadian democracy (Sauvé #2, at para. 1; Opitz, at para. 10). In Sauvé #2, a seminal decision on the right to vote, this Court reviewed the nature and purpose of s. 3 at length before striking down legislation which prevented inmates serving sentences of two years or more from voting in federal elections. McLachlin C.J., writing for the majority, stressed the critical importance of a broad and purposive interpretation of the right to vote. She stated that the framers of the Charter had "signaled the special importance of this right not only by its broad, untrammeled language, but by exempting it from legislative override under s. 33's notwithstanding clause" (para. 11). As a result, any intrusions on this core democratic right are to be reviewed on the basis of a stringent justification standard (para. 14).

[ 26 ] The central purpose of s. 3 is to ensure the right of each citizen to participate meaningfully in the electoral process (Figueroa v. Canada (Attorney General), 2003 SCC 37, [2003] 1 S.C.R. 912, at paras. 25-26). Civic participation is fundamentally important to the health of a free and democratic society. Democracy demands that each citizen have a genuine opportunity to participate in the governance of the country through the electoral process. If this right were not protected adequately, ours would not be a true democracy (Figueroa, at para. 30).

[ 27 ] Therefore, a broad interpretation of s. 3 enhances the quality of our democracy and strengthens the values on which our free and democratic state is premised (Figueroa, at para. 27). As a corollary, an overly narrow interpretation of the right to vote would diminish the quality of democracy in our system of government. As this Court observed in Sauvé #2, a government that restricted the franchise to a select group would effectively weaken the legitimacy of the country's democratic system and undermine its own claim to power (para. 34).

B. Role of Residence in the Canadian Electoral System

[ 28 ] The underlying theme of this appeal is the role of residence in Canada's electoral system. Although the AGC has conceded that the impugned provisions breach s. 3 of the Charter, her reliance on the social contract theory nevertheless raises important questions relating to the function of residence in the electoral system. In my view, residence can best be understood as an organizing mechanism for purposes of the right to vote. It is an important device which underpins our geographically determined, constituency-based system of electoral representation. However, this instrumental necessity does not elevate residence to an essential requirement of the Charter right to vote.

[ 29 ] In clear language, the Charter tethers voting rights to citizenship, and citizenship alone. Section 3 does not mention residence. Citizenship is the defining requirement of the right to vote, and the choice of the framers of the Charter to omit the residence requirement as an element of this core democratic right is significant.

[ 30 ] Further, the Act provides for exemptions from the residence requirement for citizens working abroad in Canada's public service or in or for the Canadian Forces, for citizens employed abroad by international organizations which have a specified connection to Canada, and for citizens living with members of any of these groups. These existing exemptions contradict the view that residence is essential to the right to vote.

[ 31 ] It follows that a broad and purposive interpretation of s. 3 does not allow for residence to operate as an internal limit on the right to vote. As this Court held in Sauvé #2, the ambit of the s. 3 right to vote "should not be limited by countervailing collective concerns . . . These concerns are for the government to raise under s. 1 in justifying the limits it has imposed on the right" (para. 11). This does not of course foreclose the possibility that an infringement of the right to vote could be justified in the context of a reasonable limits analysis or that legislative restrictions on the right to vote are necessarily unconstitutional. What it does mean, however, is that limits on the right to vote must, if raised, be justified under s. 1, and not indirectly incorporated into the scope of the right itself. Accordingly, residence is not an essential requirement of the right to vote; rather, it is a countervailing consideration which must be justified by the AGC under s. 1 of the Charter.

[ 32 ] Nor, in my view, does the jurisprudence establish that residence is essential to the Charter right to vote. In no case has this Court held that residence is an essential and implicit requirement of the right to vote. On the contrary, as I mentioned above, it is clear from the jurisprudence that a broad and liberal interpretation is particularly critical in the case of the right to vote and that s. 3 must be "construed as it reads" (Sauvé #2, at para. 11). And although this Court described residence as a "fundamental requirement" in Opitz, it did so in relation to the definition of the statutory entitlement to vote set out in s. 6 of the Act, not to the scope of the right to vote under s. 3 of the Charter (para. 32). Moreover, the issue in Opitz was whether an election should be set aside because of administrative errors, not the scope of the right to vote in the context of a Charter challenge to electoral legislation. I therefore do not take Opitz to stand for the proposition that residence is an essential requirement of the right to vote under the Charter.

[ 33 ] In any event, the role of residence in our electoral system must be understood in its historical context. The requirement emerged at a time when citizens were generally unable to travel as easily and extensively as they do today and tended to spend their lives in one community. At that time, the right to vote was linked to the ownership of land, and only male property owners could vote. The residence requirement was designed, in part, to prevent "plural voting", that is, to prevent a person who owned property in several ridings from casting a vote in each of them (Haig v. Canada (Chief Electoral Officer), 1993 CanLII 58 (SCC), [1993] 2 S.C.R. 995, at p. 1052).

[ 34 ] Today, in contrast, we live in a globalized society. The ability of citizens not only to move, but to remain connected and maintain communications in so doing, is unprecedented. Many Canadians live abroad, and many do so for five years or more. The application judge cited evidence showing that in 2009, approximately 2.8 million Canadians — or 8 percent of Canada's population at the time — had been living abroad for one year or more, and that there were well over one million Canadians to whom the non-residence limit in the Act applied. He also noted that the results of one research project show that non-resident Canadian citizens maintain strong connections, both family- and employment-related, to Canada, as well as a strong sense of belonging. According to Penny J., the evidence revealed that, in addition to socio-cultural connections, many non-residents maintain strong economic ties to Canada by contributing to social insurance programs, paying taxes and receiving benefits. Further, he noted that 60 percent of surveyed respondents were solely Canadian citizens and, if denied the right to vote in Canada, would be unable to vote in any other country (see application judge's reasons, at paras. 19-30).

[ 35 ] In sum, the world has changed. Canadians are both able and encouraged to live abroad, but they maintain close connections with Canada in doing so. The right to vote is no longer tied to the ownership of property and bestowed only on select members of society. And citizenship, not residence, defines our political community and underpins the right to vote.

C. Justification Analysis

[ 36 ] Section 1 of the Charter provides:

The Canadian Charter of Rights and Freedoms guarantees the rights and freedoms set out in it subject only to such reasonable limits prescribed by law as can be demonstrably justified in a free and democratic society.

[ 37 ] The impugned provisions of the Act are clearly prescribed by law (Oakes, at p. 135). The question remains whether the limit on non-residents' voting rights can be demonstrably justified in a free and democratic society.

[ 38 ] Two central criteria must be met for a limit on a Charter right to be justified under s. 1. First, the objective of the measure must be pressing and substantial in order to justify a limit on a Charter right. This is a threshold requirement, which is analyzed without considering the scope of the infringement, the means employed or the effects of the measure (R. v. K.R.J., 2016 SCC 31, [2016] 1 S.C.R. 906, at para. 61). Second, the means by which the objective is furthered must be proportionate. The proportionality inquiry comprises three components: (i) rational connection to the objective, (ii) minimal impairment of the right, and (iii) proportionality between the effects of the measure (including a balancing of its salutary and deleterious effects) and the stated legislative objective (Oakes, at pp. 138-39; Mounted Police Association of Ontario v. Canada (Attorney General), 2015 SCC 1, [2015] 1 S.C.R. 3, at para. 139; K.R.J., at para. 58). The proportionality inquiry is both normative and contextual, and requires that courts balance the interests of society with those of individuals and groups (K.R.J., at para. 58; Oakes, at p. 139).

[ 39 ] The onus in the s. 1 inquiry is on the party seeking to uphold the limit, that is, in the case at bar, the AGC (Oakes, at pp. 136-37). To discharge this burden, the AGC must satisfy the civil standard of proof on a balance of probabilities (Oakes, at p. 137; RJR-MacDonald Inc. v. Canada (Attorney General), 1995 CanLII 64 (SCC), [1995] 3 S.C.R. 199, at paras. 137-38).

[ 40 ] I appreciate that my colleagues, in dissent, have a different vision of the proper judicial approach to the reasonable limits analysis that is based on their reading of the language of s. 1 itself. This difference in opinion is largely semantic in nature, driven by a disagreement as to whether the term that should be used is "infringement" or "limit". Specifically, my colleagues would prefer that the term "infringement" apply to a limit only once it has been determined that the limit cannot be justified under s. 1. This approach is novel, since a cursory review of the jurisprudence reveals that the terms "infringement" and "limit" are often used interchangeably. This Court has consistently asked whether infringements can be justified under s. 1, and the term "infringement" has not been restricted such that it applies only to unjustified limits on Charter rights (see, e.g., Oakes, at p. 129; Harvey v. New Brunswick (Attorney General), 1996 CanLII 163 (SCC), [1996] 2 S.C.R. 876, at paras. 32 and 51; RJR-MacDonald, at paras. 125-26; Vriend v. Alberta, 1998 CanLII 816 (SCC), [1998] 1 S.C.R. 493, at para. 108; Thomson Newspapers Co. v. Canada (Attorney General), 1998 CanLII 829 (SCC), [1998] 1 S.C.R. 877, at para. 87; Dunmore v. Ontario (Attorney General), 2001 SCC 94, [2001] 3 S.C.R. 1016, at para. 48; Sauvé #2, at paras. 7 and 10; Figueroa, at para. 16; Greater Vancouver Transportation Authority v. Canadian Federation of Students — British Columbia Component, 2009 SCC 31, [2009] 2 S.C.R. 295, at para. 37; Canada (Attorney General) v. Bedford, 2013 SCC 72, [2013] 3 S.C.R. 1101, at para. 126; Carter v. Canada (Attorney General), 2015 SCC 5, [2015] 1 S.C.R. 331, at para. 94; K.R.J., at para. 79; B.C. Freedom of Information and Privacy Association v. British Columbia (Attorney General), 2017 SCC 6, [2017] 1 S.C.R. 93, at paras. 58-59).

[ 41 ] Given that my colleagues' approach would constitute a departure from decades of Charter jurisprudence, was neither raised nor argued at any stage of these proceedings and, above all, need not be considered in order to dispose of this appeal, I will decline to discuss the merits of their position on this point.

[ 42 ] To be clear, then, this case calls for an approach which — far from distorting the analysis, as my colleagues allege — is no more than the application of settled law with respect to how limits on Charter rights can be justified. As I explained above, this Court has developed an analytical approach requiring that two distinct questions be answered: the first is whether a Charter right has been infringed, while the second is whether that infringement can be justified in accordance with the Oakes framework (see, e.g., Andrews v. Law Society of British Columbia, 1989 CanLII 2 (SCC), [1989] 1 S.C.R. 143, at p. 178; Sauvé #2, at para. 10; Bedford, at paras. 125-28). This analytical separation is necessary, in part, because the burden of proof is attributed differently: the rights claimant has the burden of establishing an infringement of his or her Charter right, but it is the state that must justify the infringement (Andrews, at p. 178; Harvey, at para. 30). With respect to the s. 3 right to vote in particular, any balancing of interests must be addressed in the context of the s. 1 justification framework, as opposed to operating as an internal limit on the right (Harvey, at paras. 29-30; Sauvé #2, at para. 11).

[ 43 ] The jurisprudence also requires that a stringent standard of justification be applied when the government seeks to justify a limit on the s. 3 right to vote (Sauvé #2, at para. 14; Figueroa, at para. 60; Opitz, at para. 35). This does not necessarily mean that the government bears a heavier burden in the context of this right than in the context of other Charter rights; as I mentioned above, the standard is always proof on a balance of probabilities. What it does mean is that reviewing courts must examine the government's proffered justification carefully and rigorously in this context rather than adopting a deferential attitude. Deference may be appropriate in the case of a complex regulatory response or a decision involving competing social and political policies, but it is not the appropriate posture for a court reviewing an absolute prohibition of a core democratic right (Sauvé #2, at para. 13; Alberta v. Hutterian Brethren of Wilson Colony, 2009 SCC 37, [2009] 2 S.C.R. 567, at para. 37).

[ 44 ] Although this Court has in the past held that it owes a "natural attitude of deference" to Parliament when dealing with election legislation, it did so in referring specifically to the nuanced choices made by Parliament in selecting and implementing Canada's electoral model (R. v. Bryan, 2007 SCC 12, [2007] 1 S.C.R. 527, at para. 9; Harper v. Canada (Attorney General), 2004 SCC 33, [2004] 1 S.C.R. 827, at para. 87). In the instant case, far from a complex decision concerning the choice and implementation of Canada's electoral model — such as legislation establishing advertising spending limits for third parties in election campaigns, as in Harper, or prohibiting the broadcasting of results on election day, as in Bryan — the provisions at issue reflect, as the AGC concedes, Parliament's decision to disenfranchise long-term non-resident citizens. McLachlin C.J.'s unequivocal statement on the approach to justification in Sauvé #2 is therefore apposite: "The right to vote is fundamental to our democracy and the rule of law and cannot be lightly set aside. Limits on it require not deference, but careful examination. This is not a matter of substituting the Court's philosophical preference for that of the legislature, but ensuring that the legislature's proffered justification is supported by logic and common sense" (para. 9). In the case at bar, citizens are subject to an absolute denial of their Charter right to vote after crossing the five-year non-residence threshold. Accordingly, a stringent standard of justification must be applied to the AGC's proffered justification.

[ 45 ] As I will explain below, the limit in question, which restricts the time that a voter can have been absent from Canada to five years and requires that he or she have a fixed intention to return to Canada, cannot be justified under s. 1, primarily because it is not minimally impairing of the rights at issue.

(1) Pressing and Substantial Objective

[ 46 ] A particularly contentious issue in this appeal concerns the legislative objective which is ostensibly being furthered by the limit on the voting rights of long-term non-resident citizens. The integrity of the justification analysis requires that the legislative objective be properly stated. The relevant objective is that of the infringing measure, not, more broadly, that of the provision (K.R.J., at para. 62; Toronto Star, at para. 20; RJR-MacDonald, at para. 144). The critical importance of articulating the measure's purpose at an appropriate level of generality has also been repeatedly affirmed by this Court (see, e.g., R. v. Moriarity, 2015 SCC 55, [2015] 3 S.C.R. 485, at para. 28). If a legislative purpose is stated too broadly, the result may be to exaggerate the importance of the objective and compromise the analysis (RJR-MacDonald, at para. 144). Conversely, if the measure's purpose is construed too narrowly, its articulation may merely reiterate the means chosen to achieve it (K.R.J., at para. 63).

[ 47 ] This task is difficult in the case at bar because the AGC has framed the legislative objective inconsistently from one stage of the proceedings to another. Before the application judge, the AGC advanced two pressing and substantial objectives: to prevent unfairness for Canada's resident voters (the "fairness objective") and to maintain the proper functioning and integrity of Canada's electoral system. In the Court of Appeal, however, the AGC restated the objectives, arguing primarily that the system "promotes the fairness of the electoral process by protecting the social contract lying at the heart of Canada's constitutional democracy" (C.A. reasons, at para. 90).

[ 48 ] The social contract, famously espoused in the work of the 18th century philosopher Jean-Jacques Rousseau, is purportedly manifested in the connection between the electors and the elected; citizens have a right to elect lawmakers, and a parallel duty to obey the laws enacted by their elected representatives. The majority of the Court of Appeal accepted that preserving the social contract is a pressing and substantial objective for the purposes of the s. 1 analysis. In the majority's opinion, "[p]ermitting all non-resident citizens to vote would allow them to participate in making laws that affect Canadian residents on a daily basis, but have little to no practical consequence for their own daily lives. This would erode the social contract and undermine the legitimacy of the laws" (para. 6).

[ 49 ] I cannot accept that preserving the social contract is a pressing and substantial objective for the purposes of s. 1. In my view, this articulation of the objective is based on a misinterpretation of this Court's decision in Sauvé #2, and superficially and vaguely evokes a political philosophy which is ill-suited to withstand the rigours of the s. 1 justification analysis.

[ 50 ] The majority of the Court of Appeal relied, as does the AGC, on the following isolated passage from Sauvé #2 in support of the argument that preserving the social contract is a pressing and substantial objective:

Denying penitentiary inmates the right to vote misrepresents the nature of our rights and obligations under the law and consequently undermines them. In a democracy such as ours, the power of lawmakers flows from the voting citizens, and lawmakers act as the citizens' proxies. This delegation from voters to legislators gives the law its legitimacy or force. Correlatively, the obligation to obey the law flows from the fact that the law is made by and on behalf of the citizens. In sum, the legitimacy of the law and the obligation to obey the law flow directly from the right of every citizen to vote. As a practical matter, we require all within our country's boundaries to obey its laws, whether or not they vote. But this does not negate the vital symbolic, theoretical and practical connection between having a voice in making the law and being obliged to obey it. This connection, inherited from social contract theory and enshrined in the Charter, stands at the heart of our system of constitutional democracy. [Emphasis added; at para. 31.]

[ 51 ] Their reliance on this passage is selective and, in my view, misinterprets the basic thrust of Sauvé #2, read in its entirety. McLachlin C.J. was describing the social contract in the context of the proportionality analysis, not in articulating a pressing and substantial objective. She adverted to the social contract theory in order to buttress a fundamentally inclusive view of voting rights in Canada, and in rejecting the existence of a rational connection between denying penitentiary inmates the right to vote and enhancing respect for the law. The social contract theory was thus used to anchor the proposition that disenfranchising citizens is anti-democratic and internally contradictory; the power of government flows from its citizens, and it is wrong for the government to use that power to disenfranchise those same citizens (Sauvé #2, at para. 32). As Laskin J.A. noted in dissent in the case at bar, far from offering a compelling rationale for disenfranchising Canadian citizens, Sauvé #2 stands as an uncompromising defence of the right of every Canadian citizen to vote.

[ 52 ] Moreover, the reasons in Sauvé #2 must be read in their entirety. McLachlin C.J. noted later in them that, even though the social contract requires citizens to obey the laws created by way of the democratic process, failure to follow those laws does not nullify a citizen's membership in the Canadian polity: "[W]hether a right is justifiably limited cannot be determined by observing that an offender has, by his or her actions, withdrawn from the social compact" (Sauvé #2, at para. 47). The right to vote is a fundamental democratic right, not a mere privilege, and cannot be denied to citizens on the basis that they have chosen to "opt out" of community membership by offending, or, I would add, by residing outside Canada (Sauvé #2, at para. 42). It might in fact be argued that a citizen who has violated our most serious criminal laws has withdrawn from the social contract to a greater extent than has, for example, a citizen who resides outside Canada on a long-term basis. Be that as it may, Sauvé #2 dictates that deeming that a citizen has "withdrawn" from the social contract is not a legitimate basis for denying him or her the right to vote.

[ 53 ] Perhaps most importantly, the social contract theory is just that: a theory. Preserving it is not an objective. Although moral philosophy doubtlessly has some role to play in the legislative sphere, it cannot readily serve as a source for a pressing and substantial objective in relation to an infringement of Charter rights, and any argument to that effect will require careful scrutiny. For the purposes of the s. 1 analysis, the "social contract" model that has been advanced in this case is devoid of content, and problematically vague. It also has analytical failings: it is at once too general, providing no meaningful ability to analyze the means employed to achieve it, and too narrow, effectively collapsing any distinction between legislative means and ends. This latter point was helpfully illustrated by the intervener David Asper Centre for Constitutional Rights: "[T]he objective (limiting the right to vote to citizens sufficiently subjected to law) and the means (the selection of those citizens who are sufficiently subjected to law) are mutually defined" (I.F., at para. 13). In other words, if we were to accept preserving the social contract as a pressing and substantial objective, then the legislation would have no real objective other than the measure itself: limiting the voting rights of long-term non-resident Canadian citizens.

[ 54 ] Thus, I reject the argument that preserving the social contract suffices as a pressing and substantial objective for the purposes of the s. 1 analysis. But this does not mean the government has failed to identify a sufficiently important objective. If we disregard the use of the expression "social contract", which the AGC concedes is a "convenient shorthand" in any event, she has centrally and consistently asserted that the voting restrictions in question advance the "related" objective of maintaining the fairness of the electoral system to resident Canadians (R.F., at paras. 1 and 55; transcript, at pp. 82 and 84-85).

[ 55 ] I am willing to accept that maintaining the fairness of the electoral system to resident Canadians is a sufficiently important legislative objective to ground the s. 1 analysis. This Court has in the past accepted that variations on promoting electoral fairness and maintaining the integrity of the electoral process are pressing and substantial objectives in the election law context (see Bryan, at paras. 17-19; Harper, at paras. 91-92; Harvey, at para. 38). While I am aware that the purpose of the impugned provisions is to promote greater electoral fairness for only some Canadians — as opposed to enhancing the health of the electoral system in general — I accept that maintaining the integrity and fairness of the electoral system can be a pressing and substantial concern even if the measures taken to achieve that objective impair the democratic rights of other citizens (Harper, at para. 91; Harvey, at para. 38; Bryan, at paras. 33-34).

[ 56 ] The objective of maintaining the fairness of the electoral system is significantly different from that of preserving the social contract. While the latter involves a vague political philosophy, the former raises concerns with respect to the integrity of the electoral system and is not so broad or symbolic as to be immune from review. The fairness objective raised by the AGC and the mechanisms for interpreting it are sufficiently precise for the Court to go ahead with the justification analysis. Even though the importance of ensuring the fairness of the electoral system and, in so doing, maintaining public confidence in it may be difficult to prove empirically, this may nevertheless be a pressing and substantial objective in the election law context (Bryan, at paras. 19 and 22).

[ 57 ] The language of the impugned provisions reflects an intention to establish a connection between non-resident electors and Canada, either through the nature of the elector's employment or by limiting the accepted period of non-residence and ensuring that the elector intends to resume residence in Canada (s. 11 of the Act). I accept that these requirements are intended to advance the overarching objective of maintaining the fairness of the electoral system to resident electors. The AGC argues that maintaining the fairness of the electoral system to resident voters is advanced by ensuring that non-resident voters are sufficiently connected to Canada in terms both of the extent of their subjective commitment to the country and of the extent to which they are affected by Canadian laws. These two aspects substantially reflect the fairness objective that was articulated by the AGC in her original application, which raised such concerns as that non-residents are not as well versed in local issues and that this might unfairly influence the results of elections, and that non-residents no longer have the same connection to Canada in terms of citizenship obligations, whereas resident voters remain subject to all Canadian laws.

[ 58 ] I note in passing that accepting the electoral fairness objective does not violate the rule against shifting objectives, a point on which the majority and the dissent diverged in the Court of Appeal. According to this rule, a court must consider only Parliament's intention at the time the legislation was enacted and must not try to identify new objectives flowing from an updated interpretation of the provision (see R. v. Big M Drug Mart Ltd., 1985 CanLII 69 (SCC), [1985] 1 S.C.R. 295, at p. 335; R. v. Butler, 1992 CanLII 124 (SCC), [1992] 1 S.C.R. 452, at p. 494; R. v. Zundel, 1992 CanLII 75 (SCC), [1992] 2 S.C.R. 731, at p. 761). I am satisfied that the electoral fairness objective corresponds to the legislative intent at the time the provisions were enacted. Although the evidence from the parliamentary debate is not extensive, it shows that when Parliament passed the legislation in 1993, a concern was expressed that the right to vote should be restricted to those non-residents who have "some degree of connection with the country" and with their electoral district (C.A. reasons, at para. 101). The legislation was thus designed to ensure a sufficient connection between non-resident electors and Canada and, out of a concern for electoral fairness, to prevent non-resident electors from voting without restrictions. This is not a new objective that has been invented on the basis of how the legislation is applied today.

(2) Rational Connection

[ 59 ] The question at the first step of the proportionality inquiry is whether the measure that has been adopted is rationally connected to the objective it was designed to achieve. The rational connection step requires that the measure not be "arbitrary, unfair, or based on irrational considerations" (Oakes, at p. 139). Essentially, the government must show that there is a causal connection between the limit and the intended purpose (RJR-MacDonald, at para. 153). In cases in which a causal connection is not scientifically measurable, one can be made out on the basis of reason or logic, as opposed to concrete proof (RJR-MacDonald, at para. 154; Toronto Star, at para. 25).

[ 60 ] In the instant case, the AGC must show that the infringement of non-residents' voting rights is rationally connected to the legislative objective of ensuring electoral fairness to resident voters. In my view, the AGC has not definitively shown that a limit of any duration would be rationally connected to the electoral fairness objective advanced in this case. It is not necessary to come to a firm conclusion on this point, however, as the measure clearly fails at the minimal impairment stage (as discussed below).

[ 61 ] Be that as it may, I would like to comment here on the AGC's submission that the existence of provincial residence requirements supports her argument that there is a rational connection between placing a limit on non-residents' voting rights in federal elections and ensuring electoral fairness to resident voters. In the AGC's view, because residence is a requirement for voting in all Canadian provinces and territories and because the validity of such requirements has been upheld in three provincial and territorial courts, the connection between residence requirements and electoral fairness has been established (Storey v. Zazelenchuk (1984), 1984 CanLII 2426 (SK CA), 36 Sask. R. 103 (C.A.); Reference Re Yukon Election Residency Requirements (1986), 1986 CanLII 3944 (YK CA), 1 Y.R. 23 (C.A.); Anawak v. Nunavut (Chief Electoral Officer), 2008 NUCJ 26, 172 A.C.W.S. (3d) 391). I am not convinced by this argument. This appeal does not concern provincial (or territorial) voting rights, and the situation in the provinces is clearly distinguishable. For instance, a provincial electorate can be significantly smaller and more mobile, a province's laws do not have comparable extraterritorial reach, and provincial identity is distinct from national identity. Most importantly, there is no provincial equivalent to citizenship. Thus, the fact that Canadian provinces and territories impose residence as a condition of voter eligibility is of little relevance to voting entitlements in federal election law. In provincial and territorial elections, residence may be used to establish the necessary connection between the voter and the province or territory. In the context of federal elections, however, Canadian citizenship is itself evidence of the requisite connection.

[ 62 ] Nor does the fact that residence is a requirement of electoral laws in other Westminster democracies — a point relied on by the majority of the Court of Appeal — assist the AGC in establishing the requisite rational connection. The mere fact that a measure is in effect in other countries is of limited utility in determining whether, in the Canadian context, it is rationally connected to the specific legislative objective advanced by the AGC. More broadly, the history of the Canadian electoral system is one of progressive enfranchisement. Canada is an international leader in this respect. For example, a report adduced in evidence before the application judge showed that Canada is one of only four countries, out of 60 democracies surveyed, that does not restrict the right to vote for persons with mental disabilities (see application judge's reasons, at para. 140). In a similar vein, given that many countries restrict the voting rights of incarcerated citizens, the result in Sauvé #2 may well have been different if comparisons with other democracies had weighed more heavily in the analysis. Yet McLachlin C.J. stated in that case that the fact that other democracies adhere to different conclusions regarding the enfranchisement of penitentiary inmates "says little about what the Canadian vision of democracy embodied in the Charter permits" (Sauvé #2, at para. 41). Accordingly, the fact that other democracies have legislated residence-based voting restrictions is of little assistance to us in determining what is required by Canadian democratic rights, as enshrined in this country's Charter. We must not defer to the standards of comparable systems in other countries to establish that there is a rational connection between restricting non-residents' voting rights and ensuring electoral fairness.

[ 63 ] Finally, I would add that the AGC failed to adduce any evidence — either specific to the social contract theory or relating to electoral fairness more generally — to justify the objective of the impugned provisions. Simply put, there is no evidence of the harm that these voting restrictions are meant to address. It has now been possible to vote from outside Canada for over 20 years, and the AGC was unable to identify a single complaint that has been made with respect to voting by non-residents. The absence of any evidence or logical reasoning pointing to a specific problem that needs to be rectified weakens the argument that the limit on voting by non-residents is rationally connected to maintaining electoral fairness.

[ 64 ] I accept that some problems are not easily proved with scientific precision, and that reason and logic are important complements to tangible evidence (K.R.J., at para. 90). This is particularly true for issues in the philosophical, political and social realms. In cases involving such issues, especially in the context of the rational connection analysis, the government can rely on inferential reasoning that is premised on logic and common sense, and not only on concrete evidence, in order to discharge its burden under s. 1 (Sauvé #2, at para. 18; Harper, at para. 29; RJR-MacDonald, at para. 154). In the case at bar, however, unlike in other cases in which the government's objective was to maintain the integrity and fairness of the electoral system (see, e.g., Bryan and Harper), the government has presented essentially no evidence to show how voting by non-resident citizens might compromise the fairness or the integrity of the Canadian electoral system. On the contrary, there have been four parliamentary studies on voting rights whose authors recommended removing any residence-based limit on Canadian citizens' right to vote (see R.F., at paras. 26-32).

[ 65 ] Having made these comments on the connection between the impugned legislation and the objective advanced by the AGC, I will now turn to the minimal impairment stage of the s. 1 analysis. For the reasons that follow, I find that the legislation fails at the minimal impairment stage.

(3) Minimal Impairment

[ 66 ] The second component of the proportionality test requires the government to show that the measure at issue impairs the right as little as reasonably possible in furthering the legislative objective (RJR-MacDonald, at para. 160; Oakes, at p. 139). In other words, the measure must be "carefully tailored" to ensure that rights are impaired no more than is reasonably necessary (RJR-MacDonald, at para. 160; Mounted Police Association, at para. 149). However, some deference must be accorded to the legislature by giving it a certain latitude: "If the law falls within a range of reasonable alternatives, the courts will not find it overbroad merely because they can conceive of an alternative which might better tailor objective to infringement" (RJR-MacDonald, at para. 160).

[ 67 ] The limit on voting by non-residents that is at issue in the instant case is not minimally impairing. There is little indication as to why the specific period of five years was chosen. Far from being a measure that is carefully tailored so as to impair voting rights no more than is reasonably necessary, it seems to have been simply a "middle-of-the-road" compromise, as the application judge noted. Though the five-year limit appears to correspond to the maximum time that can elapse between federal elections, this does not explain why such a limit should be considered to be minimally impairing for the purposes of s. 1.

[ 68 ] Most importantly, the limit on voting by non-residents is not minimally impairing because it is overinclusive on a number of fronts. The fact that the time-based limit is not tailored means that it applies to people to whom it is not intended to apply in furtherance of the electoral fairness objective. The government is seeking to bar people from voting who lack a sufficient connection to Canada. However, the AGC has failed to show a correlation between, on the one hand, how long a Canadian citizen has lived abroad and when he or she intends to return and, on the other hand, the extent of his or her subjective commitment to Canada. Many non-resident citizens who live abroad for many years maintain deep and abiding connections to this country, as illustrated by the appellants in this case. Conversely, there may be citizens who have never left Canada but whose subjective commitment to the country is much weaker and who are less well versed in local issues.

[ 69 ] Moreover, many residents living abroad remain connected to and informed of current affairs in Canada. Modern communications, the international press and globalization mean that the view that time spent living abroad weakens one's subjective commitment to Canada is open to question. Indeed, the evidence adduced before the application judge shows that Canadian citizens living abroad can maintain connections to Canada in many ways, including through family, online media and visits home, and by contributing taxes and collecting social benefits. In short, the measure is not minimally impairing, given that it applies to non-resident citizens who have deep political, familial, financial or cultural roots in Canada, which it does because of a concern that the commitment of such individuals to Canada is insufficient.

[ 70 ] The AGC also submits that the legislation furthers establishing, as a condition for voting, that non-resident electors be regularly subject to laws that they have participated in making. The AGC's argument on this point must fail for three reasons.

[ 71 ] First, legislation does not have an equal impact on all resident citizens. The extent to which citizens are affected by legislation will differ depending on who they are and how they choose to live their lives, regardless of whether they live in Canada or abroad. For example, an elderly voter will likely not be affected by the government's long-term climate policy to the same extent as a young voter. Certain government programs or tax credits may have a greater impact on voters who have children than on those who do not. And the list goes on. The point is this: the attempt to quantify the extent to which a citizen is subject to Canadian legislation is clearly overbroad in the context of electoral fairness concerns. While it may be true that non-resident citizens are subject to fewer Canadian laws than are resident citizens, or are subject to them differently, attempting to tailor Charter rights to the extent to which citizens are burdened, or not burdened, by Canadian laws would be an impossible exercise.

[ 72 ] Second, non-resident citizens do live with the consequences of Canadian legislation. Many non-resident Canadians come home frequently and are subject to Canadian legislation during their visits. Many have family members living in Canada; as the application judge mentioned, Canadian laws affect the resident families of non-resident and resident Canadians alike. And there are Canadian laws that do apply to non-resident citizens while they are living abroad: many laws have extraterritorial application and confer both benefits and burdens on non-resident citizens, including laws with respect to taxation, criminal law, foreign anti-corruption measures, government benefits and citizenship (Canadian American Bar Association, I.F., at paras. 17-18). For example, a Canadian citizen living abroad on a visa obtained under a trade agreement could have his or her life disrupted should that agreement be renegotiated. Government policies can also have global consequences — for example, the government's military, environmental and trade policies have implications that extend well beyond Canada's borders — and non-resident citizens may also be acutely concerned about such issues. Finally, and importantly, Parliament can change laws on its own initiative and thus alter the extent to which Canadian legislation applies to non-resident citizens. The effect of the AGC's position is that the constitutional right to vote would be subject to shifting policy choices. Such an interpretation cannot be correct.

[ 73 ] Third, taken to its logical extreme, the AGC's electoral fairness concern would mean that every non-resident citizen, without exception, should be prevented from voting as soon as he or she takes up residence outside Canada. If the possibility that non-resident electors will not be affected by Canadian laws to a sufficient extent were a valid threat to electoral fairness, the threat would manifest itself the moment they began residing abroad. The fact that the legislation allows for a period of time before non-resident citizens are disenfranchised shows that this electoral fairness concern is not credible.

[ 74 ] In short, the five-year limit cannot be said to be minimally impairing, given that it impairs the rights of many non-resident citizens who maintain deep connections to Canada and many on whom Canadian laws continue to have a significant impact. From this perspective, the limit will in many cases undermine, rather than promote, the underlying objective of electoral fairness advanced by the AGC. It is also argued that the five-year limit is minimally impairing on the basis that it falls within the range of external voting limits drawn in countries such as Australia and New Zealand. For the reasons set out above, I place little stock in comparisons with other countries for the purpose of determining whether this legislation is constitutional. I would simply note that such comparisons are unhelpful in any event. For example, the limit the United Kingdom places on the voting rights of its non-resident citizens takes effect when they have been out of the country for 15 years. Further, the limits on voting by non-residents of six years in Australia and three years in New Zealand are actually far less stringent than the five-year limit at issue in this case: in Australia, the voter's right can be extended indefinitely upon application, while New Zealand's three-year limit is reset each time the non-resident returns to the country.

[ 75 ] The AGC has failed to show why limiting the voting rights of non-residents to citizens who have been abroad for less than five years and requiring that they have a fixed intention to return are minimally impairing in relation to the legislative objective. These requirements go too far, because they do not effectively reflect a citizen's connection to Canada, and as a result they impair the voting rights of citizens to whom, given the stated objective, the legislation should not apply. A non-resident citizen who takes the trouble to vote by way of special ballot while living abroad has demonstrated a profound attachment to Canada. We have nothing to gain from disenfranchising such citizens. Indeed, our democracy is manifestly strengthened by such demonstrations of civic commitment.

(4) Balancing the Salutary and Deleterious Effects

[ 76 ] At the final stage of the Oakes test, it must be asked whether there is proportionality between the overall effects of the Charter-infringing measure and the legislative objective (Oakes, at p. 139; Hutterian Brethren, at paras. 72-73). Whereas the preceding steps of the Oakes test are focused on the measure's purpose, at this stage the assessment is rooted in a consideration of its effects (Hutterian Brethren, at para. 76). This allows a court to determine on a normative basis whether the infringement of the right in question can be justified in a free and democratic society.

[ 77 ] At this stage, I would simply observe that any salutary effects of ensuring electoral fairness, as asserted by the AGC, are clearly outweighed by the deleterious effects of disenfranchising well over one million non-resident Canadians who are abroad for five years or more.

[ 78 ] The benefits of the impugned legislation are illusory. It is unclear how the fairness of the electoral system is enhanced when long-term non-resident citizens are denied the right to vote. Despite some vague notions that it might be unfair to allow non-resident citizens to vote indefinitely, there is little by way of either concrete evidence or principled logic to show any actual benefits that result from the limit. The few types of harm to which, according to the AGC, the limit responds, such as the concern that voting by long-term non-residents should not affect a local constituency, are speculative in nature.

[ 79 ] The deleterious effects of the infringement on affected non-resident citizens, on the other hand, are serious. As I mentioned above, certain government decisions will have a unique or disproportionate impact on non-resident citizens; on some issues, the limit would restrict the participation of the very individuals who have the most at stake.

[ 80 ] Many of Canada's best and brightest citizens live abroad — and indeed are encouraged to do so — for such purposes as pursuing educational and professional opportunities, and in the course of their endeavours, they are often ambassadors of Canadian values. Sometimes the ability of such citizens to return is hampered by circumstances beyond their control, but they nevertheless maintain strong and abiding connections to Canada. The appellants in this case are good examples of this. Dr. Frank was raised and educated in Canada. He moved to the United States on a full scholarship for graduate studies. He has not sought immigration status in that country, as he does not intend to stay there permanently. Mr. Duong also moved to the United States for educational purposes. Both appellants visit Canada frequently, continue to be actively interested in Canadian issues, and have made an effort to demonstrate their civic commitment by voting in the past. They have not voluntarily severed their connections to Canada. And yet they are denied a fundamental democratic right on this very basis.

[ 81 ] It is no answer to say that the deleterious effects are "measured" and that non-resident citizens can move back whenever they choose and in that way regain the right to vote (C.A. reasons, at para. 157). This justification is based on a misinterpretation of the nature of the choice available to non-resident citizens and also, more fundamentally, distorts the proper approach to the justification of a Charter infringement. To suggest that the infringement ends when affected citizens resume residence in Canada does not detract from the harm done by an absolute prohibition on the exercise of a core democratic right. In no other context do we tolerate the idea that a person can earn his or her Charter rights back through voluntary conduct. Reasoning to that effect must also fail here. As this Court noted in response to a similar argument that had been advanced in Sauvé #2: "The silenced messages cannot be retrieved, and the prospect of someday participating in the political system is cold comfort to those whose rights are denied in the present" (para. 60).

[ 82 ] Voting is the cornerstone of our democracy. Denying citizens the right to vote not only strikes at the heart of their fundamental rights, but also comes at the expense of their dignity and their sense of self-worth (Sauvé #2, at para. 35). As McLachlin C.J. noted in Sauvé #2: "When basic political rights are denied, proof of additional harm is not required" (para. 59). In other words, denial of the fundamental right to vote, in and of itself, inflicts harm on affected citizens. This is particularly true where there is simply no convincing rationale to support the denial. In the absence of evidence pointing to a concrete problem, the justification boils down to an argument based on worthiness: the non-resident citizens in question are deemed to be less deserving of the right to vote than the resident majority on the basis that they have voluntarily left Canada and severed their connection to the country. However, this Court has quite properly foreclosed the use of such worthiness rationales to justify restrictions on the right to vote in past cases. Worthiness cannot be used to justify the disenfranchisement of non-resident Canadian citizens in the case at bar.

VI. Conclusion

[ 83 ] I would allow the appeal, with costs throughout. The impugned provisions of the Act infringe s. 3 of the Charter, and the infringement is not justified under s. 1. It follows that these provisions are inconsistent with the Charter. Accordingly, based on s. 52 of the Constitution Act, 1982, ss. 222(1)(b) and (c), 223(1)(f) and 226(f) of the Act are declared to be of no force or effect; the words "a person who has been absent from Canada for less than five consecutive years and who intends to return to Canada as a resident" are struck from s. 11(d) of the Act and are replaced with the words "an elector who resides outside Canada"; and the word "temporarily" is struck from ss. 220, 222(1) and 223(1)(e) of the Act.

The following are the reasons delivered by

Rowe J. —

[ 84 ] At issue in this appeal is the constitutionality of the limit on voting in federal elections for citizens who have not been resident in Canada for five years or more. The federal government has conceded that this limit constitutes an infringement of s. 3 of the Canadian Charter of Rights and Freedoms. I agree that the federal government has not met the requirements to justify this limit under s. 1. However, in my view, any evaluation of this kind of limit should acknowledge the significance and centrality of residence to our system of representative democracy, and should not foreclose the possibility that residence requirements in another context might be constitutional. While I would therefore allow the appeal, I would do so on a basis different from that of the majority.

A. The Significance of Residence and the Right to Vote

[ 85 ] The impugned measures are consistent with Canada's historical approach to non-resident voting. Since 1920, federal statutes have governed voter eligibility and balloting procedures for federal elections in Canada (Dominion Elections Act, S.C. 1920, c. 46). Prior to 1920, federal electoral laws were based on provincially enacted franchise requirements. For all but a few Canadians — notably the military — residence of some term has been a requirement to vote since that time. While residence continued as a prerequisite to vote for most Canadians, beginning in 1955, non-resident voting rights were slowly expanded. That year, the non-resident voting right of Canadian Armed Forces ("CAF") electors was extended to their spouses while the CAF elector resides outside Canada. In 1970, non-resident voting was extended to the dependents of CAF electors, as well as those working abroad in embassies and consulates, and their dependents.

[ 86 ] With the advent of the Charter, a series of reports recommended comprehensive reforms to the Canada Elections Act, S.C. 2000, c. 9 ("Act"). The 1986 White Paper on Election Law Reform recommended extending the franchise to voters living abroad without limitation. Another report followed in 1991: Reforming Electoral Democracy, vol. 1, issued by the Royal Commission on Electoral Reform and Party Financing ("Lortie Commission"). The report acknowledged a concern that had been raised previously, that some Canadians living abroad may not have a sufficient connection to Canada. Nonetheless, the Lortie Commission recommended that non-resident Canadian citizens be eligible to vote provided they "demonstrate their continuing attachment to the Canadian polity if they wish to participate in its political processes" (Lortie Commission's Report, at p. 46).

[ 87 ] In 1992, a House of Commons Special Committee on Electoral Reform reviewed the Lortie Commission's Report and recommended changes to the Act. The Special Committee did not accept the Lortie Commission's recommendation on non-resident voting, but instead proposed amendments to allow voting by citizens who are: (i) outside the electoral district in which they reside but within Canada on voting day; (ii) temporarily out of the country during the election period, but resident in Canada; and (iii) living abroad for less than five years and intending to return to Canada (House of Commons, Special Committee on Electoral Reform, Third Report to the House of Commons, No. 7, 3rd Sess., 34th Parl., December 11, 1992, at pp. 7:3-7:5). While the impugned provisions had the effect of enfranchising Canadians who would have previously been denied the opportunity to vote, they represented a choice to maintain a limit on long-term non-resident voting, and an affirmation of the importance of residence from the perspective of Parliament.

[ 88 ] The pride of place given to residence in our election laws should be understood in its historical context. As Canadian citizenship did not receive legal recognition until 1947, residence was historically the only means through which electors demonstrated their connection to Canada, and in turn, their eligibility to vote. While citizenship may now be understood as a basis for connection to Canada, its presence in the text of s. 3 of the Charter does not undermine the significance of residence. Instead, residence remains fundamental to the Canadian electoral system.

[ 89 ] Canada, in line with the Westminster system, has no nationally-elected positions. For example, no one votes for the Prime Minister in the way that citizens of the French Republic vote for their President. Instead, citizens vote for representatives of their geographically defined constituencies. As the Attorney General of Canada ("AGC") points out, "a federal general election can in some important respects be understood as a collection of 338 local constituency elections held simultaneously" (R.F., at para. 14). This system "ensures that Canadian elections incorporate local as well as national influences" (R.F., at para. 14). The underlying premise is that a federal policy can impact different geographically defined communities in different ways. For example, the Patent Act, R.S.C. 1985, c. P-4, has greater importance for the pharmaceutical industry in the Montréal area than it does in Regina. And official languages policy has greater importance in New Brunswick than it does in British Columbia. Residence is significant because it establishes a connection to a particular electoral district and to the concerns of persons living in that electoral district.

[ 90 ] This aspect of our representative democracy is not constitutionally entrenched; however, residence has been historically and remains today more than just an "organizing mechanism". It is foundational to our system; it has been so since Parliaments were first convened in England in the 13th century. Increased mobility and modern communication technologies do not break the organic link between residence and voting, nor do they transform residence into an obsolete concept. I agree that its significance does not elevate residence to an inherent limit on the s. 3 right to vote. However, the fact that residence as a general concept is not an inherent limit on s. 3 says nothing about whether any given residence requirement could constitute a limit on s. 3 or whether any such limit is justified under s. 1. That s. 3 does not refer to residence speaks to but does not decide the first point; it says nothing about the second.

[ 91 ] Section 3 protects the right to vote, but it does not follow as a corollary that there is a right to vote in the constituency or province of one's choosing. The provinces and territories have each crafted residence requirements that reflect the concerns and circumstances that are particular to their jurisdiction. For example, in Nunavut, to be eligible to vote in a territorial election, a person must have been resident in that territory for "a consecutive period of at least 12 months" (Nunavut Elections Act, S. Nu. 2002, c. 17, s. 7(1)). The Nunavut Elections Act also provides that "[n]o person is to be considered resident in a home or dwelling that the person occupies seasonally for a period not exceeding a total of 180 days a year, unless at the time of an election the person has no residence in any other place" (Nunavut Elections Act, s. 4(12)). In Ontario, as a general rule, individuals who are absent for more than two years lose their eligibility to vote in a provincial election (Election Act, R.S.O. 1990, c. E.6, s. 15(1.1)). In Quebec, persons must have been "domiciled" in the province for six months to be eligible to vote in a provincial election (Election Act, R.S.Q., c. E-3.3, s. 1). In my view, the concession by the AGC that the impugned measures infringe s. 3 does not prejudice provincial/territorial governments in arguing that their legislation does not. Nor does it decide the constitutionality of any other federal residence requirement. In any event, different considerations will apply at the s. 1 analysis of any established (or conceded) breach of s. 3 at the provincial level. I would note that in Reference re Yukon Election Residence Requirements (1986), 1986 CanLII 3944 (YK CA), 1 Y.R. 23 (C.A.), in considering the constitutionality of the territory's residence requirements, the Yukon Territory Court of Appeal considered the small margin by which many constituencies in the territory are won in elections (at para. 15), and the significant transient population passing through the territory (at para. 18), which could significantly impact local interests. Yukon residence requirements were upheld as constitutional in light of these circumstances. As this Court held in Haig, "[t]erritorial exigencies, such as those present in the northern territories, may justify a host of rules particular to a given province, and the possibility of such divergence is woven into the very fabric of Canadian federalism itself" (Haig v. Canada (Chief Electoral Officer), 1993 CanLII 58 (SCC), [1993] 2 S.C.R. 995, at p. 1029). Evidence of the circumstances relating to the various residence requirements in each of the provinces and territories may well affect the analysis of any alleged Charter infringement. It seems to me that all the above residence requirements would run afoul of the Charter if we take the view, which I do not, that residence bears no rational connection to electoral fairness and can never be the basis for limiting the right to vote. While their constitutionality awaits consideration another day (should they be challenged), the approach adopted in this case will surely be relevant.

B. Section 1 Analysis

[ 92 ] Promoting electoral fairness is a pressing and substantial objective. As the AGC clarified, "preserving the social contract" was only shorthand for that objective. In my view, there is also a rational connection between the impugned measures and this objective. However, the impugned measures ultimately do not withstand s. 1 scrutiny. What is decisive for me is the balancing of salutary and deleterious effects. Accordingly, I do not address whether the impugned measures are minimally impairing.

[ 93 ] Before proceeding to this analysis, it is important to highlight the manner in which electoral fairness is engaged. Unlike other kinds of measures said to promote electoral fairness, for example to prevent fraudulent voting, the impugned measures in this case do not promote electoral fairness for all electors. Instead, they promote the interests of one group of citizens (residents) at the expense of another group of citizens (long-term non-residents). As the AGC argues, one purpose of the impugned measures is to "maintain the fairness of the electoral system to the resident Canadian" (R.F., at para. 1 (emphasis added)). This differs from other cases considered by this Court in which laws intended to promote electoral fairness did so for all electors. For example, in Harper v. Canada (Attorney General), 2004 SCC 33, [2004] 1 S.C.R. 827, this Court considered the constitutionality of laws limiting election spending during campaign periods. The provisions at issue were intended to promote equality in political discourse, and were ultimately found to be constitutional, notwithstanding the limits they imposed on freedom of expression. The inherent opposition of interests in this case is important to keep in view when evaluating the impugned measures. In particular, do the benefits (for resident citizens) of the limits on long-term non-resident voting outweigh the consequences for long-term non-residents?

(1) Rational Connection

[ 94 ] The government must demonstrate that the impugned measures are rationally connected to the pressing and substantial objective. In my view, the impugned measures are rationally connected to the objective of maintaining electoral fairness to resident Canadians.

[ 95 ] At this stage of the analysis, the government must show through evidence or common sense, that the chosen means promote the pressing and substantial objective. The government need not demonstrate this objective on a civil standard of proof. Instead, "it is sufficient for the government to demonstrate that it [has] a reasonable basis for believing such a rational connection exists" (RJR-MacDonald Inc. v. Canada (Attorney General), 1995 CanLII 64 (SCC), [1995] 3 S.C.R. 199, at para. 82). The goals of the legislature should be "logically furthered" by the means chosen (Lavigne v. Ontario Public Service Employees Union, 1991 CanLII 68 (SCC), [1991] 2 S.C.R. 211, at p. 291). Whether a law is rationally connected to the identified pressing and substantial objective does not depend on the measures being faultlessly calibrated to that objective. Simply put, the bar is not a high one.

[ 96 ] It has been suggested that the appropriate focus at this stage is on whether any temporal limit is rationally connected to the electoral fairness objective. In my view, this focus is misplaced. Rather, what must be evaluated is whether residence is rationally connected to electoral fairness. The AGC submits that the impugned provisions help to ensure that voters maintain both a subjective and objective connection to Canada, and that maintenance of this connection furthers the objective of electoral fairness. The subjective connection relates to an individual's engagement with Canadian and with local (constituency) issues. The premise is that it is unfair for those not engaged with such issues to vote. The objective connection relates to whether an individual is subject to Canadian laws. The premise is that it is unfair for those not subject (or who are markedly less so) to participate in deciding laws that govern others who are resident and who will be subject fully to Canadian laws.

[ 97 ] While the parties were not always clear, as I understand the AGC's arguments there are two ways in which long-term non-resident voting may be unfair to resident Canadians. These correspond to two different senses of residence or connection. First, long-term non-resident voting is unfair because non-residents do not have a sufficient connection to the particular constituencies in which they would vote. Second, long-term non-resident voting is unfair because non-residents do not have a sufficient connection to Canada.

[ 98 ] With respect to the first, in our electoral system, Canadians elect a Member of Parliament ("MP") to represent their constituency in the House of Commons; this MP understands and speaks for his or her communities and the mix of challenges that his or her constituents face. But, non-residents would cast votes for a candidate that they would almost certainly never meet; and that candidate would represent a constituency where non-residents have not lived for a long time (if ever) and to which they may have no intention of returning. Unfairness arises when individuals who do not have to live with the local consequences of government decisions have a say in determining which MP will represent that community.

[ 99 ] In passing, I would refer to the issue of where long-term non-residents would vote. Under the Act, temporary non-residents who want to vote by special ballot must submit an application form to be included on a register of temporary non-resident electors. In applying to be included on the register, citizens must provide the following information, inter alia (s. 223(1)(e)):

the address of the elector's last place of ordinary residence in Canada before he or she left Canada or the address of the place of ordinary residence in Canada of the spouse, the common-law partner or a relative of the elector, a relative of the elector's spouse or common-law partner, a person in relation to whom the elector is a dependant or a person with whom the elector would live but for his or her residing temporarily outside Canada.

The constituency in which one votes is determined on the basis of this information. If restrictions on voting for long-term non-residents are removed, the place where one's ballot is cast would be determined in the same way.

[ 100 ] The second unfairness relates to connection on a national rather than at a constituency level. The argument advanced by the AGC is that long-term non-residents are not sufficiently connected to Canada, and that they do not have a sufficient stake in Canadian laws and policies as compared with resident Canadians. For example, there is a common sense unfairness if a non-resident supports significant tax increases by which they will never be burdened (save for tax on any income earned in Canada).

[ 101 ] To summarize, the AGC's arguments identify two kinds of unfairness: the unfairness of non-residents voting somewhere (because they lack a sufficient connection to a particular constituency), and the unfairness of non-residents voting anywhere (because they lack a sufficient connection to Canada).

[ 102 ] The underlying logic of the foregoing is embedded in Chief Justice McLachlin's reasons in Sauvé v. Canada (Chief Electoral Officer), 2002 SCC 68, [2002] 3 S.C.R. 519 (Sauvé #2). Those reasons establish a connection between the right to vote and the obligation to obey the law (at para. 31):

This delegation from voters to legislators gives the law its legitimacy or force. Correlatively, the obligation to obey the law flows from the fact that the law is made by and on behalf of the citizens. In sum, the legitimacy of the law and the obligation to obey the law flow directly from the right of every citizen to vote. As a practical matter, we require all within our country's boundaries to obey its laws, whether or not they vote. But this does not negate the vital symbolic, theoretical, and practical connection between having a voice in making the law and being obliged to obey it.

McLachlin C.J. focused on the problem with being subject to laws when one cannot vote. Does not the logic work the other way as well? If the law's legitimacy derives from the fact that those who are subject to it are the ones who (indirectly) create it, then is it not also unfair that individuals who are not subject to or affected by the law can decide for those who are? I have difficulty finding otherwise.

[ 103 ] We must then ask whether there is a relationship between residence and an individual's connection to a particular constituency or, perhaps, to Canada generally. Asked differently, are long-term non-residents likely to be less connected to a constituency or to Canada? Especially at the constituency level, I find it difficult to escape the conclusion that long-term non-residents are likely to be less connected to their community. As a general proposition, individuals who have not lived in a constituency for over five years are less likely to be informed about the issues affecting that constituency. (How could it be otherwise?) And, if the long-term non-resident voter has no intention to return to that constituency, they will not feel the impacts of federal laws and policies as they manifest at the local level. While perhaps somewhat attenuated, the same logic applies at the national level. Those who have not lived in Canada for a long time are likely to be less connected to Canada than those who reside here. While non-resident citizens are affected to some degree by Canadian law, they are to a far lesser degree than are Canadians living in this country. Thus, there would be the unfairness of those who are largely unaffected (non-residents) participating in decisions that would affect others (residents).

(2) Proportionality of Effects

[ 104 ] This brings me to whether the salutary effects of promoting fairness for resident Canadians are outweighed by the deleterious effects of denying long-term non-resident citizens the right to vote in federal elections. For me, this stage of the proportionality analysis is decisive in this appeal.

[ 105 ] The salutary effects of an impugned measure are evaluated according to the extent to which it promotes the objective which was found to be pressing and substantial at the first stage of the analysis. The consequence of limits on long-term non-resident voting are that these individuals will never cast a decisive vote in a constituency in which they are not resident. This is a salutary effect of the impugned measures, as the local representative will not be selected by individuals from outside the constituency. While this is so as a general proposition, the AGC has offered almost no evidence of the impact that long-term non-residents would or could have had either locally or nationally if permitted to vote.

[ 106 ] What evidence we have suggests that the impact would likely be negligible. A very small number of Canadians living abroad who are currently eligible to vote choose to exercise their right to vote through a special ballot. In the election that followed the implementation of the Special Voting Rules under Part 11 of the Act, which establish the procedures for special ballot voting, including votes to be cast by non-residents, a little over 15,000 special ballots were requested and issued. In the 2011 election, in the ten Canadian ridings with the highest number of special ballots, as a percentage of total registered electors in that constituency, the non-resident votes ranged from a low of 0.05 percent to a high of 0.2 percent. Elections Canada reported that barely 6,000 votes were recorded from international electors. While these numbers pertain to non-residents who were eligible to vote under the current statutory scheme — that is, citizens who were non-resident for less than five years — they are nonetheless instructive. As Penny J. explained in the application decision, "it is hard to see what unfairness is being visited on the resident majority by the voting of other non-resident citizens" (para. 113). In such circumstances, the salutary effects are inconsequential. This would be different if large numbers of non-resident voters cast ballots, especially if they did so in a coordinated way so as to maximize the impact on election results. A million additional voters could well sway an election.

[ 107 ] By contrast, the deleterious effects of the provisions on long-term non-residents are clear: they cannot vote. While they may not feel the local consequences of particular federal policies in the constituencies in which their votes would be counted, they stand nonetheless to be affected by certain federal laws and policies, perhaps in life-altering ways. By the logic of the majority in Sauvé #2, we should be alive to the reality of having no voice in relation to matters by which one may be deeply affected. I note that the same logic may not apply at the provincial and territorial level; long-term non-residents are not likely to be significantly affected by provincial and territorial laws as provincial and territorial jurisdiction does not generally extend to matters that cross international borders.

[ 108 ] I would add that s. 3 has value beyond playing a "meaningful role in the selection of elected representatives" (Figueroa v. Canada (Attorney General), 2003 SCC 37, [2003] 1 S.C.R. 912, at para. 25). The right is not merely instrumental; it does not only protect the opportunity to be represented by one's favoured candidate. As Iacobucci J. recognized in Figueroa, the right to vote "has an intrinsic value independent of its impact upon the actual outcome of elections" (para. 29). Voting is the "primary means by which the average citizen participates in the open debate that animates the determination of social policy" (para. 29). Each vote has expressive content, communicating not only the identity of one's preferred candidate, but the values and policy choices they represent. Denying long-term non-residents the right to vote accordingly denies those citizens the "opportunity to express an opinion about the formation of social policy and the functioning of public institutions through participation in the electoral process" (para. 29).

[ 109 ] Thus, while on the evidence before us the salutary effects are inconsequential, the deleterious effects are clear, and outweigh the salutary effects.

C. Conclusion

[ 110 ] I do not want to close the door to any and all possible limits on voting federally based on residence. Nor do I want to prejudge, by implication, the constitutionality of limits based on residence in the provinces and territories. However, on the basis set out above, I would find that the limit on s. 3 conceded by the AGC has not been justified under s. 1.

The following are the reasons delivered by

Côté and Brown JJ. (dissenting) —

I. Overview

[ 111 ] In 1993, Parliament extended voting rights to a significant number of Canadians living abroad, including all citizens who have been absent from Canada for less than five consecutive years. But it left in place long-standing statutory restrictions which precluded from voting most citizens who have lived outside of Canada for five consecutive years or more and who are not otherwise eligible to vote under specific exemptions, such as public service employees and their families. This appeal requires us to consider the constitutionality of those restrictions — specifically, whether the prohibition on long-term non-resident voting is a reasonable limit on the appellants' democratic right to vote under s. 3 of Canadian Charter of Rights and Freedoms, or whether s. 3 requires that all citizens, regardless of how long they have lived abroad, or even if they have never set foot in Canada, be permitted to cast a ballot in federal (and, for that matter, provincial) elections.

[ 112 ] The Attorney General of Canada concedes that the restriction on long-term non-resident voting constitutes a limit on the appellants' s. 3 right, but says that this limit is constitutional. Subject to what we say at para. 123, we agree. The objective of the 1993 amendments was to expand the franchise to non-resident citizens temporarily living outside of Canada, while preserving a relationship of currency between electors and their communities in Canada. This objective is pressing and substantial, and the means that Parliament chose to achieve it — a voting regime akin to those in place in comparable Westminster democracies, whose parliaments, like Canada's, are constituted of elected community representatives — are demonstrably justifiable in a free and democratic society.

[ 113 ] While we therefore reach a different result than the Chief Justice, we see our fundamental point of disagreement as lying in the proper judicial approach to the limitations analysis. We start from the premise that s. 3 is a positive right which, unlike most Charter rights, requires legislative specification in order for the right to be operative. It follows that Parliament acted to define and shape the boundaries and contours of a positive entitlement which, as such, necessarily requires legislative specification.

[ 114 ] Moreover, and contrary to the seemingly absolute quality of the majority's understanding of s. 3's guarantee, such legislative specification is not only necessary, but uncontroversial. Nobody suggests that s. 3 entitles three-year-old Canadian citizens to vote. Similarly, the Canada Elections Act, S.C. 2000, c. 9, withholds the vote from Canadian citizens who have never lived in Canada. And yet, and as the appellants concede, their logic — and, we add, the logic of the majority reasons in this appeal — would necessarily invalidate that restriction as well, opening the vote to persons who have never before so much as set foot in Canada. Far from an example of "progressive enfranchisement" (majority reasons, at paras. 2 and 62), we see this development as regressive, undermining the longstanding and entirely salutary practice in Westminster parliamentary democracies of privileging local connections in deciding who may elect local representatives. In any event, "progressive" or not, the impugned limit on the right to vote is reasonable. We would therefore dismiss the appeal.

II. Background

[ 115 ] The Act governs voter eligibility and balloting procedures for federal elections in Canada. It provides citizens with three ways to vote: (1) in person at a polling station on Election Day; (2) in person at an advance poll prior to Election Day; or (3) by means of a special ballot (s. 127(a) to (c)). This appeal concerns the third category — specifically, the circumstances in which a special ballot can be issued to citizens living abroad.

[ 116 ] Prior to 1993, non-resident Canadians were not permitted to vote by special ballot — indeed, they were not permitted to vote at all — unless they were employed in the service of Canada (i.e., diplomats, military personnel, or other public servants) or were a family member of an eligible elector posted abroad. This restriction on voting applied regardless of how long a citizen lived outside the country.

[ 117 ] Beginning in the early 1980s, a series of reports, royal commissions, and a House of Commons Special Committee considered the issue of absentee voting. These efforts culminated in Parliament's decision in 1993 to expand the franchise for Canadians living abroad by amending the Act to permit citizens who have been absent from Canada for fewer than five consecutive years to vote, for the first time, by special ballot — irrespective of their employment or the reason they left Canada.

[ 118 ] Today, the process of voting by special ballot — the provisions at the crux of this appeal — is outlined in s. 11 and Part 11 of the Act. The effect of s. 11 is that long-term non-residents — citizens who have lived outside of Canada for five consecutive years or more, or those who do not intend to return — cannot vote by special ballot, unless they fall into one of the exceptions. In substance, this restriction pre-dated the 1993 amendments. Long-term non-residents were never permitted to vote by special ballot from abroad. The 1993 amendments, however, granted citizens who have been away for fewer than five years the ability to vote by special ballot, so long as they intend to re-establish residence in Canada in the future. Thus, far from "disenfranchising" Canadians (majority reasons, at para. 75), the legislation at issue in this appeal had precisely the opposite effect. It enfranchised a segment of Canadian citizenry which had never before been able to vote. As we describe below (at paras. 126, 132 and 137), this is precisely how Parliament saw the matter. And while it is obviously not determinative, we further observe that, contrary to the majority's statement, the Crown has not conceded (nor can the Crown be fairly taken to have conceded) that the impugned provisions reflect "Parliament's decision to disenfranchise" Canadians (majority reasons, at para. 44).

[ 119 ] A separate question — not directly at issue in this appeal — is whether the Act permits long-term non-residents to vote in person. In practice, an Elections Canada administrative policy permits any non-resident to vote by physically attending a polling station, either on Election Day or during the advance voting period. The Attorney General submits that this policy rests on an erroneous interpretation of the Act, and argues that long-term non-residents (except those who are enfranchised under other provisions) cannot vote in any manner (transcript, at pp. 76-77). Since it is not strictly necessary to resolve this question, we limit our discussion below to the constitutionality of the special ballot restriction. The appellants themselves conceded that in-person voting is largely "an illusory possibility" for long-term non-residents (transcript, at p. 13). The restriction on casting a special ballot therefore amounts to a de facto restriction on voting altogether, and it should be treated as such for purposes of the limitations analysis.

III. Analysis

[ 120 ] We begin by observing, respectfully, that the majority distorts the limitations analysis to speak of an "infringement" or "breach" based solely on the fact that an impugned measure imposes a limit on a Charter right (see, e.g., majority reasons, at paras. 24, 31, 38, 42, 53, 60, 76, 79, 81 and 83). By way of explanation, s. 1 "permits and envisions that the rights and freedoms enshrined in the Charter" are subject to "rational and reasonable limitations" (Sauvé v. Canada (Chief Electoral Officer), 2002 SCC 68, [2002] 3 S.C.R. 519 ("Sauvé #2"), at paras. 84 and 89 (emphasis added); see also R. v. Oakes, 1986 CanLII 46 (SCC), [1986] 1 S.C.R. 103, at p. 136) and not "rational and reasonable infringements". This is, we note, supported by the text of s. 1 itself, which speaks not of reasonable and demonstrably justifiable infringements, but of reasonable and demonstrably justifiable limits. This is simply one instance (albeit one that is constitutionally enshrined) of the principle that no right is absolute, including Charter rights such as the s. 3 right to vote. To be clear, then, a reasonable limit does not justify an infringement, but is inherent in the right itself, shaping the right's outer boundaries (B. W. Miller, "Justification and Rights Limitations", in G. Huscroft, ed., Expounding the Constitution: Essays in Constitutional Theory (2008), 93, at p. 96). Our recent jurisprudence confirms this treatment of s. 1's provision for "reasonable limits" (see, e.g., R. v. K.R.J., 2016 SCC 31, [2016] 1 S.C.R. 906). In short, a right is infringed only where the right, as reasonably limited, is breached. The issue presented by this appeal, then, is not whether the limit to the right to vote effected by the restriction on long-term non-resident voting justifies an infringement of s. 3, but whether that limit is unreasonable, such that s. 3 is infringed.

[ 121 ] We do not dispute, as the majority seems to suggest (at para. 42), that an analytical separation is necessary so as to distinguish the claimant's burden of establishing a limit on his or her rights from the government's burden of justifying the limit under s. 1. Our reasons do not raise a question of burden; we accept that the burden always rests upon the state to justify limits to rights (Law Society of British Columbia v. Trinity Western University, 2018 SCC 32, [2018] 2 S.C.R. 453, at para. 314). The point of difference between us and the majority is much narrower than the majority conceives, and is reduced to this question: what is it that is being justified — an infringement, or a limit? We do not believe the Charter can (or should) be read so as to ever allow for justified "infringements". It does, however, allow for justified "limits". An "infringement" is a limit that is not justified (K.R.J., at paras. 91, 115 and 116; Bracken v. Niagara Parks Police, 2018 ONCA 261, 141 O.R. (3d) 168, at para. 33). We respectfully observe that the majority's reasons confuse those concepts and elide the text of s. 1. A conceptually sensible and textually faithful account of the s. 1 analysis properly focuses on whether a limit on a Charter right is justified.

[ 122 ] The majority's response to this is, in our respectful view, internally inconsistent. On the one hand, it sees the matter as being of sufficient importance to insist upon departing from the text of s. 1. But it also sees the matter as one of "semantics", noting that the Court has, over the years, used the terms "infringement" and "limit" interchangeably. We agree that the Court has drifted in and out of speaking of justifying "infringements". Our point is not that it has never done so. Our point is that it should stop doing so. If this is, as the majority says, a matter of "semantics", then there is no good reason for continuing to ignore constitutional text. In any event, we view the question as transcending mere semantics. It distorts our constitutional discourse, and our understanding of rights and of the legitimate boundaries of state action, to speak of individuals having rights which may be justifiably violated by the state. Indeed, the majority's reasons furnish an example, by suggesting that s. 3 grants an absolute right that is violated by the restrictions on long-term non-resident voting. Reasonable limits are "inherent in the rights themselves", and it is those limits which must be justified (Miller, at p. 96).

[ 123 ] We note that, before us, counsel for the Attorney General also perpetuated this analytical error by conceding an "infringement". Were the Attorney General to concede an infringement, there would be no point to hearing this appeal: she would be conceding the case. As we have already noted, however, we read the Attorney General's concession as being that the restriction on long-term non-resident voting constitutes a limit on the s. 3 Charter right to vote, leaving as the only issue whether this limit is demonstrably justified in a free and democratic society. We are fortified in this view by the Attorney General's statement in her Principles guiding the Attorney General of Canada in Charter litigation:

. . . Parliament may enact laws that limit rights and freedoms, and . . . the Charter will be violated only where a limit is without justification.

As a result, the Attorney General will sometimes apply the principle of constitutionalism and the rule of law by recognizing that a right or freedom has been limited, but without conceding that the limitation is without justification. Instead, the Attorney General may seek to demonstrate through litigation that federal legislation is justified in limiting rights and freedoms, thereby respecting the Charter. [Emphasis added; p. 6.]

(Canada, Department of Justice, Principles guiding the Attorney General of Canada in Charter litigation (2017) (online))

[ 124 ] The 1993 amendments record Parliament's effort to specify the limits of the s. 3 right — that is, to define and shape the boundaries and contours of a positive entitlement which, as such, necessarily requires legislative specification. To determine whether a limit is justified, this Court has endorsed a "flexible contextual approach" (Sauvé #2, at para. 84) which takes into account the full factual context in which an alleged infringement occurs — one that eschews rigid and technical application and that "will vary depending on the circumstances" (Sauvé #2, at para. 80, citing R. v. Keegstra, 1990 CanLII 24 (SCC), [1990] 3 S.C.R. 697, at pp. 737-38; see also R. v. Sharpe, 2001 SCC 2, [2001] 1 S.C.R. 45, at para. 154, per L'Heureux-Dubé, Gonthier and Bastarache JJ., Thomson Newspapers Co. v. Canada (Attorney General), 1998 CanLII 829 (SCC), [1998] 1 S.C.R. 877, at para. 87, per Bastarache J., and Oakes, at p. 139). This makes good sense. Justification is an unavoidably contextual exercise.

[ 125 ] That made clear, the limitations analysis proceeds in two broad steps. First, we must identify the objective behind the impugned measure and determine whether it is sufficiently important to justify the limit on a Charter right. Second, we must assess whether the means that Parliament selected to pursue its objective are proportionate to the rights limitation — an inquiry that considers whether the measure is rationally connected to the objective; whether the measure minimally impairs a claimant's Charter rights; and whether there is proportionality between the effects of the measure and the objective identified at the first step (Oakes, at p. 139).

A. The Government's Objective

[ 126 ] Much of the argument before this Court focused on the government's objective at the time of the 1993 amendments to the Act, and whether the Attorney General has identified a sufficiently important purpose to justify limiting a Charter right. In our respectful view, the majority errs by focusing on the absence of a concrete problem or mischief. Proceeding in this way sidesteps the undeniable point that Parliament can constitutionally legislate in pursuit of, or in response to, considerations of political morality or philosophy. The limitations analysis must be flexible enough to accommodate concerns of this nature and to afford due respect to Parliament's policy-making expertise and to the full range of its law-making capacity. The moral nuance inherent in defining and defending the boundaries of rights — that is, in justifying rights limitations — is not a mechanical or purely empirical exercise that can be reduced to "technical questions of weight and balance" (G. C. N. Webber, The Negotiable Constitution: On the Limitation of Rights (2009), at p. 104). Rather, it requires premises and principles to be taken seriously (R. Pound, "Mechanical Jurisprudence" (1908), 8 Colum. L. Rev. 605, at p. 612).

[ 127 ] As we explain below, Parliament's objective in maintaining the limit on long-term non-resident voting was to privilege a relationship of some currency between electors and their communities. In our view, this objective is sufficiently pressing and substantial to survive scrutiny under s. 1.

(1) Identifying the Objective

[ 128 ] Identifying the objective of a rights-limiting measure enacted by a legislature raises the methodological difficulty that the objective — or the level of generality at which the objective should be framed — may not be immediately apparent, especially where a constitutional challenge relates to one component of a broader statutory scheme or legislative enactment (see, e.g., P. W. Hogg, Constitutional Law of Canada (5th ed. 2017), vol. 2, c. 38.9(a)). This difficulty is augmented where, as here, the impugned rights limitation is said to arise not by legislative action, but by its absence — in this case, by Parliament's failure to extend the right to vote by special ballot to long-term non-residents.

[ 129 ] Such challenges reinforce the importance of context as "the indispensable handmaiden to the proper characterization of the objective" — and, indeed, to all stages of the limitations analysis (Thomson Newspapers Co., at para. 87, per Bastarache J.). Here (and, we suggest, generally), two contextual elements — the impugned legislation itself (in this case, the 1993 amendments), and the state of the law prior to the impugned legislation — provide compelling evidence of the relevant legislative objective behind the failure to extend voting rights to all non-residents.

[ 130 ] First, the best way of discerning a legislature's purpose will usually be to look to the legislation itself (J. Sikkema, "The 'Basic Bedford Rule' and Substantive Review of Criminal Law Prohibitions Under Section 7 of the Charter", in D. B. M. Ross, ed., Assisted Death: Legal, Social and Ethical Issues after Carter (2018), 49, at pp. 51 and 71). Examining the scope of what a legislature sought to regulate — and relatedly, the scope and nature of any exceptions or exclusions that are provided for in a statute — will almost always give meaningful insight into the "corporate will of the legislature" as an entity (R. v. Heywood, 1994 CanLII 34 (SCC), [1994] 3 S.C.R. 761, at p. 788; see also R. Sullivan, Sullivan on the Construction of Statutes (6th ed. 2014), at paras. 9.56 and 9.57). In other words, the means that Parliament adopted will usually make plain the objective that Parliament sought to achieve. Further, an express statutory statement of purpose, where it exists, will generally be determinative, as it was available to and voted on by all members of the legislature, knowing that it would represent a corporate statement of legislative purpose.

[ 131 ] But historical context is also relevant. The state of the law as it existed prior to an impugned provision coming into force can, in light of the subsequent legislative provision under review, give insight into why the provision was enacted and what purpose it was designed to serve (R. Ekins, The Nature of Legislative Intent (2012), at pp. 258-59).

[ 132 ] Here, these two factors — the present and past state of the law — go hand in hand. The 1993 amendments brought about a significant expansion of the franchise to non-resident Canadians, but an expansion with limits. This is immediately evident from the broader statutory context of the impugned voting restriction. The 1993 amendments to the Act permit temporary non-residents to vote by special ballot, but require long-term non-residents to re-establish residence in Canada before voting in federal elections. The expansion of the franchise to temporary non-residents, coupled with the retention of the prior restriction for long-term non-residents, points to a clear and readily discernable purpose behind the legislative design as a whole: Parliament sought to privilege a relationship of some currency between electors and the communities in which they are eligible to vote. This objective can be reasonably inferred from the shift to permitting temporary non-residents to vote, while continuing to exclude long-term non-residents from voting; and, from the provision immediately restoring the right to vote for any long-term non-resident (in person, at an advance poll, or by special ballot) upon re-establishing residence in Canada. This purpose, moreover, is affirmed by the other exceptions to the long-term non-resident voting restriction in the Act — electors serving abroad in the Canadian Forces or in the public service, for example — being cases where, by virtue of the circumstances of their residence abroad, Canadian citizens necessarily maintain a connection to their communities in Canada.

[ 133 ] In this appeal, the parties have pointed to excerpts from Hansard which they say shed light on the purpose behind the 1993 amendments. Although such evidence may be relevant in limited circumstances (for example, where the minister's speech on second reading addresses the objective behind the specific impugned provision), reliance upon snippets of parliamentary debates as a means of identifying the legislative objective of a provision is, as a general proposition, a questionable practice. Statements by individual legislators may not (and often could not) represent the collective intent of the legislature as jointly expressed in the legislative act. As this Court has recognized, "the intent of particular members of Parliament is not the same as the intent of Parliament as a whole" (Heywood, at p. 788).

[ 134 ] Circumspection in relying upon Hansard is particularly important where members' statements, taken together, express competing interpretations or purposes. "Whose version, in one chamber of the bicameral Parliament, can be said to unlock any secrets of interpretation?" (Ruparel v. Canada (Minister of Employment and Immigration), 1990 CanLII 13102 (FC), [1990] 3 F.C. 615 (T.D.), at p. 625). Conflicting members' statements aside, the practice of relying on statements by individual members of a legislature — statements that are neither voted on nor necessarily agreed to by other legislators — is fundamentally at odds with the typical conception of a legislature as an institution which acts only through the corporate legislative expression of its membership on the proposed course of action set out in a bill. That corporate expression "does not reduce to the intention of any one or more individual members" (Ekins, The Nature of Legislative Intent, at p. 10). How, then, the converse could be true — that the expressed intentions of individual legislators should (or could) be aggregated or otherwise be taken to identify the legislature's corporate intention — is far from obvious. This is not to suggest that legislatures cannot respond to the views of individual members, but rather that the possibility of doing so is built into the give-and-take of the law-making process (which will often not be apparent in a public record such as Hansard) (R. Ekins, "Legislation as Reasoned Action", in G. Webber et al., Legislated Rights: Securing Human Rights through Legislation (2018), 86, at p. 111).

[ 135 ] And, even if this problem could be overcome, practical difficulties will persist, going to the reliability of such statements as indicators of legislative purpose. For example, statements by individual members may be partial, inchoate, tentative, or simply mistaken. Further, consistent judicial reliance on Hansard materials may encourage strategic behaviour on the part of legislators, thereby obscuring rather than illuminating any "true" underlying purpose of a legislative enactment. Legislators advocating in favour of a bill may not give sufficient attention to the limited means by which the bill seeks to achieve a stated objective, thereby creating the risk — where relied upon by courts — of supporting unduly broad interpretations of purpose. These concerns, among others, have been canvassed at length by a range of leading scholars (see, e.g., W. N. Eskridge, Jr., Dynamic Statutory Interpretation (1994), at pp. 222-23; C. R. Sunstein, "Interpreting Statutes in the Regulatory State" (1989), 103 Harv. L. Rev. 405, at p. 433; A. Scalia and B. A. Garner, Reading Law: The Interpretation of Legal Texts (2012), at pp. 369-90; J. H. Baker, "Case and Comment: Statutory Interpretation and Parliamentary Intention" (1993), 52 Cambridge L.J. 353, at pp. 356-57; J. Waldron, Law and Disagreement (2001), at pp. 142-46; and Ekins, The Nature of Legislative Intent).

[ 136 ] Our Court has also warned about the "inherent unreliability" and "indeterminate nature" of speeches and statements made by legislators as a means of discerning legislative purpose (Re B.C. Motor Vehicle Act, 1985 CanLII 81 (SCC), [1985] 2 S.C.R. 486, at pp. 508-9). While records of these statements are admissible, we have repeatedly emphasized that they will usually be of limited reliability and weight (see, e.g., Reference re Upper Churchill Water Rights Reversion Act, 1984 CanLII 17 (SCC), [1984] 1 S.C.R. 297, at p. 319; Rizzo & Rizzo Shoes Ltd. (Re), 1998 CanLII 837 (SCC), [1998] 1 S.C.R. 27, at para. 35; Placer Dome Canada Ltd. v. Ontario (Minister of Finance), 2006 SCC 20, [2006] 1 S.C.R. 715, at para. 39; Canadian National Railway Co. v. Canada (Attorney General), 2014 SCC 40, [2014] 2 S.C.R. 135, at paras. 44-47; Sullivan, at para. 23.88; see also R. v. Morgentaler, 1993 CanLII 74 (SCC), [1993] 3 S.C.R. 463, at p. 484).

[ 137 ] All this said, and just to put the matter to bed in this case, we note that the few haphazard Hansard materials put before us here tend to confirm the objective that we have identified in considering the 1993 amendments and the prior state of the law. For example, members of the Special Committee on Electoral Reform (whose work resulted in the bill put before Parliament) expressed a desire to ensure that electors retained a "close enough connection" to Canada (House of Commons, Minutes of Proceedings and Evidence of the Special Committee on Electoral Reform, No. 5, 3rd Sess., 34th Parl., November 24, 1992, at p. 5:61). The Member of Parliament for Halifax West stated that he was not in favour of non-residents voting "in an unrestricted way", and that they should therefore "have a connection with . . . their constituency before they should be allowed to vote in [that] constituency" (House of Commons, Minutes of Proceedings and Evidence of the Special Committee on Electoral Reform, No. 3, 3rd Sess., 34th Parl., March 25, 1992, at p. 3:30). These comments echo earlier discussions from 1984, when the Member of Parliament from Kenora-Rainy River cautioned that non-resident voting should not be permitted to the point of "interfer[ing] with the local constituency" (House of Commons, Minutes of Proceedings and Evidence of the Standing Committee on Privileges and Elections (1983 Statutory Report of the Chief Electoral Officer), No. 1, 2nd Sess., 32nd Parl., March 27, 1984, at p. 1:30). As these excerpts highlight, legislative discussions beginning in the 1980s, and continuing through to the 1993 amendment process, reflect an ongoing concern that voters retain a relationship of some currency between electors and their communities.

(2) Assessing the Objective

[ 138 ] Oakes requires us to consider whether the objective underlying the rights limitation — in this case, expanding the franchise while ensuring that electors maintain a relationship of some currency to their communities — is "pressing and substantial". We conclude that it is. In our analysis below, we first address the preliminary question of how courts should consider legislative objectives in the voting rights context, such as the one at issue here, that are philosophically grounded. Then, we turn to the specific objective we have identified and explain why it is pressing and substantial.

(a) Defining the Boundaries of the Right to Vote

[ 139 ] When assessing legislative objectives under a limitations analysis, courts must bear in mind that legislatures can pursue a range of different kinds of objectives. While some statutes are targeted at addressing discrete and specific problems, other statutes pursue broader philosophical goals without necessarily being targeted at an identifiable "problem" or "mischief". Just as Parliament can validly "legislate on the basis of some fundamental conception of morality" (R. v. Butler, 1992 CanLII 124 (SCC), [1992] 1 S.C.R. 452, at p. 493), so too can Parliament legislate in pursuit of normative conceptions of what the Canadian political community is, and how it can best be protected and made to flourish. In either case, evidence of "actual harm" is not required to demonstrate that an asserted objective is pressing and substantial (Harper v. Canada (Attorney General), 2004 SCC 33, [2004] 1 S.C.R. 827, at para. 93).

[ 140 ] Focussing on the problem or mischief therefore misses the point of Parliament having acted as it did in this case. Parliament's objective here, to preserve a relationship of some currency between electors and their communities, is clearly inspired by a particular moral philosophical understanding of the relationship between citizen and state in a democracy. It is also entirely consistent with "the need", previously recognized by this Court in Reference re Prov. Electoral Boundaries (Sask.), 1991 CanLII 61 (SCC), [1991] 2 S.C.R. 158, at p. 188, "to recognize cultural and group identity and to enhance the participation of individuals in the electoral process and society [which] requires that other concerns also be accommodated". The measure was, accordingly, not targeted at an empirically demonstrable "problem" or "mischief" but to shape the contours of a Charter right which, while drafted in absolute terms ("[e]very citizen of Canada has the right to vote"), is not absolute. Were it otherwise, s. 1 would be rendered meaningless in this case. In other words, Parliament was quite properly striving to shape the boundaries of the right by enacting legislation governing the terms on which elections are conducted — in this case, by drawing a line at citizens who have a current relationship to the community in which they seek to cast a ballot.

[ 141 ] The majority's approach leaves no room for a meaningful limitations analysis by reasoning that "the Charter tethers voting rights to citizenship, and citizenship alone" (para. 29). This Court did not go so far in Sauvé #2 as to state the s. 3 right in such absolutist terms; indeed, the very indication that the s. 3 right to vote "cannot lightly be cast aside" and is subject to a "stringent justification standard" confirms that this right can be justifiably limited (Sauvé #2, at para. 14). Further, the majority's point that s. 3 cannot be overridden by s. 33's notwithstanding clause should not, in our respectful view, be taken as conferring judicial license to strike down legislation on the basis of disagreement with Parliament's policy objectives in legislating on the exercise of the right to vote. Rather, it should encourage judicial restraint in considering those objectives, knowing that a finding of constitutional invalidity would absolutely foreclose Parliament from pursuing them. Three considerations support the view that Parliament is permitted, within limits, to shape the scope of voting rights under s. 1.

[ 142 ] First, while most Charter rights are negative in the sense that they preclude the state from acting in ways that would impair them, the right to vote is a positive entitlement. It is therefore, at least in part, a necessarily legislated right. Meaning, it is "given form and content in legislation" which specifies "the distinctions on which the general affirmation is silent" (Ekins, "Legislation as Reasoned Action", at pp. 102-3). Such legislation is never designed to solve a problem or address a particular mischief. Rather, it breathes life into the right so that it may be recognized and exercised. This would apply to the restriction at issue here, but also to mundane regulations governing polling locations, polling hours, and voter identification requirements, all of which determine, to some extent, the degree of access that Canadians have to the ballot box — or, stated more broadly, the effective scope of the s. 3 right. On this question, the majority also overstates our respective points of difference, when it states that limits to the right to vote must be raised and justified under s. 1 rather than incorporated into the scope of the right itself (paras. 31 and 42). Of course limits to the right to vote must be justified; nobody disputes that. Where we differ from the majority is in our view that limits to the right to vote can be justified, because some specification of the right to vote (whether to account for age, or the currency of relationship between electors and the communities they represent) is necessary. The majority's absolutist approach to the limitations analysis takes the opposing view: that "citizenship", being the only qualification mentioned in s. 3, is the only requirement, and that all other specifications are necessarily unconstitutional (paras. 29 and 35). In other words, the majority effectively treats s. 1 as a dead letter where s. 3 is concerned.

[ 143 ] Secondly, the Act contains a range of restrictions on voting which, while not challenged here are no less the product of principled and unavoidably philosophical reasoning about the democratic franchise than is the long-term non-resident restriction. In particular, the Act withholds the vote from Canadian citizens who have never before lived in Canada. Again, Parliament, in shaping the boundaries of the s. 3 right, took the view that a person who was not born in Canada, has never lived in Canada, and may never intend to live in Canada does not have an unconditional right to vote. But of course, the logical implication of the majority's argument that "the Charter tethers voting rights to citizenship, and citizenship alone" (para. 29) is that this restriction is unjustifiable and therefore unconstitutional — meaning that Parliament cannot prohibit a citizen who has never set foot in Canada, and never will, from voting in Canadian elections (transcript, at pp. 6 and 25). In our view, such a state of affairs, which would represent a radical judicial expansion of voting rights of a sort that Parliament has clearly deemed undesirable, inevitably follows from the majority's decision on this appeal.

[ 144 ] Another voting rights restriction in the Act prevents minors from voting. Again, such a restriction is irreconcilable with an absolutist approach to s. 3's guarantee that "[e]very citizen . . . has the right to vote". We appreciate that this Court, in Sauvé #2, sought to distinguish the age restriction as "regulating a modality of the universal franchise" (para. 37 (emphasis added)), thereby casting it as a categorically different restriction from other limits to the s. 3 right. Parliament, we are told in Sauvé #2, "is making a decision based on the experiential situation of all citizens when they are young" (para. 37). But this rationale would presumably apply to support any line-drawing exercise. For example, on this logic, Parliament, in "making a decision based on the experiential situation of all citizens when they are young", could permissibly set the voting age at 75 without running afoul of s. 3. More to the point of this appeal, however, the distinction between a limit on the s. 3 right and a mere restriction on the modality of voting is not evident. Why is an age requirement a permissible "modality" of regulating the right to vote, but a residence requirement is not? And more particularly, if (as the appellants suggest) the distinguishing feature of the age requirement is that "it's just a timing [issue] as to when [a citizen can vote]" (transcript, at p. 21), why would that same rationale not also apply to a residence requirement, since a long-term non-resident is permitted to vote if and when they re-establish residence in Canada — a matter that is, unlike age, entirely within his or her control?

[ 145 ] The only coherent account for the Act's restriction on the right to vote for Canadian citizens under the age of 18 — especially in light of the appellants' contention that the s. 3 right necessarily applies to all citizens — is that such a limit to the s. 3 right is justified. And as with the impugned measure here, that justification necessarily reflects a certain normative or philosophical theory — one that trades off some considerations (ensuring an informed electorate that is capable of exercising rational and independent choice) against others (extending the scope of the franchise to all citizens) in a particular manner. Properly understood, the uncontested capacity of lawmakers to prohibit minors from voting — whether cast as a limit, or as a regulation of a modality of the right to vote — supports the notion that Parliament has a critical role to play in shaping the boundaries of the s. 3 right, and that it necessarily exercises that role in pursuit of certain philosophical, moral, or otherwise normative objectives. Further, the age restriction on voting plainly belies the notion that a lack of Canadian citizenship is the only constitutionally permissible basis upon which Parliament may deny someone the right to vote.

[ 146 ] The third consideration supporting the view that the 1993 amendments represented Parliament's legitimate attempt to shape the boundaries of the s. 3 right emerges from this Court's analysis in Sauvé #2. Writing for the majority, Chief Justice McLachlin, in striking down a former provision in the Act disenfranchising "[e]very person who is imprisoned in a correctional institution serving a sentence of two years or more" (Canada Elections Act, R.S.C. 1985, c. E-2, s. 51(e)), invoked a purely deontological argument — namely, that the power held by lawmakers, and hence the legitimacy or force of the law they make, is rooted in a delegation from voters; and that the moral obligation to obey the law flows from having a voice in making it (para. 31). Respectfully, we see the merits of this moral philosophical argument quite differently. Indeed, this rationale for legislative action based on the "consent of the governed" is unsourced and unelaborated, and has been quite rightly rejected in favour of one conceiving legislative authority as "turn[ing] on the capacity to secure the common good", and its object as the common good itself (Ekins, "Legislation as Reasoned Action", at p. 92). This sees, as we see, the exercise of legislative authority as "a moral choice made in response to reasons: a choice about the propositions that ought to frame our common life" (p. 99). But that disagreement aside, the very fact that the majority in Sauvé #2 invoked a deontological argument to buttress its position as to the scope of the s. 3 right demonstrates the unavoidability of moral philosophical considerations in defining the boundaries of that right. And, where such considerations drive legislative action, and where as a result the comparative expertise which courts have in legal reasoning no longer applies, courts ought to be particularly circumspect in second-guessing Parliament's assessment of those boundaries.

[ 147 ] Indeed, the majority's approach here is itself no less rooted in a philosophical or theoretical understanding about the right to vote than are the impugned restrictions. The majority, while purporting to dismiss philosophically based perspectives on the relationship between elector and state as unworthy of consideration as such, in substance simply prefers its own philosophically based perspective to Parliament's. For instance, when the majority speaks of a "fundamentally inclusive view of voting rights in Canada" (para. 51), it is referring to just that — a view of the s. 3 right to vote and the implications that follow. Whether, therefore, one takes a broad or a narrow view of s. 3, though, Parliament — no less than the majority — must unavoidably shape the right based on philosophical considerations. The appropriate inquiry in limitations analysis is, therefore, not whether Parliament's legislative objective rests on theoretical considerations, but whether the objective that it pursues is pressing and substantial.

[ 148 ] We therefore maintain that the limitations analysis must be flexible enough to account for Parliament's ability to legislate in pursuit of philosophical, moral or otherwise normative considerations. Our fundamental objection to the majority's limitations analysis is that it is unjustifiably absolutist: it necessitates that all citizens be permitted to vote at all times and in all contexts. The majority's analysis is incongruent with current practices (especially with regard to minors, and to citizens who have never lived in Canada and never intend to) and leaves no role for Parliament to craft reasonable limits in pursuit of compelling objectives. And, just as incongruously, while the majority readily cites and defers to the findings of parliamentary policy studies (at para. 64), it refuses to afford any such respect for parliamentary policy decisions.

[ 149 ] In our view, s. 1's direction that no right is absolute should not be treated as a dead letter when considering claims under s. 3. In what follows, we apply what we view as the proper approach to analysing the proper limits to the s. 3 right — one that affords proper appropriate respect for Parliament's legislative judgments in pursuit of salutary, but philosophically grounded, objectives.

(b) Preserving a Relationship of Currency Between Electors and Their Communities

[ 150 ] In Opitz v. Wrzesnewskyj, 2012 SCC 55, [2012] 3 S.C.R. 76, at para. 32, this Court spoke of residence as a "fundamental requirement" to the entitlement to vote, along with age and citizenship. The majority observes that the Court in Opitz discussed residence in the context of s. 6 of the Act, not s. 3 of the Charter. That is so. But significantly, the Court qualified the criterion of residence as "fundamental" where it did not have to do so, since the Act itself does not use such language. On the assumption that the Court does not speak superfluously, we take this as a statement defining residence as a fundamental requirement of the right to vote. While, therefore, citizenship is a necessary requirement to vote, this does not mean that it is, as the majority suggests (at para. 29) the only constitutionally permissible limit. Citizenship is a status; it appertains to those qualified under the Citizenship Act, R.S.C. 1985, c. C-29, s. 3, including persons born to a Canadian citizen outside Canada. It does not itself indicate a relationship of any currency to a particular Canadian community. Parliament, not unreasonably, deemed residence or recent residence to be indicative of this relationship — hence the five-year rule. The fact that the Act includes certain exceptions to the five-year residence rule (for instance, for public servants stationed abroad) supports, rather than undercuts, the notion that a relationship of currency is essential. Each of the categories of exceptions prove the rule, since they cover persons who, by virtue of their circumstances abroad, necessarily maintain a connection to their communities in Canada.

[ 151 ] As we have described, Parliament's limit on voting for long-term non-residents was intended to preserve a relationship of some currency between electors and their communities. With great respect, and even though it acknowledges the term is merely shorthand, the majority in its analysis places exaggerated significance on the notion, advanced by the Attorney General and the Court of Appeal, of preserving the "social contract" (paras. 47-54). Social contract theory is neither uniformly defined nor static: while the majority refers to Jean-Jacques Rousseau (at para. 48), many other philosophers have developed various theories of the social contract through the centuries. Contrary to what the majority says, the Attorney General did not change the substance of her argument regarding the legislative objective before this Court (at para. 47). She merely used the concept of the social contract as a label, which — while perhaps unhelpful and imprecise — is not inconsistent with the legislative objective that can be clearly discerned from the Act. In our respectful view, preserving a relationship of currency between electors and their communities is a pressing and substantial objective addressing two key concerns.

[ 152 ] First, preserving the relationship between electors and their communities by limiting long-term non-resident voting ensures reciprocity between exercising the right to vote and bearing the burden of Canadian laws. Chief Justice McLachlin acknowledged the constitutional significance of this notion of reciprocity with powerful language in Sauvé #2 (at para. 31):

In a democracy such as ours, the power of lawmakers flows from the voting citizens, and lawmakers act as the citizens' proxies. This delegation from voters to legislators gives the law its legitimacy or force. Correlatively, the obligation to obey the law flows from the fact that the law is made by and on behalf of the citizens. In sum, the legitimacy of the law and the obligation to obey the law flow directly from the right of every citizen to vote. As a practical matter, we require all within our country's boundaries to obey its laws, whether or not they vote. But this does not negate the vital symbolic, theoretical and practical connection between having a voice in making the law and being obliged to obey it. This connection, inherited from social contract theory and enshrined in the Charter, stands at the heart of our system of constitutional democracy.

[ 153 ] In that case, the majority invoked the reciprocity argument to justify prisoners' right to vote on the basis that prisoners, who are deprived of liberty by the state, are exceptionally affected by Canadian laws. In the present appeal, however, the reciprocity principle, which Sauvé #2 determined to be "enshrined in the Charter [and] stand[ing] at the heart of our system of constitutional democracy", justifies limiting non-resident voting precisely because long-term non-residents are not generally subject to Canadian laws. For this reason, the Act enables long-term non-residents to exercise their right to vote immediately upon re-establishing residence in Canada: since they have rejoined the community, they are again entitled to elect its lawmakers. Preserving the reciprocity between electing lawmakers and bearing the burden of obedience to the law is necessary to ensure fairness to resident Canadians. Indeed, it is unfair to Canadian residents for their lawmakers to be elected by long-term non-residents who have no connection of any currency to their electoral district (J. P. Boyer, Election Law in Canada: The Law and Procedure of Federal, Provincial and Territorial Elections (1987), vol. 1, at p. 393).

[ 154 ] Second, preserving the relationship between electors and their communities through limits on long-term non-resident voting protects the integrity of the Canadian electoral system, which is founded on geographical representation. In this sense, it serves "the broader concept of effective representation which best serves the interests of a free and democratic society" (Ref. re Prov. Electoral Boundaries (Sask.), at p. 189). The Constitution Act, 1867, establishes Canada as a parliamentary democracy with a "Constitution similar in Principle to that of the United Kingdom" having, accordingly, an electoral system resting on geographically-defined electoral districts for the House of Commons, the Senate and the provincial legislatures (Constitution Act, 1867, preamble, ss. 21, 22, 37 to 40, 52 and 69 to 72). In Canada, as in the United Kingdom and other Westminster democracies, the origins of the voting system are rooted in "the notion of territorial representation" (V. Bogdanor, "Introduction", in V. Bogdanor and D. Butler, eds., Democracy and Elections: Electoral Systems and Their Political Consequences (1983), at p. 2). Before us, the appellants acknowledged that "residence was a defining feature" of Canada's democratic system (transcript, at pp. 19-20).

[ 155 ] Under our Westminster parliamentary system, then, Canadian electors vote for a Member of Parliament (specifically, a member of the House of Commons), who serves as the representative of the electorate in a geographically defined community (Haig v. Canada (Chief Electoral Officer), 1993 CanLII 58 (SCC), [1993] 2 S.C.R. 995, at p. 1031). At Confederation, this regional nature of the composition of the House of Commons was constitutionally entrenched (Constitution Act, 1867, ss. 37 and 40). (Although s. 40 is now considered spent, Parliament continues to provide for electoral districts under the Electoral Boundaries Readjustment Act, R.S.C. 1985, c. E-3; see A. Dodek, The Canadian Constitution (2nd ed. 2016), at p. 54.) While Parliament, as the aggregate of its members, represents "all constituencies, all of the territory, all parties, all interests, all citizens, all inhabitants" (J. Ajzenstat, The Canadian Founding: John Locke and Parliament (2007), at p. 61), the s. 3 voting right is premised upon electors voting for a representative of their community. This regional structure must therefore inform any consideration of the electoral system, and Canadians' participation therein.

[ 156 ] Indeed, the regionally based composition of the House of Commons features prominently in our jurisprudence on the redrawing of boundaries, which permits "geography, demography and communities of interest" to be considered in determining constituency boundaries, even where such considerations may result in otherwise inequitable voting power disparities (Reference re Prov. Electoral Boundaries (Sask.), at p. 172; see also R. MacGregor Dawson, The Government of Canada (4th ed. 1963), at p. 341). Likewise, the notion of "ordinarily resident" is central to the Act, as it determines the polling division in which an elector exercises his or her right to vote (ss. 3 and 6, and Opitz, at para. 11). It is defined as "the place that has always been, or that has been adopted as, his or her dwelling place, and to which the person intends to return when away from it" (s. 8). Limiting long-term non-resident voting thus ensures that the electors residing in a particular constituency, who share a community of interests that is typically derived from (at least in part) geographical proximity, retain the power to decide for themselves who would best advance those shared interests on their behalf in the House of Commons. It follows that the appellants' and the majority's (at para. 69) insistence that connections to Canada can be maintained via, for example, online news websites, misses the point of the 1993 amendments: connections to particular communities are not so easily maintained.

[ 157 ] The majority itself equivocates as to the role that residence plays, or should play, in Canada's voting rights regime. While recognizing that residence is an "important device", and that it "underpins our geographically determined, constituency-based system of electoral representation", it then minimizes residence as merely an "instrumental necessity," and not an "essential requirement" of the right to vote (paras. 28 and 31-32). With respect, the distinction between an instrumental necessity and an essential quality — a distinction that underpins the majority's entire analysis — eludes us. It also flies in the face of this Court's acknowledgement in Opitz that residence is a fundamental requirement of the right to vote, not simply an organizing mechanism — an observation that applies equally in the Charter context (para. 32). And, in any event, the majority offers no reason why residence should not be understood as an essential requirement — a judgement that is clearly at odds with that of Parliament and this Court's jurisprudence and one which finds no justification in the majority's reasons.

[ 158 ] In summary, we would conclude that Parliament's objective is pressing and substantial. We turn now to the question of proportionality.

B. Proportionality

[ 159 ] The second question in the limitations analysis asks whether the voting restriction is proportionate. In answering this question, we must exercise caution to avoid usurping Parliament's policy-making function. The inquiry is not what members of the Court prefer, but whether the limit was one that Parliament could reasonably impose: "The courts are not called upon to substitute judicial opinions for legislative ones as to the place at which to draw a precise line" (R. v. Edwards Books and Art Ltd., 1986 CanLII 12 (SCC), [1986] 2 S.C.R. 713, at p. 782; see also Stoffman v. Vancouver General Hospital, 1990 CanLII 62 (SCC), [1990] 3 S.C.R. 483, at p. 531, per La Forest J.; Irwin Toy Ltd. v. Quebec (Attorney General), 1989 CanLII 87 (SCC), [1989] 1 S.C.R. 927, at p. 993, per Dickson C.J. and Lamer and Wilson JJ.). This is particularly so when we are confronted with a challenge to Canada's election laws, in respect of which body of laws we have previously held that "a natural attitude of deference" is required (R. v. Bryan, 2007 SCC 12, [2007] 1 S.C.R. 527, at para. 9; see also Harper, at para. 87).

[ 160 ] Accepting as we do that Parliament's objective — privileging a relationship of currency between electors and their communities — is pressing and substantial, we have little difficulty in concluding that the means by which it chose to pursue that objective — a five-year non-residence cut-off — are proportionate. The Attorney General has satisfied her burden of showing that five consecutive years (as opposed to one year, ten years, or some other number) was a reasonable — and therefore constitutionally permissible — demarcation.

[ 161 ] First, five years corresponds to the maximum length of a Parliament, thereby ensuring that all non-residents can vote in at least one election after leaving Canada. Five years is also sufficiently long to encompass the vast majority of post-secondary educational programs (from undergraduate degrees to PhDs), permitting students who travel abroad to study to resume residence in Canada after completing their programs without foregoing the ability to vote. Inasmuch as the limit is designed to reasonably distinguish short-term from long-term non-residence, it cannot be said that the figure of five years does not reasonably succeed in doing so.

[ 162 ] The majority contends that the Act's five-year limit on non-resident voting is arbitrary (para. 4). In response, we simply observe that, as the appellants conceded in their submissions (transcript, at p. 25), the majority's absolutist stance on s. 3 tethering the right to vote to citizenship alone would render any time limit arbitrary — and not only in the context of federal voting rights. Although the majority suggests that provincial residence requirements are distinguishable since there is no provincial equivalent to citizenship (at para. 61), s. 3 refers not only to citizens' right to vote in an election of members of the House of Commons, but also of a legislative assembly. The majority's suggestion, therefore, that citizenship + residence can supply the requisite connection between the voter and the province, but that citizenship alone is sufficient to show that connection in the context of federal elections, is to read a distinction into the wording of s. 3 that simply does not exist.

[ 163 ] Moreover, the majority's tenuous effort to validate provincial residence requirements is cast into doubt by everything else the majority says about s. 3 in its reasons. After straining to explain how residents abroad "remain connected to and informed of current affairs" (para. 69), how legislation does not impact resident citizens equally (at para. 71), and how citizens abroad become "ambassadors" of their home jurisdiction (at para. 80), why would those arguments not apply with equal force to, for example, a Manitoban who moves to British Columbia? Unfortunately for provincial election laws, the majority cannot have it both ways: its absolutist approach to s. 3 leaves no room for justifying residence-based limits on the s. 3 right, irrespective of whether such limit is imposed by Parliament or by a provincial legislature.

[ 164 ] Secondly, five years falls well within the range of limits adopted by other Westminster democracies in which election regimes are, as here, structured around geographically defined electoral districts. It is worth observing that inheritors of Westminster-style democracy tend to have more restrictive voting regulations with respect to non-resident voting (A. Blais et al., "Deciding Who Has the Right to Vote: A Comparative Analysis of Election Laws" (2001), Electoral Studies 20). Indeed, when elections are structured as a choice of representative for specific regionally defined communities rather than national candidates running at large, the justification for limiting voting rights to individuals who have a connection to those particular local communities is stronger. For example, New Zealand permits citizens abroad to vote for three years after leaving the country. Australia sets the limit at six years. And in the United Kingdom, the limit is 15 years.

[ 165 ] To all this, the majority answers by dismissing all non-resident comparative examples. "[T]he history of the Canadian electoral system is one of progressive enfranchisement", the majority proclaims (para. 62). "Canada is an international leader in this respect" and "[w]e must not defer to the standards of comparable systems in other countries" (para. 62). And even if they were apt comparators, the majority adds, Australia and New Zealand's respective six-year and three-year limits are implemented differently (para. 74).

[ 166 ] We first respectfully observe that, in rejecting other Westminster parliamentary comparisons, the majority's patriotism risks descending into exceptionalism. At different times in its history, Canada would well have taken some lessons from other countries. For example, New Zealand, granted the franchise to its indigenous people, the Māori, in 1867, nearly a century before Canada did so (N. Atkinson, "Voting rights", in Te Ara — The Encyclopedia of New Zealand (2015) (online)). Women in New Zealand were enfranchised in 1893, nearly a quarter century before Canada extended a limited franchise to women (ibid.). Why, given this history, the majority would simply dismiss the choices of New Zealand legislators as unworthy of any consideration by this Court, mystifies us. Granted, the majority alternatively points to small differences between how New Zealand and Australia implement their limits (majority reasons, at para. 74), but in doing so the majority is merely seizing upon the minutiae of policy preferences of Canadian legislators (and, for that matter, legislators in Australia and New Zealand) on matters about which this Court as an institution has little comparative expertise.

[ 167 ] The point is this simple. Where other internationally respected liberal democracies sharing our Westminster parliamentary framework have adopted comparable time limits on voting for long-term non-residents, such examples provide compelling evidence to satisfy the low bar of rationality. To the extent that the majority suggests otherwise on the basis that Canada is a purported "international leader" in respect of voting rights and should therefore "not defer" to other countries (para. 62), it deploys highly political, rhetorical arguments — arguments that, we observe, stand in tension with the majority's own invocation of internationalism and of a "globalized" world of connectivity and communication (para. 34). In any event, Parliament would have been perfectly entitled to consider such arguments in crafting Canada's laws in the first instance. But with respect, these are not legal arguments justifying this Court's suggestion that the impugned legislation is irrational.

[ 168 ] Further, any adult Canadian citizen can still exercise the right to vote at any point, provided that he or she re-establishes residence in Canada. Thus, the restriction at issue is not a permanent denial of the right to vote. Just like the age requirement, it represents a distinction "based on the experiential situation of all citizens" in that category (Sauvé #2, at para. 37); it is not a distinction based on moral worth (which was the Court's concern in Sauvé #2). The deleterious effects of the limit are therefore of less significance, and are outweighed by the other salutary effects that we have discussed — namely, ensuring reciprocity between exercising the right to vote and bearing the burden of Canadian laws, and protecting the integrity of Canada's electoral system.

[ 169 ] The majority discounts the salutary effects as speculative and illusory (at para. 78) — overlooking, among other things, the historical pedigree of residence in our jurisprudence (see Opitz, at para. 32) and the importance of "effective representation," which we have found lies "at the heart of the right to vote" (Reference re Prov. Electoral Boundaries (Sask.), at p. 188), and which is clearly undermined if Parliament cannot legislate in furtherance of a geographically based electoral system that is responsive to the needs of local constituencies.

[ 170 ] Further, it is, in our respectful view, unfortunate that the majority adopts the hyperbolic submission of the appellants that Canadians living abroad are often "ambassadors of Canadian values" (para. 80). While "Canadian values" (whatever those are) may hold appeal, their invocation here is a poor substitute for legal reasoning explaining precisely why the impugned legislation does not represent a reasonable limit to the s. 3 right to vote. Further, and even on its substance, this statement exaggerates non-resident Canadians' status, and unreasonably assumes their motives for leaving. Canadian citizens may leave Canada for all sorts of "non-ambassadorial" reasons, ranging from better career prospects, to lower taxes, to a preference for the "values" of other countries.

[ 171 ] Finally, the majority also asserts, in conclusory terms, that the denial of a right in and of itself inflicts harm (para. 82). But if this is true here, it would be true in every case where a right is limited. As we have already explained, no right is absolute. Were it otherwise, s. 1 would serve no purpose. To presume some quantum of harm from the very fact that a right is limited is, once again, to take an impermissibly absolutist approach to Charter rights, and in effect to read s. 1 out of the Charter entirely. In any event, this approach turns the majority's limitations analysis into something of a one-sided affair: the putative harm which the majority sees arising from the very fact that a right is limited is exactly the type of speculative and illusory impact that the majority disavows in considering the salutary effects of the limit.

IV. Conclusion

[ 172 ] In light of these considerations, we conclude that the Attorney General has demonstrated that the non-resident voting restriction represents a reasonable limit on the s. 3 right, demonstrably justifiable in a free and democratic society. The measure is rationally connected to the objective of preserving a relationship of currency between electors and their communities because it logically distinguishes short-term from long-term non-residents. The limit is minimally impairing because, on balance, a five-year time period falls within the range of reasonable options that were open to Parliament, and it is not this Court's prerogative, let alone within this Court's expertise, to second-guess the precise location at which Parliament chose to draw the line. And, in the final balancing, the salutary effects of preserving the integrity of Canada's geographically based electoral system and upholding a democratically enacted conception of the scope of the right to vote in Canada are significant. The deleterious effect of denying some citizens the right to vote is not insubstantial, but it is tempered by the fact that the restriction is reversible rather than permanent — any Canadian can immediately retain the right to vote upon re-establishing residence in Canada.

[ 173 ] We would dismiss the appeal.

Appeal allowed with costs throughout, Côté and Brown JJ. dissenting.

Solicitors for the appellants: Cavalluzzo, Toronto.

Solicitor for the respondent: Attorney General of Canada, Toronto.

Solicitors for the intervener the Attorney General of Quebec: Attorney General of Quebec, Québec.

Solicitors for the intervener the Canadian American Bar Association: Blake, Cassels & Graydon, Toronto.

Solicitors for the intervener the Canadian Expat Association: Osler, Hoskin & Harcourt, Calgary.

Solicitors for the intervener the David Asper Centre for Constitutional Rights: Goldblatt Partners, Toronto; University of Toronto Faculty of Law, Toronto.

Solicitors for the intervener the Canadian Civil Liberties Association: Lerners, Toronto.

Solicitor for the intervener the Chinese and Southeast Asian Legal Clinic: Chinese and Southeast Asian Legal Clinic, Toronto.

Solicitors for the intervener the British Columbia Civil Liberties Association: Stockwoods, Toronto; Michael Sobkin, Ottawa.

Supreme Court of Canada

Headnote

Cases Cited

Statutes and Regulations Cited

Authors Cited

The Chief Justice —

I. Overview

II. Background

III. Judicial History

A. Ontario Superior Court of Justice, [2014 ONSC 907](https://www.canlii.org/en/on/onsc/doc/2014/2014onsc907/2014onsc907.html), 119 O.R. (3d) 662

B. Court of Appeal for Ontario (Strathy C.J.O., Brown J.A. Concurring; Laskin J.A. Dissenting), [2015 ONCA 536](https://www.minicounsel.ca/oca/2015/536), 126 O.R. (3d) 321

IV. Issue

V. Analysis

A. Right to Vote

B. Role of Residence in the Canadian Electoral System

C. Justification Analysis

(1) Pressing and Substantial Objective

(2) Rational Connection

(3) Minimal Impairment

(4) Balancing the Salutary and Deleterious Effects

VI. Conclusion

Rowe J. —

A. The Significance of Residence and the Right to Vote

B. Section 1 Analysis

(1) Rational Connection

(2) Proportionality of Effects

C. Conclusion

Côté and Brown JJ. (dissenting) —

I. Overview

II. Background

III. Analysis

A. The Government's Objective

(1) Identifying the Objective

(2) Assessing the Objective

(a) Defining the Boundaries of the Right to Vote

(b) Preserving a Relationship of Currency Between Electors and Their Communities

B. Proportionality

IV. Conclusion

Cited By (12 cases)

R. v. Kim, 2025 ONCA 478

Drover v. Canada (Attorney General), 2025 ONCA 468

Hillier v. Ontario, 2025 ONCA 259

Ontario (Attorney General) v. Working Families Coalition (Canada) Inc., 2025 SCC 5

R. v. Brar, 2024 ONCA 254

Fair Voting BC and Springtide Collective for Democracy Society v. Attorney General of Canada, 2023 ONSC 6516

Working Families Coalition (Canada) Inc. v. Ontario (Attorney General), 2023 ONCA 139

Ontario (Attorney General) v. G, 2020 SCC 38, [2020] 3 S.C.R. 629

CCLA v. Attorney General of Ontario, 2020 ONSC 4838

R. v. A.C., 2019 ONSC 4270