Frank v. Canada (Attorney General): Renewing Voting Rights

In its first major decision of 2019, the Supreme Court of Canada has held that Canadian citizens who have been residing outside Canada for at least five consecutive years are entitled to vote in federal elections, contrary to provisions in the Canada Election Act. Although the prohibition was first enacted in 1993, until 2007 Canadians living abroad could reestablish the connection considered necessary to vote — and begin to recount their five years — by returning from time to time; however, as of 2007, Canadian had to reestablish residency in Canada before they could vote. (Exceptions were members of the armed forces and teachers and administrative staff for an armed forces school, persons in the federal or provincial public service posted abroad or working for an international organization to which Canada contributes or persons living with someone in these categories.)

A challenge to the provisions was successful in the Ontario Superior Court; however, that decision was overturned by the Court of Appeal. The Supreme Court of Canada held the provisions to be unconstitutional. The Frank decision is worth noting not only because of its substantive reasons, however, but also because the majority and the dissent based their reasoning, according to the dissent, on different approaches to Charter interpretation.

The reasons Canadians go abroad for extensive lengths of time are varied, but the two who challenged the Canada Elections Act both fit a particular pattern in one respect: they have family in Canada, return for visits and have not been able to obtain employment in Canada, although they would like to work here. They also have different connections to the U.S. One has become an American citizen; the other has not. One has lived there for about 13 years, primarily to go to school, while the other moved in grade 10 and has spent most of his adult life there. In the end, the degree of connection did not matter.

At the Court of Appeal, Strathy C.J.O. focused on the role of voting in reflecting local interests, effectively adopting the government’s position, which it did not raise in the initial application:

…[T]he residency requirement fulfills the pressing and substantial objective of preserving the social contract at the heart of Canada’s system of constitutional democracy. It ensures citizens are both subjectively connected to Canada through their knowledge and affiliation and objectively connected through holding citizenship responsibilities and duties to obey domestic laws. The connection between having a voice in making the laws and being obliged to obey them is what gives the laws legitimacy.

This emphasis on the social contract reflects the relationship between the local community determined by ridings and the everyday relationship between Canadians and their obligations to obey the laws. The majority at the Supreme Court, however, rejected the social contract as a “pressing and substantial objective” under section 1, and emphasized the global nature of Canada’s place in the world ( “[t]he ability of citizens not only to move, but to remain connected and maintain communications in so doing, is unprecedented.”). The Attorney General argued that residence was a crucial requirement in the right to vote, despite the clear wording of section 3 of the Charter (“Every citizen of Canada has the right to vote in an election of the members of the House of Commons or of a legislative assembly and to be qualified for membership therein.”) The Chief Justice, writing for three other justices, rejected the significance of residency, stating that it is not an internal limitation on section 3, although it may be raised under section 1: “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.” (Rowe J.A., while agreeing in the result maintained that residence might be relevant in other situations.)

While it may be more easily addressed under a provision such as section 3, which is clearly written, requiring residence to be addressed under section 1 respects the Charter‘s architecture, too often ignored, which requires that once a plaintiff/complainant has established a breach of a Charter guarantee, the inquiry should shift to section 1. In some cases, internal limits have been read into the guarantee, making it easier for government to justify the infringement (see, for example, Withler v. Canada (Attorney General) and my brief comment, in the context of a broader discussion of SCC equality jurisprudence at p. 18 of “Supreme Court of Canada Equality Jurisprudence and ‘Everyday Life’“.) and despite the argument by the Attorney General, the majority in Frank rejected this approach. Some limitations will be easily addressed under section 1 and may seem so obvious as to constitute an internal limit on the right (the dissent in this case, for example, refer to three year olds voting), but the clearer approach is to put the onus clearly on the (usually) government under section 1. (An obvious exception are those rights that have an internal limitation, such as section 8’s unreasonable search and seizure or section 7’s limitation on the right to life, liberty and security, but in these cases, the internal limitations should be read narrowly before shifting to section 1, not always followed.)

The majority did accept that fairness of the electoral system constitutes a pressing and substantial objective, but, it appears, not a rational connection between the objective and the provisions; the more significant finding is that the provisions did not constitute a minimal impairment. There is no reason for selecting the five year limitation, which has no relationship to the connection people have to Canada (people living abroad may have close connections, people living in Canada may actually have little) and which is not indicative of the impact of Canadian laws on Canadians, whether living elsewhere or not. Without commenting on the substance of this part of the Chief Justice’s analysis, I tend to agree that the deleterious effects of the infringement outweigh any benefits of the impugned provisions.

The two dissenting justices introduce a different approach to interpreting section 1, one that the Chief Justice rejects more or less out of hand (probably correctly). On the merits, they find the limit on the right to vote reasonable because there do need to be limits: no one suggests three year olds should be able to vote, but the majority decision means that Canadians who have never lived in Canada should be able to vote in federal elections. They conclude, “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”.

The dissent does undertake an analysis of the application of the Charter to the provisions; however, what stands out in their opinion, is the distinction they make between “limit” and “infringement”. They ask, “what is it that is being justified — an infringement, or a limit?”. Infringements can never be justified, limits can be. Thus “[a]n ‘infringement’ is a limit that is not justified”, reflecting the wording of section 1 (“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.”) Therefore, “The issue presented by this appeal,…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 dissent notes that counsel for the Attorney General added to the confusion by conceding an “infringement”; if this were taken properly in their view, it would render the appeal moot. Instead, they refer to the Attorney General’s statement in 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.

In their substantive analysis, the dissent dispute the Chief Justice’s broad statements about the extent to which Canadians living abroad are affected by Canadian laws, which is in its generality a weakness of his analysis. They
link the five years to resulting in allowing Canadians to vote in at least one election before they lose, at least temporarily, the right to vote and to covering most educational programs. Unlike the Chief Justice, who dismisses the experiences offered by other comparable countries, the dissent find them influential. They are also critical of the political and philosophical approach they believe the Chief Justice takes, including generalized statements about why Canadians leave Canada.

The difference between the majority and the dissent in Frank does not rest on whether provisions negatively affecting a Charter guarantee are limitations or infringements (although here, it is hard to say that the dissent is not correct, as the Chief Justice relates, the terms have been used interchangeably without serious harm, hardly an appealing defence). Rather, it rests on fundamental differences about the kinds of factors we bring to Charter analysis. They represent the debate about how we perceive our legal system and what it represents and reflects: and yes, that is a philosophical debate about the values we bring to governing Canada.


  1. I was disappointed by several aspects of Wagner’s reasoning (although the first complain applies equally to all reasons).

    First, before hopping into the section 1 analysis, surely some consideration needs to be given to what the “right to vote” in section 3 actually means. Canada’s democratic system (like Britains) has also required a local connection to the riding in which you propose to vote. It is not a “right to vote” in the abstract, it’s a right to vote under the parliamentary system reflected in our Constitution, which provides for representation based on geographical region. Does that include a right to vote in a riding in which you do not have some meaningful connection? That’s not obvious – Canadian resident citizens, after all, don’t get to vote in ridings in which they don’t reside. The the form of our parliamentary democracy, to extend a free-standing right to vote independent of local connections under one part of the constitution does considerable violence to another part of the constitution – they need to be interpreted harmoniously with one another.

    Second, I found Wagner CJ’s reasoning on section 1 to be superficial and unpursuasive. To dismissive the beneficial effects of the existing law as speculative is to simply avoid the issue – the beneficial effects of a law which prohibits certain activities will always be speculative if it is generally observed, since we cannot see the adverse consequence of conduct that people don’t engage in.

    Third, to suggests that the adverse effects are implausible betrays a lack of imagination on the part of the Majority. For example, non-resident citizens may now be able to engage in riding shopping by choosing to vote in close or competitive ridings with which they have some nominal connection. Given that MP elections are frequently decided by a small number of votes, it wouldn’t take much coordination to change the outcomes of local elections.

    Fourth, and finally,, given that Canada is fairly promiscuous with its citizenship, and that a large number of Canadian citizens live more or less permanently in countries (i) whose interests are hostile to ours, and (ii) who have no regard for democratic principles, this decision meaningfully increases the risk of foreign interference in our election (for example, by pressuring Canadian citizens living abroad to vote a particular way or in particular ridings). The current row with China in which China is using the lives Canadian citizens in China as bargaining chips suggests that this is hardly a far-fetched concern.

    That said, we’re stuck with the Court’s interpretation on the point. Given that, and given that it’s interpretation is so badly at odds with our Parliamentary System, perhaps we need to amend the latter to accommodate this new right. One option would be to do as some European countries do and create special ridings for non-resident citizens (perhaps just a single seat or two for the rest of the world). That would give non-residents citizens a voice in parliament, while limiting the scope for riding shopping (and also limiting the scope for adverse consequence of foreign influence). It would also ensure that local representatives are actually chosen by people in the locality they represent. Furthermore, given the foreseeable expansion in scope of voting abroad (and the increased risk of foreign influence on citizens subject to the authority of foreign governments), we should also consider increasing security procedures for voting (e.g., eliminating mail-in ballots for everyone and requiring people to physically present themselves to an embassy or consulate to vote).

  2. Perhaps the categories as outlined and defined on the Government of Canada website could be used to determine eligibility to vote if living outside of Canada. Should voting be limited to those who pay taxes in Canada, i.e., taxation as a qualification for political representation.

  3. I don’t think taxation is a qualification for political representation. That said, I do think the concept of “ordinary residence” – which is what we use to establish liability for tax (and entitlement to benefits such as welfare or healthcare) – is a reasonable connecting factor.

    If you “ordinarily reside” in a riding, you should be entitled to vote in that riding. So the kid from Toronto who goes to university in Kingston for 4 years should be able to vote in Toronto, ditto if she goes to university in NYC for 4 years, in each case because they have a real connection to the riding in which they vote – they ordinarily reside there. De facto, that was the existing rule (since, saying that you were “temporarily” outside of Canada implies that you are ordinarily resident in Canada).

    Ultimately, our parliamentary system is premised on MPs representing geographically defined constituencies. For that to make sense, eligibility to vote in a particular constituency requires some connecting factor between voter and constituency- if people can vote in any riding they want, our system doesn’t work the way it’s designed. (Mind, some people might be OK with that, but if they want to change our parliamentary system, the avenue to do that is to amend the constitution, not reinterpreting one part of it in a way that is inconsistent with another) Ordinary residence is a well-defined legal concept for connecting rights (and liabilities) to geographically defined polities, the ability to vote in a riding should be limited to people who ordinarily reside in it.

  4. Expats like those domestic voters without a “fixed address” may be required to have a registered voter resident in the country vouch for them allowing them to vote in that individual’s riding. It’s interesting to note that in the U.S. according to the Federal Voting Assistance Program many states allow U.S. citizens 18 and older who have never resided in the U.S. to vote in a state in which a parent who is a U.S. citizen and a registered voter resides.

Leave a Reply

(Your email address will not be published or distributed)