On May 2, 2011, Canadians voted in the 41st federal election. Voters in the riding of Etobicoke Centre elected Ted Opitz to represent them in Parliament. The race was hotly contested. So too was the result.
A judicial recount showed that Mr. Opitz won by a plurality of just 26 votes. Boris Wrzesnewskyj, the runner-up, applied to the Ontario Superior Court of Justice under s. 524(1)(b) of the Canada Elections Act (the “Act”) to annul the election based on “irregularities … that affected the result of the election”.
524. (1) Any elector who was eligible to vote in an electoral district, and any candidate in an electoral district, may, by application to a competent court, contest the election in that electoral district on the grounds that…(b) there were irregularities, fraud or corrupt or illegal practices that affected the result of the election.
The alleged irregularities were, by and large, administrative errors relating to registration certificates and vouching – methods, provided for under the Act, that permit a person not on the official list of electors to prove their entitlement to vote on election day.
The application judge, Mr. Justice Lederer, declared the election “null and void”.
He found irregularities in 79 votes, including failures to follow statutory procedures designed to ensure that only those entitled to vote cast a ballot on election day.
Justice Lederer relied on the “magic number” test to determine whether these 79 irregularities “affected the result of the election”: where the number of irregular votes equals or exceeds the plurality of the winning candidate, the result of the election was affected.
Based on the evidence and the “magic number” test, Mr. Wrzesnewskyj succeeded on his application; Mr. Opitz’s victory in Etobicoke Centre was annulled.
Mr. Opitz appealed to the Supreme Court of Canada, as of right.
The Supreme Court of Canada
The Court, like the election, was closely divided. Justices Rothstein and Moldaver, writing for a 4-3 majority, reversed the application judge, allowed the appeal and restored the election result.
Of central concern was the Charter-protected right to vote and the enfranchising purpose of the Canada Elections Act. Time and again, the majority emphasized that the decision to annul an election disenfranchises all voters, not just those whose votes are set aside.
It was against this backdrop that the majority considered the appeal.
“We are asked to disqualify the votes of several Canadian citizens based on administrative mistakes, notwithstanding evidence that those citizens were in fact entitled to vote. We decline the invitation to do so. The Canadian Charter of Rights and Freedoms and the Canada Elections Act … have the clear and historic purposes of enfranchising Canadian citizens, such that they may express their democratic preference, and of protecting the integrity of our electoral process. Following these objectives and the wording of the Act, we reject the candidate’s attempt to disenfranchise entitled voters and so undermine public confidence in the electoral process.”
The majority judgment begins by putting Mr. Wrzesnewskyj’s application in context: the challenge was grounded in administrative errors; there were no allegations of fraud, corruption, illegal practices, or wrongdoing.
Given the complexity of administering a federal election, administrative errors are inevitable. If these errors are technical mistakes, the integrity of the electoral process remains intact. Only serious mistakes that affect the election result undermine the integrity of the process.
The task for the Court was to interpret the statutory language. With a clear definition of what constitutes “irregularities … that affected the result of the election”, the judges could then determine whether the application judge properly set aside the 79 votes in question.
“The current system of election administration in Canada is not designed to achieve perfection, but to come as close to the ideal of enfranchising all entitled voters as possible. Since the system and the Act are not designed for certainty alone, courts cannot demand perfect certainty. Rather, courts must be concerned with the integrity of the electoral system. This overarching concern informs our interpretation of the phrase “irregularities … that affected the result”.”
In defining this section of the Act, the majority relied on three interpretive aides: the constitutional right to vote and the objectives of the Act; the text and context of s. 524(1)(b); and the need to balance competing democratic values.
The purpose of the Act is to enfranchise all persons entitled to vote in order to permit them to freely express their democratic preferences. The Act has other purposes, but the majority identified enfranchisement as a cornerstone of the Act. This enfranchising purpose – coupled with the constitutional right to vote, guaranteed by s. 3 of the Charter – played a central role in defining s. 524(1)(b).
The text and context of the section also informed the analysis. Parliament chose its words carefully. It would have used the term “non-compliance” if any deviation from statutory procedure was a basis on which to annul an election. The Act requires something more. Relying on the “noscitur a sociis” rule of interpretation (“associated words”), the majority took meaning from the entire phrase: “irregularities, fraud or corrupt or illegal practices”. The common thread tying these words together is the seriousness of the conduct and the impact of that conduct on the integrity of the electoral process. By associating the word “irregularities” with the words “fraud” and “corrupt or illegal practices”, Parliament must have contemplated serious administrative errors capable of undermining the integrity of the electoral process.
Finally, the majority recognized and balanced competing democratic values in defining s. 524(1)(b). The Act has safeguards in place to prevent abuse but, at the same time, accepts a degree of uncertainty in the administration of elections. Interrelated and sometimes conflicting values must be balanced, including certainty, accuracy, fairness, accessibility, voter anonymity, promptness, finality, legitimacy, efficiency and cost. But the central value is the Charter-protected right to vote.
Taking into account the enfranchising purpose of the Act, the impact of any impugned conduct on the integrity of the electoral process, and the primacy of our Charter-protected right to vote, the majority defined an “irregularit[y] … that affected the result of an election” as a breach of a statutory procedure that results in an individual voting who was not entitled to vote. Entitlement consists of the fundamental requirements of age, citizenship, and residence.
Having interpreted the statutory language, the majority then turned to the analytical framework governing an application under s. 524(1)(b). The process involves two steps and the applicant who seeks to annul an election bears the legal burden of proof throughout.
First, an applicant contesting an election must establish an irregularity – a breach of a procedural safeguard designed to establish an elector’s entitlement to vote. If established, an applicant must then demonstrate that the irregularity “affected the result” of the election – someone not entitled to vote, voted. An error that does not go to entitlement cannot satisfy s. 524(1)(b). For the purposes of this second step, the majority considered evidence of entitlement unavailable at the poll on election day. Unlike the minority, the majority held that the judge may consider any evidence capable of establishing that a person was, in fact, entitled to vote despite the existence of an irregularity.
If a court is satisfied that irregularities affected the result of the election, the judge must fashion an appropriate remedy.
“Overturning an election is but one of several consequences that may flow from the failure of election officers to follow rules. A declaration that an election is annulled may be considered the ultimate public consequence of violating provisions of the Act, and accordingly should be reserved for serious cases.”
The decision to annul is discretionary under the Act. Only where a court is satisfied that the winner is in doubt should the results be set aside. The majority relied on the “magic number” test to make this determination. The “magic number” test requires a court to annul the election if the rejected votes are equal to or outnumber the winner’s plurality. Despite obvious deficiencies with this test – for example, it inherently favours the applicant by assuming all rejected votes were cast for the successful candidate – the majority applied it here, leaving open the possibility that a more realistic method may be adopted in the future.
Crucial to the majority’s decision was the lack of deference accorded to the application judge’s findings of fact.
Rothstein and Moldaver identified two errors of law made by the application judge. First, Lederer J. misstated the onus of proof in his analysis of Polls 31 and 426. Rather than asking whether Mr. Wrzesnewskyj had satisfied his onus of establishing an irregularity, Lederer J. asked whether Mr. Opitz had established that there was no irregularity, thereby shifting the burden of proof. The second error of law was Lederer J.’s failure to consider material evidence regarding Polls 174 and 89. For both Polls, evidence in the record enabled the majority to discern the identity of impugned voters. Even though formal procedures in the Act were broken by, for example, missing signatures, declarations, or forms, the record supported the inference that at least 59 of the 79 voters in question were entitled to cast their votes.
Absent these two errors, the application judge’s findings of fact would have been entitled to deference unless there was a palpable and overriding error. As a result of these two legal errors, Lederer J.’s factual findings were not entitled to deference, enabling the Court to examine the evidence and draw its own conclusions about the impugned votes at these four Polls.
“In order to uphold the findings of the application judge, we must be satisfied that he not only appreciated which of the parties bore the onus, but also that he applied the correct onus in arriving at his critical findings of fact. We cannot be so satisfied.”
For Polls 31, 426 and 174, the majority concluded that mistakes occurred in 49 votes, but none rose to the level of an “irregularity”. For Poll 89, Mr. Wrzesnewskyj established “irregularities” in 10 votes, but was unable to prove that these “irregularities” actually “affected the result of the election”. The majority concluded that all 59 of the impugned votes at these four Polls were wrongly set aside, leaving only 20 contested votes. That was sufficient to dispose of the appeal; even if the majority set all 20 aside, the “magic number” test could not be met.
Writing for the minority, the Chief Justice of Canada would have dismissed the appeal and upheld the application judge’s decision to annul the election result in Etobicoke Centre.
The minority departed from the majority in four significant ways: purpose of the Act; definition of “irregularities”; admissible evidence to prove entitlement; and deference to the application judge.
According to the minority, the overarching purpose of the Act is not enfranchisement. Rather, the purpose is to ensure the democratic legitimacy of federal elections in Canada. Turning to the language of the section, “irregularities” suggests non-compliance with provisions of the Act. “Irregularities” in s. 524(1)(b) are “failures to comply with the requirements of the Act, unless the deficiency is merely technical or trivial.” The minority therefore placed greater emphasis on adherence to procedure.
If an irregularity results in a vote improperly cast, the irregularity is one that could affect the result of the election. Like the majority, the minority used the “magic number” test to determine whether an irregularity “affected the result of an election”.
The types of irregularities that result in a vote improperly being cast are determined by reference to a central pillar of Canada’s electoral system: the principle of entitlement to vote.
“Every electoral system must strike a balance between enabling those who have the constitutional right to vote to do so, and ensuring that those who do not have that right are not allowed to vote. The formal system of entitlement is our mechanism for striking the right balance between these two concerns, while ensuring the efficiency and certainty of the electoral process. It aims to safeguard both the right to vote and the integrity of elections.”
There are three prerequisites to entitlement to vote: qualification, registration, and identification. A person who is a Canadian citizen and is at least 18 years old is qualified to vote. In order to be registered, a person must be included on the list of electors in their polling division or must file a registration certificate. Finally, a prospective voter must prove their identity and residence at the poll in accordance with the Act.
All three prerequisites to entitlement to vote must be satisfied before a person is permitted to cast a ballot in a Canadian federal election. Qualification alone is not enough. In the minority’s view, the majority erred by conflating qualification and entitlement. In so doing, they offended the plain words of the Act, the objective of certainty in the electoral process, and fundamental principles of fairness.
“My colleagues, with respect, merge the concepts of qualification and entitlement. They take the position that everyone who is qualified to vote and ordinarily resident in the electoral district is entitled to vote. Thus a voter who is not on the electoral list and has not filed a registration certificate (s. 149) can be later held to have been “entitled” to vote if he was qualified to vote and ordinarily resident in the electoral district. I cannot accept this view.”
Unlike the majority, the minority concluded that the application judge did not reverse the onus of proof. The Chief Justice pointed to clear statements in the decision below where Lederer J. articulated the proper onus and applied it accordingly. This finding was critical to the analysis because it meant that absent a palpable and overriding error, Lederer J.’s conclusions on findings of fact should not be disturbed.
The Chief Justice went through each of the Polls and, by and large, found evidence to support the application judge’s conclusions. Ultimately, the minority concluded that 65 of the 79 votes were properly set aside. The plurality being only 26 votes, the election result was properly set aside and the appeal should be dismissed.
“Votes were set aside in this case because of failures in the registration and identification prerequisites of entitlement. These cannot be remedied by after-the-fact proofs of qualification. Without the voter establishing his qualifications in an approved manner prior to voting, the Act is clear that he is not entitled to vote (ss. 6, 148.1 and 149).”
This decision is significant because it defines s. 524(1)(b) of the Canada Elections Act and provides the analytical framework to scrutinize an election based on administrative errors. The majority judgment demonstrates that courts will undertake a substantive, rather than a procedural, approach when reviewing an application to set aside an election based on administrative errors.
Another significant aspect of the decision is that the Court applied, but did not endorse, the “magic number” test. No other test was put forward in argument and, given the time constraints in play, it is unsurprising the Court did not formulate a new test. However, the majority explicitly identified difficulties with this test and left it open as to whether a new test would be adopted in the future. It would be desirable if a new test develops that remedies the current bias in favour of the applicant.
SCC Court File No.: 34845
Date of Decision: October 25, 2012