COURT OF APPEAL FOR ONTARIO

CITATION: Cusimano v. Toronto (City), 2012 ONCA 907

DATE: 20121224

DOCKET: C55365

Weiler, Juriansz, and Tulloch JJ.A.

BETWEEN

Agustine G. Cusimano

Applicant (Appellant/ Respondent by way of cross-appeal)

and

City of Toronto and Maria Augimeri

Respondents (Respondents/ Appellant by way of cross-appeal/ Respondent by way of cross/appeal)

Lorne Honickman and Rory Barnable, for the appellant/respondent by way of cross-appeal, Agustine G. Cusimano

Diana Dimmer and Tim Carre, for the respondent/respondent by way of cross-appeal, City of Toronto

Julian Heller and Zabi Yaqeen, for the respondent/appellant by way of cross-appeal, Maria Augimeri

Marcia J. Taggart, for the intervener, City of Mississauga

Caroline V. Jones and Michael Fenrick, for the intervener, Gloria Saccon

Heard: November 8, 2012

On appeal and cross-appeal from the judgment of the Divisional Court (Justices John R.R. Jennings, Michael R. Dambrot, and John R. Sproat), dated December 19, 2011, with reasons reported at 2011 ONSC 7271 and 2012 ONSC 2629.

Table of Contents

A. Overview..................................................................................................... 3

B. Facts.......................................................................................................... 7

(1) The Statutory Scheme for Municipal Elections and the 2010 Ward 9 City Councillor Election................................................................................................... 7

(2) Decisions Prior to this Appeal....................................................... 12

(a) The Decision of the Application Judge................................... 12

(b) The Decision of the Divisional Court...................................... 15

(c) Opitz v. Wrzesnewskyj.......................................................... 20

C. Legal Issues and Analysis.......................................................................... 24

(1) Are the irregularities in the 275 VLCRFs sufficient in and of themselves to discount the votes they represent?....................................................................... 24

(2) What evidence was the Divisional Court entitled to consider in concluding that the omission of the election official’s signature did not affect the result of the election? 29

(3) Conclusion on the Main Appeal..................................................... 34

D. Costs........................................................................................................ 35

(1) Facts........................................................................................... 36

(a) Costs Before the Application Judge....................................... 36

(b) Augimeri’s Motion to Intervene as a Party.............................. 36

(c) Costs Before the Divisional Court.......................................... 39

(d) Opitz and Costs.................................................................... 45

(2) The Parties’ Positions Respecting the Divisional Court’s Decision on Costs 47

(3) Analysis of Costs for Augimeri’s Motion to Intervene and Proceedings Before the Application Judge and the Divisional Court.............................................. 49

(4) Costs of this Appeal..................................................................... 57

(5) Summary of Costs....................................................................... 59

Weiler J.A.:

A. Overview

[1] This appeal requires the court to determine whether Maria Augimeri was validly elected as a City Councillor in Ward 9 of the City of Toronto.

[2] In the October 25, 2010 municipal election conducted pursuant to the Ontario Municipal Elections Act, 1996, S.O. 1996, c. 32 (“MEA”), Maria Augimeri defeated Agustine Cusimano by a plurality of 89 votes. There is no suggestion of fraud, although one person voted twice. Cusimano contested the validity of Augimeri’s election,[^1] primarily on the basis of irregularities contained in the “Voters’ List Change Request Forms” (VLCRFs).

[3] VLCRFs are used when a person’s name is not on the voters’ list for a particular ward. In such instances, that person will be asked to do two things: first, to present identification verifying his or her name, qualifying address, and signature, like all other voters who attend at a voting place on election day; and second, to complete and sign a VLCRF containing a declaration of entitlement to vote. An election official, having verified that identification was provided and the form was properly completed, will then add his or her signature to the end of the VLCRF and provide the voter with a ballot.

[4] In the Ward 9 election, 275 VLCRFs used to add voters to the voters’ list were submitted without an election official’s signature, though each contained a signed declaration of the voter attesting to his entitlement to vote.[^2] The omission of the election official’s signature on these VLCRFs was a clerical mistake that was not within the voters’ control or that of the candidates. Although Cusimano complains of hundreds of irregularities in the Ward 9 election, these 275 VLCRFs that lack an election official’s signature but include a voter declaration are the focus of this appeal.

[5] Cusimano applied to the Superior Court to determine the validity of the election pursuant to s. 83(1) of the MEA. Section 83(6) states that the court shall not determine an election to be invalid if (a) a procedural irregularity occurred that “did not affect the result of the election” and (b) “the election was conducted in accordance with the principles of this Act.”

[6] As discussed in greater detail below, the application judge and the Divisional Court adopted two divergent approaches to determining whether the irregularities in Ward 9 did, in fact, “affect the result of the election.”

[7] Following the Divisional Court’s decision, the Supreme Court of Canada released its reasons in Opitz v. Wrzesnewskyj, 2012 SCC 55. The majority’s reasons, also discussed in greater detail below, provide support for the Divisional Court’s approach and conclusion.

[8] In Opitz, the Supreme Court of Canada dealt with a challenge to the validity of an election under the Canada Elections Act, S.C. 2000, c. 9 (“CEA”). A number of voter declaration forms, which are functionally equivalent to VLCRFs, were not signed by the voter, but were signed by the election official. As is the case here, the Supreme Court had to choose between two divergent approaches in determining whether the omission affected the result of the election.

[9] The two approaches were the procedural approach, whereby failure to follow a procedural safeguard will invalidate the vote cast, and the substantive approach, whereby primacy is given to the underlying right to vote and failure to follow a procedural safeguard is not determinative. Rather, when the court follows the substantive approach, it will have regard to all of the evidence respecting the vote, and decide whether persons not entitled to vote, voted. The majority of the Supreme Court in Opitz rejected the procedural approach in favour of the substantive approach.

[10] The Supreme Court also adopted the “magic number test.” Under the “magic number test,” the court considers whether the number of discounted ballots was greater than the plurality by which the person originally elected won. If so, the court should exercise its discretion to annul the election. Applying the “magic number test” to this case, if the votes represented by the 275 VLCRFs are discounted, then the election is invalid because the discounted votes exceed the 89 votes by which Augimeri won.

[11] Applying the specific wording of the MEAand the substantive approach endorsed in Opitz, I conclude that the omission of the election official’s signature on those 275 VLCRFs did not compromise the voter’s underlying entitlement to vote. The 275 ballots represented by those VLCRFs should not be discounted, as the procedural irregularities in issue did not “affect the result of the election.”

[12] For that reason and the reasons that follow, I would therefore uphold the decision of the Divisional Court and dismiss the appeal.

[13] Certain additional issues were argued before this court by Augimeri, the winning candidate in Ward 9, who became a party to these proceedings after the application judge’s decision. These issues were also raised before the Divisional Court, but it did not decide them. I do not propose to decide them here. There is no decision at first instance to review and the resolution of these issues is not essential to the ultimate outcome of this appeal.

[14] A subsidiary but important issue to the main appeal is who should pay the costs of the proceedings. I will consider the question of costs after I deal with the main appeal.

B. Facts

(1) The Statutory Scheme for Municipal Elections and the 2010 Ward 9 City Councillor Election

[15] The MEA is the governing statute for municipal elections held in the province of Ontario. There are several notable elements of the statutory scheme for the purposes of this appeal.

[16] Section 17(2) sets out a list of qualifications which entitle an individual to vote in a municipal election on voting day. It stipulates that a qualified elector will:

• Reside in or own property located in the municipality or be a spouse of such an owner or tenant;

• Be a Canadian citizen;

• Be at least 18 years old; and

• Not be prohibited from voting under s. 17(3) or otherwise by law.[^3]

[17] Prior to election day, a preliminary voters’ list is prepared by the Municipal Property Assessment Corporation (MPAC) for each polling division. Typically, the voters’ list does not contain complete information and requires amendment on the day of the election to accommodate qualified and eligible individuals who have been left off the list by MPAC.

[18] An individual who has been left off of the voters’ list and is seeking to be added must complete a VLCRF. A VLCRF has three sections that the voter is instructed to complete. The first section requires the voter to specify what kind of change is being requested – for example, a voter might check the box which says “Add my name to the list”. The second section requires the voter to fill in his or her name, date of birth, and address, and to specify his or her entitlement to vote – for example, a voter might check the box that says “Resident” if he or she is a resident of the ward in which the polling station is located. The third section requires the voter to sign and date a declaration that he or she is “a Canadian citizen, at least 18 years old on voting day, and entitled to be an elector” and that “the information provided…is true and accurate.”

[19] The VLCRF also has a section that the election official manning the polling station is instructed to complete. The election official is to review the form to ensure it contains complete information, mark a box indicating that the voter’s identification has been checked, and place his or her signature at the bottom of the form. A substantial number of the VLCRFs that were not signed by the election official had a checkmark confirming that the election official had checked the voter’s identification.[^4]

[20] In support of his submission that the 275 VLCRFs not signed by an election official should be declared invalid, Cusimano points to s. 24 of the MEA. Section 24(3) stipulates that if the clerk is “satisfied that the applicant is entitled to have the requested change made, the clerk shall…endorse the application to indicate approval…and the voters’ list will be changed to reflect the approved application.”

[21] Cusimano submits that in order to comply with the statutory requirement that the clerk “endorse the application,” the election officials processing the VLCRFs were required to place their signatures on each submitted form. Their failure to do so amounts to an irregularity that provides a basis for annulling the election in Ward 9 and ordering a new one.

[22] Section 83(1) permits a person who is entitled to vote in an election to contest the validity of that election. The person may make an application to the Superior Court of Justice for a determination of whether an election is valid and, if it is not, whether a by-election should be held. The MEA does not provide any specific guidelines indicating when it is appropriate for a judge to invalidate an election. However, by including s. 83(6), the legislature has seen fit to restrict a judge’s discretion in cases of procedural irregularities.

[23] Section 83(6) stipulates that “[t]he court shall not determine an election to be invalid if,

a) an irregularity described in subsection (7) occurred at the election but did not affect the result of the election; and

b) the election was conducted in accordance with the principles of this Act.”

[24] The procedural irregularities described in subsection (7) are:

1. An irregularity on the part of the clerk or in any of the procedures before voting day.

2. Failure to have a voting place open at the appointed location and time.

3. Non-compliance with a provision of this Act or of a regulation, by-law, resolution or procedure made, passed or established under this Act, dealing with voting, counting of votes or time requirements.

4. A mistake in the use of forms, whether prescribed or not.

[25] Together, ss. 83(6) and (7) prevent a court from invalidating an election that was conducted in accordance with the principles of the MEA, where the procedural irregularities did not affect the result. In other words, s. 83(6) amounts to a saving provision for elections in which certain specified irregularities occurred when those irregularities did not affect the outcome or integrity of the electoral process.

[26] Before the application judge, both parties conceded that the errors in the impugned VLCRFs fell within either ss. 83(7)(3) or (7)(4), and were therefore procedural irregularities subject to the saving provision in s. 83(6). The Divisional Court also accepted this.

[27] The central issue on appeal is therefore whether these procedural irregularities “affect[ed] the result of the election”. Before us, Cusimano made no specific submissions respecting s. 83(6)(b), namely, whether the election was conducted in accordance with the principles of the Act. Nonetheless, because the words of an Act must be considered in context, the principles underlying the MEA should be borne in mind.

[28] The “principles of the Act” referred to in s. 83(6)(b) are not expressly set out in the legislation. Rather, they are to be inferred from the provisions of the Act. These inferred principles were considered at some length by Dambrot J. at paras. 104-114 of the Divisional Court’s decision in this case. For the purposes of this appeal, it is necessary for me only to refer to one of them.

[29] In Montgomery v. Balkissoon (1998), 1998 CanLII 1993 (ON CA), 38 O.R. (3d) 321, at p. 322, this court explained that the “the proper majority vote decides the election.” This principle is enacted by “ensuring, so far as is reasonably possible, that valid votes be counted and invalid votes be rejected”: Montgomery, at p. 322.

[30] In articulating this principle, the court warned against confusing the principles of the Act with the means used to ensure conformity with them. This court’s comments in Montgomery speak to the enfranchising objectives of the MEA and the MEA’s overridingaim of ensuring that the ballots cast by individuals who are entitled to vote are not discounted by virtue of administrative or procedural errors that do not affect the result of the election.

[31] I turn now to the decisions of the application judge and the Divisional Court.

(2) Decisions Prior to this Appeal

(a) The Decision of the Application Judge

[32] The application judge was not satisfied that the irregularities in the VLCRFs had not affected the outcome of the election, and therefore declined to apply the saving provision under s. 83(6). She declared the Ward 9 election invalid and ordered a by-election.

[33] Cusimano argued before the application judge, as he did again on appeal, that the failure of an election official to sign a VLCRF was a failure to comply with the statutory requirement that a clerk or a delegate of the clerk “endorse” the application. He submitted that any individual added to the voters’ list on the basis of a VLCRF missing an election official’s signature was, as a result of the statutory breach, not entitled to vote. Because the number of voters added to the list on the basis of a VLCRF missing an election official’s signature (275) exceeded the margin of votes by which he had lost (89), he argued that the result of the election was “affect[ed]” within the meaning of s. 83(6).

[34] The application judge accepted Cusimano’s submission that the word “endorse,” contextually interpreted within the phrase “the clerk shall…endorse the application to indicate approval”, means “sign.” By extension, she concluded that the failure of an election official to sign the VLCRF was a breach of s. 24 of the MEA, that the irregularities in the VLCRF fell within either s. 83(7)(3) or (7)(4), and that the election was therefore subject to the saving provision in s. 83(6).

[35] In considering whether the irregularities had “not affected the result of the election” within the meaning of s. 83(6), she was guided by the decision in O’Brien v. Hamel (1990), 1990 CanLII 6834 (ON SC), 73 O.R. (2d) 87. The O’Brien court took what the majority of the Supreme Court in Opitzrefers to as a “procedural approach” to this determination: Opitz, at para. 54. Under this approach, failure to follow a procedural safeguard set out in the governing legislation is alone sufficient to discount the vote cast. If the number of discounted votes exceeds the margin of victory, it cannot be said that the statutory non-compliance did not affect the result of the election: O’Brien v. Hamel, at pp. 95-96.

[36] The application judge observed that the failure to comply with procedural safeguards set out in the MEA was widespread in Ward 9. At para. 49 of her reasons, she stated:

Hundreds of VLCRFs in various polling divisions were not signed by an election official. This is not a case of isolated irregularities. Nor are the irregularities trifling.

[37] Applying the reasoning in O’Brien, she concluded that the number of votes tainted by these irregularities exceeded the margin of votes by which Cusimano lost, and therefore that the result of the election had been affected by the irregularities. Though she considered whether it was open to her to order the City to verify the entitlement of the voters who cast the impugned ballots, she ultimately felt she was limited by the MEA to determining whether or not the election was valid on the record before her.

[38] Further, while she felt it was unclear from the jurisprudence which party bore the onus of demonstrating that an irregularity had affected the result of the election under s. 83(6), if it was on Cusimano, it had been satisfied for the reasons above.

[39] Finally, the application judge considered the respondent’s submission that the election could be saved by a presumption of regularity – the presumption that only individuals entitled to vote had in fact received ballots. The respondents argued that such a presumption should arise from evidence in the record that the election officials were competent and hard-working, that they had followed the regulatory procedures on which they were trained, and that they believed only those entitled to vote would have received ballots.

[40] The application judge declined to apply a presumption of regularity in this case. She felt it would be inappropriate in light of the proof presented by Cusimano that at least one person voted twice at a Ward 9 polling station – definitive evidence that not only individuals entitled to vote had received ballots.

[41] She concluded that the failure of the election officials to comply with the statute was fatal to the Ward 9 election, noting at para. 65:

…O’Brien v. Hamel held that the fact that the methods used by election officials may have been as good as that provided for in the statute did not save the votes. Compliance with the statute was required to ensure the integrity of the system.

[42] In the result, the application judge declared the Ward 9 election invalid and ordered a by-election be held.

(b) The Decision of the Divisional Court

[43] The respondents appealed the application judge’s order to the Divisional Court. Maria Augimeri, the victor in Ward 9, was granted leave to join as a party to the appeal. The City of Toronto and Augimeri were supported by two interveners, both of whom intervened again in the appeal before this court: the City of Mississauga, and Gloria Saccon, one of the voters in Ward 9. Dambrot J., writing for the court, allowed the appeal, declared the Ward 9 election valid, and set aside the application judge’s order for a by-election.

[44] Dambrot J. accepted, at paras. 54 and 60, that “endorse” within the meaning of s. 24(3) required at least “some physical act applied to the form,” such as a signature, initial, or stamp, and that the failure of the election officials to endorse the VLCRFs amounted to an “irregularity” within the meaning of s. 83(7) of the MEA.

[45] He then dealt with whether the election could nonetheless be saved under s. 83(6) – whether it could be said that the result had not been affected by the irregularities and that the election had been conducted in accordance with the principles of the MEA. Dambrot J.’s description of s. 83(6), at para. 62 of the Divisional Court’s reasons, bears repeating here:

…[Section] 83(6) is a very broad saving provision. As I have already noted, it provides a sweeping definition of “irregularity,” while narrowly circumscribing the circumstances in which an irregularity will be fatal to an election. It is a recognition that irregularities are inevitable in an election and an affirmation that the democratically expressed will of the electorate should not lightly be overturned. An election will only be set aside where the irregularity either violates a fundamental democratic principle or calls into a question whether the tabulated vote actually reflects the will of the electorate.

[46] As I will discuss in greater detail below, this conception of s. 83(6) foreshadows the Supreme Court’s decision in Opitz.

[47] Before deciding whether the irregularities in the 275 VLCRFs did not affect the result of the election, Dambrot J. considered which party bears the standard of proof in such a determination. He held that the applicant has the burden of establishing, on a balance of probabilities, both that an irregularity existed and that it affected the outcome.

[48] In deciding the issue of whether the result of the election was affected by the irregularity, Dambrot J. held that the application judge had erred in law in her approach, for three reasons.

[49] First, he found her refusal to apply a presumption of regularity, on the basis of evidence that one person was permitted to vote twice, unreasonable.

[50] Second, he found the application judge’s use of an arithmetic formula – the number of votes at issue less the number of votes representing the margin of victory – to determine the impact of the irregularities on the result of the election to be insufficient. Instead, Dambrot J. was of the opinion that the application judge was obliged to examine all the evidence to ascertain the “true effect” of the irregularities.

[51] Determining the “true effect” of the irregularities requires the court to consider whether they resulted in either (a) votes being counted that were cast by individuals who were not entitled to cast them, or (b) votes not being counted that were cast by individuals who were entitled to cast them. Where those votes actually changed the outcome of the election, because they exceeded the margin of victory, the irregularities could properly be said to have affected the result of the election. This approach closely parallels what the Supreme Court in Opitz calls the “substantive approach,” discussed in greater detail below.

[52] Third, Dambrot J. found that the application judge erred in law by finding there was no evidence in the record upon which she could conclude that the impugned VLCRFs were submitted by voters who were in fact entitled to vote. In particular, she overlooked the fact that the scheme in place for processing the VLCRFs provided a series of procedural safeguards to ensure that each ballot cast was submitted by an entitled voter. As Dambrot J. describes, at para. 95-96:

All electors were required by the Identification Policy to produce specific identification verifying their name, qualifying address, and signature in order to vote. Once it was confirmed by an election official that an elector whose name was not on the voters’ list was in the correct ward, the official was required by the Voting Procedure to give the elector a VLCRF and have the elector complete it. The elector was required to record his or her name and address on the form; to tick boxes indicating occupancy, residency, and school support; to sign and date a declaration that he or she was a Canadian citizen, at least 18 years old on voting day, entitled to be an elector, and that the information provided was true and accurate; and to present the completed form to an official. The official was then required to review the form, tick a box confirming that identification had been checked, and, if satisfied of the elector’s eligibility, sign the form, add the elector’s name to the front of the voters’ list and place the original VLCRF and a duplicate in appropriate envelopes. The Ballot Officer was then required to provide the elector with a ballot and follow the ordinary voting procedure.

There is no reason to doubt that this scheme was generally followed by the election officials. [Emphasis added.]

[53] While Dambrot J. acknowledged the errors in hundreds of VLCRFs identified by Cusimano, he concluded that there was a significant body of evidence that in each case a voter’s name was added to the voters’ list, that voter was in fact eligible to vote. The evidence included the voter procedure described above, the affidavits of seventeen election officials that, to the best of their knowledge, they had followed the voting procedure, and declarations signed by 275 of 286 voters attesting to being Canadian citizens, 18 years of age, and entitled to vote.

[54] Dambrot J. placed particular emphasis on these declarations because they were made under threat of penalty of a fine of up to $25,000, pursuant to s. 89(h) and 94.1(1) of the MEA. The declarations thus provided compelling evidence that the individuals who signed them were entitled to vote. Dambrot J. also pointed out that “relying on the declarations of voters that they are entitled to vote reinforces the enfranchisement of citizens and guards against disenfranchisement”: para. 102.

[55] As a result, the Divisional Court allowed the appeal, declared the Ward 9 election valid, and set aside the order that a by-election be held.

(c) Opitz v. Wrzesnewskyj

[56] As indicated at the outset of these reasons, the adoption of the substantive approach by the majority of the Supreme Court in Opitz has a direct impact on the proper resolution of this appeal. I will therefore briefly review the majority’s decision before discussing the legal issues raised in this case.

[57] Opitz won in the Etobicoke Centre electoral district by a plurality of 26 votes. Wrzesnewskyj was the runner-up. He applied to have the election annulled under s. 524(1)(b) of the CEA, which allows a candidate to contest the validity of an election on the grounds that there were “irregularities…that affected the result of the election”.

[58] Recall that, in order for the saving provision in s. 83(6)(a) of the MEA to be applied, the court must determine that the irregularity “did not affect the result of the election”. This phrase is simply the negative construction of the language contained in the CEA enabling a candidate to obtain a declaration of invalidity.

[59] The majority of the Supreme Court, whose reasons were co-authored by Rothstein and Moldaver JJ., with Deschamps and Abella JJ. concurring, dismissed the appeal. They emphasized two principles. The first is that the purpose of an election statute is “to enfranchise all persons entitled to vote and to allow them to express their democratic preferences”: Opitz, at para. 35. The second is that a remedial by-election is not a perfect substitute because it disenfranchises all those electors who voted in the original election: Opitz, at para. 48.

[60] The Supreme Court was unanimous in holding that the challenger bears the burden of proving, on a balance of probabilities, that irregularities have affected the result of the election: see Opitz, at paras. 52 and 172. Indeed, the majority cites the Divisional Court’s decision in this case in support of its conclusion: Opitz, at para. 52.

[61] The majority then proceeded to discuss the two approaches historically taken to determining when irregularities will justify invalidating an election. As I have indicated, these two approaches, the procedural approach and the substantive approach, closely parallel the approaches taken by the application judge and the Divisional Court respectively in this case.

[62] The procedural approach, followed in O’Brien v. Hamel, requires the court to discount any vote cast where there was a failure to comply with procedural steps set out in the governing act aimed at establishing an individual’s entitlement to vote: Opitz, at para. 54. This approach places a premium on form over substance, and risks overturning an election that reflects the will of voters who in fact had the right to vote. The procedural approach also enlarges the likelihood of litigation, because losing candidates will look for “technical administrative errors in the hopes of getting a second chance”: Opitz, at para. 56.

[63] Unlike the procedural approach, the substantive approach looks beyond procedural non-compliance and emphasizes the purpose of an election statute, namely, “to enfranchise all persons entitled to vote and to allow them to express their democratic preferences”: Opitz, at para 35. In adopting the substantive approach, the majority stated, at para. 57:

The substantive approach is recommended by the fact that it focuses on the underlying right to vote, not merely on the procedures used to facilitate and protect that right. In our view, an approach that places a premium on substance is the approach to follow in determining whether there were “irregularities…that affected the result of the election”.

[64] The substantive approach, as formulated by the Opitz majority, entails two steps. First, an applicant seeking to invalidate an election must demonstrate that an “irregularity” has occurred. Because an “irregularity” is not specifically defined in the CEA, the majority defines an “irregularity” as a “breach of a statutory provision designed to establish a person’s entitlement to vote”: Opitz, at para. 58. By “entitlement”, the court means an individual’s age, citizenship, and residence – not compliance with procedural safeguards under the governing act: Opitz, at para. 63.

[65] The second step an applicant must take under the substantive approach is to demonstrate that the irregularity “affected the result” of the election by establishing that someone not entitled to vote, voted. A vote cast by someone not entitled to vote will be considered invalid and discounted. A discounted vote affects the result of the election, because it changes the vote count: Opitz, at para. 59.

[66] Once an applicant has led evidence of an irregularity respecting entitlement to vote, his or her prima facie evidentiary burden is met. The respondent then must adduce evidence that either no irregularity occurred or, despite the irregularity, that the votes in question are valid: Opitz, at para. 61. A court may consider any evidence, produced by either party, including after the fact evidence, which is relevant to the three criteria of entitlement: age, citizenship and residence: Opitz, at paras. 62-63.

[67] If it is established on the basis of all the evidence that there were irregularities that “affected the result of the election”, a court may annul the election. This discretion should be exercised when the court is satisfied that the true winner of the election is in doubt. The winner of the election will be in doubt when a sufficient number of votes have been discounted on the basis that they were cast by persons not entitled to vote.

[68] Specifically, while leaving room for the possibility of another, “more realistic” method for assessing controverted elections, the majority endorsed the “magic number test”: Opitz, at paras. 71-73. As discussed above, under this approach, an election will be declared invalid where the number of discounted votes exceeds the margin of victory. This is the same formula used by the application judge and the Divisional Court in this case.

C. Legal Issues and Analysis

[69] Cusimano argues that each of the 275 VLCRFs in issue contain an irregularity which is alone sufficient to discount the votes represented: the omission of an election official’s signature. Because of this irregularity, he submits that there is no assurance that those individuals who were permitted to cast votes were qualified and entitled to do so.

[70] I will first discuss the nature of the irregularity in issue and then the evidence respecting whether the persons who voted were, in fact, entitled to vote. I conclude, as did the Divisional Court, that the omission of the election officials’ signatures on the 275 VLCRFs did not compromise the voters’ entitlement to vote and therefore “did not affect the result of the election.”

(1) Are the irregularities in the 275 VLCRFs sufficient in and of themselves to discount the votes they represent?

[71] Both the wording of the MEA and the majority decision in Opitz are fatal to Cusimano’s argument.

[72] As I have indicated, the irregularity in this case is a procedural irregularity described in s. 83(7) of the MEA. Section 83(6) of the MEA makes it clear that a procedural irregularity described in s. 83(7) will not invalidate an election so long as (a) the irregularity did not affect the result and (b) the election was conducted in accordance with the principles of the MEA. Thus, on a plain reading of the MEA, the procedural irregularities in the 275 VLCRFs are not sufficient in and of themselves to discount the votes they represent. The court must, under the clear legislative direction in the MEA, go on and consider whether the saving provision of s. 83(6) applies. Cusimano’s argument is therefore not supported by the wording of the MEA.

[73] Despite this, Cusimano points this court to the majority’s observation in Opitzat para. 57 that, “[p]roof of an irregularity may itself be sufficient to discount a vote.” He says this statement supports his argument that an irregularity alone may be sufficient to discount a vote, and that the irregularity in question – the failure of the election officials to sign the VLCRFs – is one such irregularity.

[74] Cusimano’s submission fails to place the majority’s comment in Opitz in context. First, as Cusimano acknowledges in his factum, the CEA does not contain a definition of “irregularity”. The majority instead defined an “irregularity” serious enough to trigger the application of s. 524(1)(b) as a breach of a statutory provision designed to determine entitlement: Opitz, at para. 58. When the majority refers to an irregularity that may alone be sufficient to discount a vote, it is therefore referring to non-compliance with a statutory provision designed to establish a voter’s entitlement to vote.

[75] The wording of the MEA is complimentary to this aspect of the decision in Opitz. Under the MEA, “irregularities” that trigger the saving provision of s. 83(6) are expressly defined in s. 83(7); they are procedural irregularities. Indeed, the heading of the saving provision in s. 83(6) is “Effect of procedural irregularities”. Entitlement to vote under the MEA is governed by s. 17(2), which lists the same three requirements as set out in Opitz – age, citizenship and residence, plus the requirement under s. 17(2)(d) that the elector not otherwise be prohibited by law from voting.

[76] Irregularities respecting entitlement to vote under s. 17(2) are not included in the saving provision of s. 83(6). Therefore, applying Opitz, it is only irregularities that call into question the criteria set out in s. 17(2) that may in themselves be sufficient to discount a vote. By extension, the failure of an election official to sign a VLCRF as required under the MEA is not the type of irregularity that will alone be sufficient to discount the underlying vote.

[77] Indeed, if this court were considering the CEA instead of the MEA, it would be questionable whether the omission of the election officials’ signatures even rises to the level of seriousness required to trigger the application of s. 524 under the first step of the Opitz analysis. Under the MEA, however, a court confronted with a procedural irregularity under s. 83(7) in a validity application must proceed to the second stage of the Opitz analysis pursuant to s. 83(6). It is therefore unnecessary to determine whether the procedural irregularities in this case would fall within the definition of “irregularities…that affected the result” put forward in Opitz.

[78] Second, the majority in Opitz draws an important distinction between “entitlement” to vote and procedural mechanisms used to satisfy election officials that voters are entitled to vote. “Entitlement” relates to the “fundamental requirements of age, citizenship and residence”: para. 63. Procedural mechanisms allow those citizens who meet these requirements, and are therefore entitled to vote, to actually cast a ballot on election day. Procedural safeguards in electoral statutes are not to be treated as ends in themselves: Opitz, at paras. 33-34.

[79] Thus, the Supreme Court has made clear that procedural requirements should not trump the substantive right to vote. Similarly, here, entitlement to vote under s. 17(2) of the MEA does not depend on compliance with the procedural safeguards set out in the statute.

[80] The requirement that an election official sign a VLCRF is a procedural safeguard. It is intended to ensure that the election official has confirmed that only a person entitled to vote is voting, by checking the voter’s identity, age and residence.

[81] While the breach of a procedural irregularity does not compromise the voter’s underlying and fundamental entitlement to vote, it is “evidence from which it may be inferred that a voter was not entitled to vote”: Opitz, at para 60. Cusimano was not required to prove the votes were fraudulently cast or to lead direct evidence of a lack of entitlement: Opitz, at para. 57. However, the respondents in this appeal were entitled to present, as they did, evidence that, despite the procedural irregularity, the ballots were cast by voters who were entitled to vote: Opitz, at para. 61.

[82] Under the substantive approach set out in Opitz and the clear statutory language of the MEA, the court mustlook to all of the evidence to determine whether the procedural irregularities did, in fact, affect the result of the election. Without this second step, the carefully drawn distinction between entitlement and procedural safeguards drawn by the majority in Opitz would be turned on its head.

[83] For these reasons, I would reject Cusimano’s argument that the omission of the election official’s signature on the 275 VLCRFs is sufficient in and of itself to discount the votes in question. His argument does not give effect to either the wording of the MEA or the substantive approach adopted by the Supreme Court in Opitz.

[84] I will now consider the second prong of Cusimano’s argument, namely, that there was insufficient evidence to assure the Divisional Court that the 275 VLCRFs lacking an election official’s signature nonetheless represented ballots cast by voters who were entitled to vote.

(2) What evidence was the Divisional Court entitled to consider in concluding that the omission of the election official’s signature did not affect the result of the election?

[85] Cusimano raises two specific concerns about the evidence the Divisional Court used to conclude that the impugned ballots were, in fact, cast by electors who were entitled to vote.

[86] First, he argues that the voter declarations are insufficient, in the absence of an election official’s signature or other concrete evidence of qualification, to validate the ballots in issue. As I noted above, all 275 VLCRFs in issue in this case were signed by a voter, declaring that he or she was a Canadian citizen, 18 years of age, and entitled to vote on election day. Cusimano submits that no weight should be given to these declarations or, alternatively, that they are insufficient to show that, despite the lack of an election official’s signature, the ballots cast were valid.

[87] In support of this argument, he points out that the registration certificates impugned in Opitz, though lacking voters’ signed declarations, were signed by an election official – “concrete evidence,” according to Cusimano, “confirming qualification and entitlement”. Because there are no election officials’ signatures on the VLCRFs in this case, Cusimano submits that there is no such direct evidence from which it can be inferred that the ballots arising from the 275 VLCRFs were cast by qualified voters. In his words, “[t]he presence of the declaration does not show that the ballots belonged in the ballot box” (emphasis omitted).

[88] Second, Cusimano argues that the Divisional Court should not have accorded as much, if any, weight to the after the fact affidavit evidence of election officials manning Ward 9 polling stations. Many of the affidavits include sworn statements that the election officials’ believed only qualified electors were permitted to vote on election day. Cusimano submits that such generalized, after the fact evidence involves “a speculative approach” not sanctioned by Opitz and cannot be used to save the votes cast under the 275 irregular VLCRFs.

[89] The affidavits were completed by DROs, MDROs and Ballot Officers who were involved in the election in Ward 9. The election officials describe the procedures and policies surrounding the administration of the election, the mandatory training they were required to complete, and their recollection of voting day. Most swear that, while it is possible, due to inadvertence, or to heavy voter traffic, that not every VLCRF was signed, they would have only added voters to the voters’ list and distributed ballots after being satisfied that those persons were entitled to vote.

[90] Cusimano’s submissions do not accurately reflect the decision in Opitz. In determining whether the result of an election is affected, the Supreme Court is clear that a judge is entitled to consider more than direct evidence.The majority stated, at para. 74 of their reasons:

In determining whether the result was affected, an application judge may consider any evidence in the record capable of establishing that the person was in fact entitled to vote despite the irregularity, or that the person was not in fact entitled to vote.

The minority similarly observed, at para. 170, that an applicant may lead either direct or circumstantial evidence to demonstrate, on a balance of probabilities, that persons not entitled to vote, voted.

[91] In relation to Cusimano’s first argument respecting the effect of signed voter declarations, the majority in Opitz indicated that such declarations – which establish age and citizenship – could be evidence that the voters in question were in fact entitled to vote: Opitz, at paras. 110-111. The dissenting reasons, authored by McLachlin C.J., state that the signed declaration is “a vital prerequisite of entitlement to vote”: Opitz, at para. 159. I therefore reject Cusimano’s argument and agree with the Divisional Court that the voter declarations are important evidence of voter qualification and entitlement.

[92] In relation to Cusimano’s second argument that after the fact evidence should not be considered, the majority of the court in Opitz did consider after the fact evidence. For example, at para. 61, the majority suggests that where registration certificates are missing, a party “may point to evidence showing that [the registration certificates] were completed, such as….evidence from the polling officials that registration took place.”

[93] In relation to poll 426, where 26 votes were at issue, the majority states at para. 102 of its reasons, “[t]he DRO, when asked eight months after the election, said she thought she had completed the 33 certificates.” Although the application judge placed little weight on the DRO’s comments, the majority of the Supreme Court, at para. 103, disagreed and accepted the DRO’s recollection as direct evidence of her familiarity with proper protocol:

With respect, we see the matter differently. Any vagueness in the DRO’s evidence is consistent with her being asked to recall something that occurred eight months earlier. Her recollection provides direct evidence that she knew she had to complete the registration certificates.

[94] The majority concluded, at para. 104, that the evidence of the DRO’s recollection, along with the evidence that the poll clerk filled out the relevant poll book page, made it “improbable, to say the least” that the DRO failed to comply with her obligation to verify voter entitlement.

[95] It is thus clear that the majority in Opitz considered the election official’s recollection to be direct evidence. The affidavit evidence of the election officials in this case similarly conveys their recollection that they were familiar with and followed proper protocols with respect to verifying voter entitlement. The affidavit evidence is therefore properly considered direct evidence, along with the voter declarations, and Cusimano’s argument that no direct evidence exists must fail.

[96] I must point out that I am in disagreement with one statement made by Dambrot J. with respect to the election officials’ affidavit evidence. He states, at para. 99 of his reasons:

While of course there were more than these seventeen officials who added voters to the list in Ward 9 and Ward 4 TDSB, I consider this sample to be large enough to give me significant confidence that very few and most likely no voters were permitted to vote who did not present the required proof of eligibility.

[97] This sentence is problematic. There was no evidence that the affidavits of the officials were statistically significant, because there was no evidence that they were taken from a sufficiently large or randomized sample to represent the hundreds of election officials manning the Ward 9 polling stations. In this regard, I agree with Cusimano’s submission that Dambrot J. was not entitled to reach the conclusion he did based on these affidavits alone.

[98] That said, evidence must be considered as a whole, and Dambrot J.’s comments in para. 99 must be placed in the context of paras. 94-100 of his reasons. When read in its entirety, Dambrot J.’s analysis makes it clear that he viewed the affidavits as part of a larger body of evidence that included, most significantly, 275 voter declarations.

[99] I agree with the Divisional Court that the affidavits and the declarations, taken together, can properly be used in the circumstances of this case to conclude that it is likely that only entitled electors voted and, therefore, that the result of the election was not affected by the omitted signatures. This conclusion is fortified by the City’s evidence that in a large number of cases there is a checkmark in the box on the VLCRF indicating that voter entitlement was, in fact, verified by the election officer, although the officer did not sign the form.

[100] For these reasons, the Divisional Court did not err in considering both the declarations on the VLCRFs and the after the fact affidavit evidence in determining whether the persons who voted were entitled to vote.

(3) Conclusion on the Main Appeal

[101] The omission of the election official’s signature on the 275 VLCRFs in issue was a procedural irregularity under s. 83(7) of the MEA. Consequently, the court must determine whether the saving provision in s. 83(6) can be applied. Here, that determination turns on whether the result of the election was affected by the omission. In making this determination, the substantive approach endorsed in Opitzrequires the court to decide, considering the evidence as a whole, whether the person casting the impugned ballot was entitled to vote. There is strong evidence in this case that, despite the omission of the election official’s signature, the persons who signed the 275 VLCRFs were nonetheless entitled to vote. Cusimano has therefore failed to discharge the onus on him to show that the result of the election was affected. The votes represented by the 275 VLCRFs should not be discounted. The saving provision under s. 83(6) of the MEA can be applied. The election in Ward 9 is valid.

D. Costs

[102] The costs appeal in this case requires this court to determine how the following costs should be apportioned:

1. The costs of the proceedings before the application judge;

2. The costs of the appeal to the Divisional Court;

3. The costs of the motion to have Augimeri added as a party;

4. The costs of the appeal before this court.

[103] I will first set out, in chronological order, the decision on costs before the application judge, Augimeri’s motion to intervene, the decision on costs before the Divisional Court, and the decision on costs in Opitz. I will then outline the parties’ positions before this court, and my analysis of their respective entitlements and liabilities. Finally, I will consider who should bear the burden of costs in this appeal.

(1) Facts

(a) Costs Before the Application Judge

[104] The application judge released her reasons in Cusimano’s favour on April 21, 2011. In her reasons on costs, released separately, she observed that the previous version of the MEA specifically provided that the costs of a recount were to be borne by the municipality, unless a court ordered otherwise: para. 13. Commenting on the jurisprudence under the existing MEA, which makes no provision for costs, she stated:

[T]he two cases addressing costs since the MEA was enacted – Di Biase v. Vaughan (City), [2007] O.J. No. 3516 (S.C.J.) and Goldie v. Brock (Township) – establish that the norm is partial indemnity costs, determined in accordance with customary costs principles: para. 11.

[105] On the basis of these two cases, the application judge awarded costs to Cusimano on a partial indemnity scale, payable by the City, amounting to $65,000.

(b) Augimeri’s Motion to Intervene as a Party

[106] Augimeri was not a party to the proceedings before the application judge. However, the process set out under the MEA for contesting elections clearly contemplates that the winning candidate will be a party to such an application. It is true that s. 83(3.1) of the MEAonly requires the applicant to serve the municipality with a copy of the application. However, s. 84(7) of the Act also allows a winning candidate whose election is challenged to disclaim the seat and thereby avoid any liability for costs of the application. A winning candidate would only be liable for costs if he or she were a party to the proceedings challenging the validity of the election.

[107] Augimeri was not a named respondent in Cusimano’s application (though his Notice of Application refers to her as “the respondent”), and she was never served. Though Augimeri became aware of Cusimano’s application before it was heard, she says she believed from her conversations with the City that it would protect her interests. She says she was told by the City that the application had nothing to do with her.

[108] When the application judge’s decision was issued in April 2011, the City initially indicated that it would not appeal the decision. Augimeri immediately began exploring her options for challenging the application judge’s order. Shortly thereafter, the City changed its position and elected to appeal, announcing that it was “clear that the decision would be appealed to the Divisional Court by other parties…” In July 2011, Augimeri did seek leave to be added as a party to the pending appeal, and to adduce fresh evidence.

[109] Prior to the hearing on Augimeri’s motion, Cusimano provided an offer to settle on a without costs basis under Rule 49 of the Rules of Civil Procedure, R.R.O. 1990, Reg. 194. The offer was that Augimeri would:

• intervene in the appeal to the Divisional Court as a friend of the court (rather than as a party);

• accept the record and file no additional evidence;

• not seek costs against Cusimano; and

• undertake not to delay the appeal, then scheduled to be heard by the Divisional Court on September 19 and 20, 2011.

[110] Augimeri declined Cusimano’s offer to settle.

[111] Lederman J. heard Augimeri’s motion on August 5, 2011. He ruled, at para. 23 of his decision, that Augimeri should be granted party status in the appeal because she had an interest in the subject matter of the proceeding and would be adversely affected by the decision.

[112] Because she did not have an opportunity to file evidence on the initial application, he allowed her to deliver an affidavit “to show that those voters listed on the VLCRFs were in fact eligible to vote”, and “to provid[e] evidence from voters establishing that the elections officials in question at each poll did take the necessary steps to verify identities, notwithstanding that the correct box on the form was not filled out, or signed”.

[113] The evidence Augimeri was seeking to introduce was responsive to the application judge’s reasons for judgment. At para. 59 of her reasons, the application judge commented that the information necessary to contact most of the voters whose VLCRFs were not signed was available, that it may have been possible for the City officials to contact those voters, verify their entitlement to vote, and sign the VLCRFs after the fact. If this had been done, the application judge noted that the City might have been in a position to establish that the statutory non-compliance had not affected the result of the election.

[114] Lederman J. left it to the panel of the Divisional Court hearing the appeal to determine the ultimate admissibility and the weight of Augimeri’s evidence, as well as the costs of the motion.

(c) Costs Before the Divisional Court

[115] Before outlining the Divisional Court’s decision on costs, I will briefly review two[^5] preliminary issues the Divisional Court was asked to determine at the outset of the appeal which are relevant to the costs portion of this judgment.

[116] First, Augimeri sought to raise a constitutional argument not raised in the original application, namely that the requirement under the MEA that a person who is not on the voters’ list complete a VLCRF in order to cast a vote was a constitutional violation under s. 2(b) and s. 15 of the Canadian Charter of Rights and Freedoms. In its preliminary reasons, reported at 2011 ONSC 5570, the Divisional Court declined to hear the constitutional argument. Jennings J., writing for the Court, noted that the application judge who allowed Augimeri to be added as a party had directed that her participation not unduly widen the focus of the appeal. Jennings J. also observed, at para. 2, that the record was “woefully inadequate” to support the proposed challenge.

[117] Second, Augimeri sought to file certain affidavit material before the Divisional Court in support of her position that the election was valid. The court admitted some of Augimeri’s affidavits but declined to admit others. The Divisional Court declined “to admit that part of her material relating to efforts made on her behalf to determine the eligibility of voters by canvassing them and having them sign forms,” primarily on the basis that these affidavits were hearsay and raised concerns about reliability: see the Divisional Court’s preliminary reasons, reported at 2011 ONSC 5611, and its judgment on the main appeal, at paras. 42-43.

[118] I will explore the significance of these preliminary matters to the issue of costs below.

[119] The parties’ submissions on costs before the Divisional Court were as follows. Cusimano asked that he be permitted to retain the $65,000 in costs awarded to him by the application judge, and that there be no costs of the appeal before the Divisional Court ordered on the grounds that it was a public interest litigation. The City sought $5,000 from Cusimano for its costs of the appeal. Augimeri sought costs of her appeal payable by the City either on a substantial or partial indemnity basis, including her costs of the motion to be added as a party.

[120] Sproat J., writing for the Court, set out several principles guiding the court’s exercise of discretion of whether to order costs, at paras. 12-31:

• While a statutory predecessor of the MEA provided for a presumption that the costs of a candidate in a recount case would be paid by the municipality, the MEA no longer contains that provision. There is therefore no presumption under the MEA that the City is obliged to pay the costs of all litigants.

• The MEA contemplates that costs awards could be made between challengers and elected officials in s. 84(7), which relieves a candidate from liability for costs if he or she disclaims all right to the office.

• Candidates are able to raise money through public fundraising to cover campaign expenses, and s. 67(2) of the MEA specifically includes “[e]xpenses relating to proceedings under section 83 (controverted elections)” as a category of campaign expenses. Candidates may therefore use campaign donations to defray litigation expenses.

• As a general rule, in these cases, a losing litigant will pay costs to a winning litigant.

• It is only in “truly extraordinary” cases that a successful party would be required to pay costs to an unsuccessful public interest litigant.

• The court may require a successful municipality to pay costs to another successful party, in appropriate circumstances.

[121] Applying these principles to the costs award made by the application judge, Sproat J. ordered that the $65,000 costs award against the City in favour of Cusimano be set aside. In doing so, he observed that Cusimano had a significant personal interest in the litigation, and that he could have chosen other means besides a court proceeding to ensure strict compliance with the MEA in the future.

[122] Sproat J. also dismissed the objection that setting aside the costs order made by the application judge would deter all but very wealthy candidates from contesting elections. He held that alternate sources of funding would likely be readily available for meritorious cases and that “[a]n arguable, interesting case is likely to attract able counsel who will take the case on other than…a straight hourly basis”: para. 34. Finally, he ordered Cusimano to bear his own costs of the appeal before the Divisional Court.

[123] In regard to Augimeri’s costs before the Divisional Court, Sproat J. held that she was a “blameless and successful” party entitled to her costs: para. 48. She also deserved at least partial credit for the City’s decision to appeal the application judge’s order.

[124] Having concluded that either the City or Cusimano would be liable for Augimeri’s costs, Sproat J. noted that the City was ultimately responsible for the election and that it was the mistakes of election officials that had resulted in so many unsigned VLCRFs. The City was also better equipped than a party in Cusimano’s position – now responsible for his own legal expenses – to bear the costs incurred by Augimeri, who was “simply in the wrong place at the wrong time and so embroiled in litigation”: para. 37. He therefore held, although the City was “successful” on the appeal, it would nonetheless bear the cost of Augimeri’s appeal.

[125] However, as there was no misconduct on the part of the City, there was no justification for awarding costs on a substantial, as opposed to a partial, indemnity basis. In addition, Sproat J. noted that Augimeri had expended considerable time and effort pursuing unmeritorious issues or duplicating the efforts of the City. He therefore reduced the partial indemnity costs award to Augimeri in two ways:

• First, he reduced the award by $10,000 to reflect costs incurred in relation to the constitutional challenge. Sproat J. noted, at para. 55, that the Divisional Court found the record “woefully inadequate” to support the challenge, and that Augimeri had been ordered by Lederman J. not to “unduly [widen] the focus, scope, or nature of the appeals.”

• Second, he reduced the award by an additional $15,000 to reflect costs incurred by Augimeri in obtaining voter affidavits ultimately ruled inadmissible by the Divisional Court. Sproat J. held, at para. 56, that Augimeri ought not to have undertaken such a canvass. He repeated the application judge’s comment, at para. 59 of her reasons, that such a canvass may have been possible for the City, but that it was “not something that the applicants could have done”. He felt her comment should have alerted Augimeri not to undertake such a canvass.

In the result, he ordered the City to pay Augimeri $40,000 for the costs of her appeal.

[126] In addition, Sproat J. considered the costs of Augimeri’s motion to be added as a party. He held that Augimeri had bettered that offer by being granted party status and leave to file additional evidence, and that Cusimano’s opposition to Augimeri’s motion’s was unreasonable and driven only by personal or political motives. He ordered Cusimano to pay Augimeri $32,500 for the costs of her motion. This amount was less than the amount of costs Augimeri sought on a substantial indemnity basis, $39,500, but greater than her partial indemnity costs of $26,500.

[127] Given the risk that Augimeri would not be able to collect her costs against Cusimano, he made the City jointly and severally liable for the costs of the motion, but ordered Cusimano to indemnify the City for any costs of the motion it paid.

(d) Opitz and Costs

[128] The decision on costs in Opitz has obvious relevance for this portion of the appeal. Although the Supreme Court was considering a federal election under the CEA, the same considerations are applicable in the municipal context. In addition, both the CEA and the MEA are similarly silent with respect to who should bear the cost of a challenge to the validity of an election.[^6]

[129] As in this case, in Opitz the application judge, Lederer J., held that a by-election would have to be held on account of irregularities in the electoral process. The runner-up candidate, who instigated the challenge before the application judge, sought costs from Elections Canada.

[130] Lederer J. began his decision by commenting that costs, while always in the discretion of the court, are generally awarded to the successful party. Proceedings challenging the validity of elections, however, are unique because there will often be no “successful” or “unsuccessful” party, at least in the traditional sense. Unless misconduct is alleged, both candidates and the governmental body administering the election stand on an equal footing in seeking the true result of the election.

[131] Lederer J. noted that neither electoral candidate had been the cause of, or implicated in, the ballots he set aside. Mistakes in the conduct of an election are inevitable. One of the checks on these errors is an application to set aside an election. Both the person who had been declared the victor and the person seeking to invalidate the result were in a similar position insofar as blame was concerned and each had the same potential claim for costs.

[132] Lederer J. further commented that in the application before him the Chief Electoral Officer had taken and maintained a position of neutrality. Counsel had merely advised the court of policy considerations and provided an understanding of the legislative framework governing elections. Lederer J. held that the Chief Electoral Officer’s neutrality was entirely appropriate, stating“[i]t is not for the person holding [that] office to take a position that favours one candidate to the detriment of others”: para. 7. However, as a consequence, the Chief Electoral Officer could not be said to be an unsuccessful party.

[133] For these reasons, Lederer J. held that each party should be responsible for its own costs. In so concluding, he acknowledged that coming to court can be expensive. However, he observed that proceedings of this nature are a matter of public interest, and that it is part of the collective responsibility of being a Canadian citizen to bring such matters to court. There was also no evidence before him that the parties did not have the necessary resources to bear their own costs.

[134] It does not appear that the application judge’s order on costs was appealed, and the Supreme Court does not address it. With respect to costs of the appeal before it, the majority held, without commenting further, that no award as to costs would be made “[i]n view of the circumstances of this case”: Opitz, at para. 133.

(2) The Parties’ Positions Respecting the Divisional Court’s Decision on Costs

[135] Before this court, Cusimano seeks costs of the proceedings before the application judge and the Divisional Court on the basis that this case is meritorious public interest litigation. He submits that he is in the same position as Augimeri; both are the victims of mistakes made by City officials. He also asks this court to overturn the award made against him by the Divisional Court for Augimeri’s costs on the intervention motion. He makes two basic submissions: a) the Divisional Court overstated her success on the motion and there was no reason to award her elevated costs; and b) he is unable to pay the costs award.

[136] Augimeri has cross-appealed the Divisional Court’s decision on costs. Augimeri submits that the Divisional Court erred in awarding her too low an amount for costs or alternatively that she should have received costs on a substantial indemnity basis. She submits that the Divisional Court incorrectly minimized the importance of the role she played. Sproat J. acknowledged that her stated intention to become a party and challenge the application judge’s decision seems to have been a determining factor in the City’s decision to appeal.

[137] She also argues that the reduction in costs of $10,000 to reflect her unsuccessful constitutional challenge was miscalculated. She says that a junior lawyer completed almost all the work done in relation to that argument, and the reduction should have been, at most, $5,500. In relation to the further reduction of $15,000 for the voter canvassing costs, Augimeri submits that the evidence she provided was justified by the application judge’s suggestion that such evidence would be helpful, the City’s failure to obtain that evidence itself, and the scope of Lederman J.’s order allowing her to intervene as a party.

[138] Finally, Augimeri submits that it is unrealistic to expect candidates to raise funds for their own legal expenses, given the finite pool of potential contributors and the restrictions in the MEA on maximum contributions and timeframes for fundraising.

[139] The City has not appealed the award of costs against it in favour of Augimeri. The City asks that Cusimano’s appeal of the Divisional Court’s costs order be dismissed, and that Augimeri’s cross-appeal as to costs also be dismissed. The City submits that the Divisional Court made no error in its decision on costs and that the court’s award should be upheld for the reasons it gave. In addition, the City raises a preliminary objection to Augimeri’s cross-appeal on the basis that she was required to obtain leave to appeal costs, but failed to raise any arguments in her factum as to why leave should be granted.

(3) Analysis of Costs for Augimeri’s Motion to Intervene and Proceedings Before the Application Judge and the Divisional Court

[140] I turn now to my analysis. As I will explain below, subject to the impact Opitz may have on costs awards in future cases of this nature, I agree generally with the principles enunciated by the Divisional Court. However, with respect to both the application judge’s order of costs in favour of Cusimano and the quantum of Augimeri’s costs before the Divisional Court, I believe the Divisional Court erred in applying those principles.

[141] I will first consider Cusimano’s request that we award him costs of all the proceedings and dismiss the Divisional Court’s order against him for the costs of Augimeri’s motion. I agree with Cusimano that he should receive a portion of his costs throughout. I see no grounds, however, for disturbing the costs award against him for Augimeri’s motion.

[142] Although Cusimano initiated the court proceedings, the Divisional Court acknowledged in its reasons on costs that it was reasonable for him to do so. Indeed, the narrow majority by which the Supreme Court resolved Opitz, a case in which nearly identical legislative language was at issue, amply supports this conclusion.

[143] In refusing Cusimano’s request for costs, the Divisional Court acknowledged that costs could be awarded against a blameless, successful party such as the City in “truly extraordinary” cases – for example, where the issue litigated was of national and international significance – but held that this case was not one of them.

[144] I agree that the City may be required to pay costs to an unsuccessful applicant in “truly extraordinary” cases. I disagree that this case is not one of them. In my opinion, the Divisional Court erroneously minimized the public interest component of this case.

[145] The Divisional Court held that “[a]t its highest, Cusimano…[was] endeavouring to hold election officials to strict compliance with the technical requirements of the M.E.A.” and that Cusimano might have pursued the same public interest objective by alternate means: para. 33. For example, Sproat J. suggested Cusimano might have brought “perceived failings in the training and diligence of election officials to the attention of the City Clerk, the Mayor and members of Council, the media and the public”: para 33.

[146] The public interest went far beyond this. These proceedings dealt with whether the omission of the election official’s signature on the VLCRFs, a procedural irregularity, “did not affect the result of the election” of Augimeri as a City Councillor pursuant to s. 83(6) of the MEA. The interpretation of these words was of great importance to all municipal elections held in Ontario, as the intervener City of Mississauga’s factum stressed.

[147] The question of whether the strict procedural approach or the substantive approach should be adopted was also of importance throughout Canada, as indicated by the appeal in Opitz, a case of national interest concerning the equivalent phrase in the CEA. Notably, on the important issue of onus of proof, the majority of the Supreme Court agreed with and cited the Divisional Court’s reasons in this case, and the approach the majority adopted closely paralleled that of the Divisional Court.

[148] The case before us is thus clearly a “truly extraordinary” case.

[149] That said, the principles underpinning the Divisional Court’s decision on costs may have been overtaken by Opitz. As outlined above, the application judge places both the successful and unsuccessful party and the election administrator (who was playing a neutral role in the application) on an equal footing. In a case like Opitz, where there is no allegation of misconduct and when inability to pay costs has not been shown, he requires each party to bear his or her own costs. The Supreme Court also elected not to order costs of the appeal before it (though its justification for doing so is unclear). On a go forward basis, when the City has taken a position of neutrality, it may therefore be appropriate for the parties disputing the validity of an election in good faith on the basis of procedural errors to be on an equal footing with respect to costs.

[150] I cannot, however, place Cusimano and Augimeri on an equal footing by awarding no costs of the appeal before the Divisional Court. As I have indicated, the City did not appeal the partial award of costs against it in favour of Augimeri. Thus, this court has no jurisdiction to make an order of no costs respecting the costs incurred before the Divisional Court. The only way in which Cusimano and Augimeri can be placed on an equal footing would be for this court to make an award of costs on a partial indemnity scale in Cusimano’s favour.

[151] In any event, a “no costs” order would not be appropriate with respect to the costs incurred prior to the Supreme Court’s decision in Opitz, having regard to the existing jurisprudence at the time of the Divisional Court’s decision, the City’s active participation throughout the proceedings, and the fact that a costs award would not have been outside the City’s expectations. I also note that bearing Cusimano’s costs will cost the City much less than conducting a by-election. As the Divisional Court noted, the City of Mississauga’s evidence indicated it budgeted $445,000 to conduct one such by-election.

[152] For these reasons, I agree with Cusimano that, although he was not successful before the Divisional Court, the City should pay his costs of the application fixed by the application judge at $65,000 and his reasonable costs before the Divisional Court on a partial indemnity basis. In his factum, Cusimano did not state the amount of costs he was seeking for the appeal before the Divisional Court. Having regard to the costs awarded to Augimeri by the Divisional Court, the fact that Cusimano was a respondent (and therefore expected to incur somewhat lower costs than the appellant), and the costs outline Cusimano filed on the appeal to this court, I would award him costs of his appeal to the Divisional Court on a partial indemnity basis of $30,000, inclusive of all disbursements and taxes.

[153] I must now deal with Cusimano’s appeal of the costs awarded against him by the Divisional Court on Augimeri’s intervention motion to be added as a party. Cusimano submits that the Divisional Court overstated Augimeri’s success on the motion, that rule 49.10 ought not to be applied, and that there was no basis on which to award Augimeri elevated costs.

[154] The Divisional Court did not err in making the award it did and I would maintain its award of costs to Augimeri fixed at $32,500. For the reasons given by the Divisional Court, Augimeri clearly bettered Cusimano’s Rule 49 offer to settle and this fact entitles her to costs on a higher scale than partial indemnity costs. Furthermore, as the Divisional Court observed, there was no public interest aspect to Cusimano’s opposition to the motion.

[155] With respect to Augimeri’s cross-appeal of the Divisional Court’s award of costs, the City submits that Augimeri was required to do more than simply request that, if necessary, leave to appeal be granted in her Notice of Cross-Appeal, as she did. I do not agree. Rule 61.07 of the Rules of Civil Procedure governs cross-appeals to this court. Rule 61.07(1.2) provides that a “respondent shall obtain leave to appeal in the manner provided by subrule…61.03.1(18)…if the cross-appeal is taken under a statute that requires leave for an appeal.” The City is correct in its submission that the Divisional Court’s decision on costs falls within the ambit of s. 133(b) of the Courts of Justice Act, R.S.O. 1990, c. C. 43, which is a statute requiring leave for an appeal within the meaning of rule 61.07(1.2). Augimeri was, therefore, required to obtain leave in the manner provided by rule 61.03.1(18). Rule 61.03.1(18)(a) stipulates that “the request for leave to appeal shall be included in the notice of…. cross-appeal…” This is precisely what Augimeri did. Augimeri therefore complied with her obligations under the Courts of Justice Act and the Rules of Civil Procedure.

[156] My decision as to whether or not to grant leave is informed by this court’s statement in Brad-Jay Investments Ltd. v. Szijjarto (2006), 2006 CanLII 42636 (ON CA), 218 O.A.C. 315 (C.A.), leave to appeal to the S.C.C. refused, [2007] S.C.C.A. No. 92. In that case, the court stated, at para. 21, “[l]eave to appeal a costs order will not be granted save in obvious cases where the party seeking leave convinces the court there are ‘strong grounds upon which the appellate court could find that the judge erred in exercising his discretion.’” As I explain below, I am of the opinion that this appeal is one such case.

[157] However, I do not agree with Augimeri that Cusimano’s conduct in these proceedings is such that elevated costs are warranted. As noted by Augimeri, normally substantial indemnity costs are awarded “where there has been reprehensible, scandalous or outrageous conduct on the part of one of the parties”: Young v. Young, 1993 CanLII 34 (SCC), [1993] 4 S.C.R. 3, at p. 134. Cusimano’s conduct, though questionable at times, did not rise to this level.

[158] I also do not accept Augimeri’s complaint regarding the Divisional Court’s reduction of her costs on account of her unsuccessful constitutional argument. In my opinion, her submission is completely met by the Divisional Court’s comment, at para. 57 of its reasons, that the lower figure for these costs in Augimeri’s costs outline assumed no time was spent by either the junior lawyer or senior counsel in preparing for the appeal. I would also note that the court’s award of costs, and for that matter its reduction of costs, is not an exact mathematical calculation. Save as indicated below, I would not increase the quantum of costs awarded to Augimeri.

[159] The exception I would make with respect to Augimeri’s costs relates to the voter canvass issue, for which her costs award was reduced by $15,000. She submits that it was reasonable for her to incur the expense to obtain such evidence because the application judge suggested that it was something the City could have done to validate the election after the fact.

[160] I agree with Augimeri. Furthermore, as I noted above, Augimeri was permitted by the motion judge “to provid[e] evidence from voters establishing that the elections officials in question at each poll did take the necessary steps to verify identities” and “to show that those voters listed on the VLCRFs were in fact eligible to vote”. Short of doing what she did, I do not see how Augimeri could have obtained the evidence she was given permission to present.

[161] The Divisional Court observed that voter secrecy was not breached by Augimeri’s efforts. Its concern that some persons who refused to cooperate with Augimeri’s volunteers and that those persons who provided answers to them “may have provided answers that were influenced one way or the other by the partisanship of the interrogators” is speculation. In the circumstances the Divisional Court’s criticism of Augimeri for obtaining this evidence is misplaced.

[162] While the evidence obtained from the voters may have been hearsay, in the sense that a representative of Cusimano was not present when the evidence was obtained, Cusimano could have cross-examined on the sworn affidavits had he chosen to do so. He did not. The affidavits provided direct evidence as to what happened at the polls, evidence that Cusimano submitted was lacking. Although, in the end, the Divisional Court had the discretion not to admit this evidence, it was reasonable in the circumstances for Augimeri to obtain it. The Divisional Court erred in penalizing Augimeri in costs for doing so. I would therefore allow the cross-appeal to the extent of $15,000.

(4) Costs of this Appeal

[163] I must now deal with the costs of this appeal. Cusimano seeks costs of the appeal on a partial indemnity basis in the amount of $30,257.50. Augimeri seeks costs of $65,239.45 on a partial indemnity basis. The City’s position is that, based on the decision in Opitz, no costs should be awarded and each party should bear his or her own costs.

[164] Unlike the situation in Opitz, where the Chief Electoral Officer took and maintained a position of neutrality, the City took a position in favour of maintaining the election result before the application judge and, after some hesitation, took on the main role of appellant before the Divisional Court. The City also played an active role as a respondent in the appeal before us. It is therefore too late, in the light of the role the City has played in these proceedings, for the City to submit that there should be no costs award made against it on the basis of Opitz.

[165] With respect to Cusimano’s position, the decision in Opitz was released by the Supreme Court on October 25, 2012, shortly before this appeal was heard on November 8, 2012. The outcome of this appeal should have been clear to Cusimano. He did not, however, abandon his appeal. All parties were therefore forced to undertake the additional expense of filing supplementary facta and appearing for a full day of argument.

[166] Because of this, I am inclined not to award Cusimano any costs of the appeal before us. However, I recognize that he incurred costs in preparing for this appeal prior to the Supreme Court’s decision in Opitz and that he was partially successful in his appeal on the issue of costs. Balancing these factors, I am therefore of the opinion that a more measured response is appropriate. I would order costs of this appeal payable by the City in favour of Cusimano, fixed in the amount of $10,000, inclusive of all taxes and disbursements.

[167] Augimeri is entitled to her costs of the appeal on a partial indemnity basis, but not in the amount of $65,239.45 she is seeking. I agree with the City that the costs being sought are “beyond the amount any party would reasonably expect to pay.” I also agree with the City that one counsel fee, and not three, is appropriate. That said, Augimeri’s factum, supplementary factum, and oral submissions were of assistance. She also achieved partial success on her cross-appeal with respect to costs. I would, therefore, order costs payable by the City, fixed in the amount of $30,000, inclusive of all taxes and disbursements.

(5) Summary of Costs

[168] Based on Opitz, on a go forward basis, when the validity of an election is challenged on account of procedural errors rather than misconduct, the City adopts a position of neutrality, and there is no indication that the parties do not have the necessary resources to bear their own costs, the parties may generally expect to bear their own costs. In the litigation before us, however, Opitz cannot be applied, nor would it be appropriate to do so.

[169] Cusimano is entitled to have his costs before the application judge, fixed in the amount of $65,000, restored. He is also entitled to his costs before the Divisional Court, fixed in the amount of $30,000, and his costs of this appeal, fixed in the amount of $10,000, payable by the City.

[170] Augimeri is entitled to have her costs of the intervention motion as ordered by the Divisional Court. She is also entitled to her costs of the appeal before the Divisional Court of $40,000 increased by $15,000, to reflect the extent to which her cross-appeal is allowed. In all other respects, her cross-appeal is dismissed.

[171] Augimeri is entitled to have her costs of the appeal before us paid by the City. These costs have been fixed in the amount of $30,000 inclusive of all taxes and disbursements.

Released: Dec. 24, 2012 “Karen M. Weiler J.A.”

“KMW” “I agree R.G. Juriansz J.A.”

“I agree M. Tulloch J.A.”

[^1]: Cusimano was joined in his application by Michael Sullivan, the unsuccessful candidate by 56 votes for the position of School Trustee for Ward 4 of Toronto District School Board. Sullivan also participated in the proceeding before the Divisional Court, but was not a party to the appeal before this court. Therefore, I will confine these reasons to the Ward 9 election.

[^2]: According to the Divisional Court, there were some 1100 VLCRFs submitted in Ward 9. 374 were not signed by an election official. Only 286 of those VLCRFs were used to add a voter to the voters’ list. Of these 286, all but 11 had signed declarations by the voter seeking to be added.

[^3]: Persons prohibited from voting under s. 17(3) are inmates serving a sentence in a penal or correctional institution, a corporation, an executor, trustee or any other person acting in a representative capacity, except as a voting proxy in accordance with s. 44 of the MEA, and persons convicted of a corrupt practice under the MEA, if the election is less than five years after the date of the election in respect of which they were convicted.

[^4]: According to Cusimano, 108 VLCRFs that were not signed by an election official did have a confirming checkmark. However, not all of these VLCRFs were completed for the purpose of adding a voter to the voters’ list.

[^5]: Augimeri asked the Divisional Court to consider a third preliminary issue, which she raised again in her cross-appeal before this court. She submits that Cusimano lacked standing to bring the application in the first place. Without determining whether Cusimano had standing or not, Dambrot J. held, at para. 45, that it would be contrary to the public interest to set aside the decision of the application judge on such a “technical basis”. I too decline to address this issue, as I have no decision at first instance to review. I would note, however, that Augimeri’s request to adduce evidence of misconduct on the part of Cusimano was expressly denied by Lederman J. when he granted her leave to be added as a party to the appeal.

[^6]: Section 526(1) of the CEA does require an application under s. 524 to be accompanied by security for costs in the amount of $1,000. However, it is clear from the decision in Opitz that this provision is not determinative of who will ultimately bear the costs of a proceeding challenging the validity of an election. Section 306 of the CEA, for example, expressly provides that if a winning candidate prevails on a recount that has been ordered pursuant to the act, the person who applied for the recount shall pay the winning candidate’s costs. Section 310 permits a candidate to make an application to the Chief Electoral Officer for reimbursement of his or her costs in respect of a recount.