IN THE SUPREME COURT OF BRITISH COLUMBIA
Henry v. Canada (Attorney General),
2010 BCSC 610
The Attorney General of Canada and
The Chief Electoral Officer of Canada
British Columbia Civil Liberties Association
Before: The Honourable Madam Justice Lynn Smith
Reasons for Judgment
Counsel for Plaintiffs:
J. L. Quail and C. Allevato
Counsel for The Attorney General of Canada:
Wruck, A. J. Semple
Counsel for The Chief Electoral Officer of Canada:
G. Whitehall, Q.C.
Counsel for Intervenor:
Place and Date of Hearing:
June 1-5, 8-11, 15, and
October 26-27, 2009
Place and Date of Judgment:
May 3, 2010
 The issue in this case is the constitutional validity of the voter identification rules in federal elections.
 The plaintiffs challenge recent amendments (effective in June 2007) to the Canada Elections Act, S.C. 2000, c. 9 [Act], as infringing the right to vote guaranteed under s. 3 of the Canadian Charter of Rights and Freedoms, Part I of the Constitution Act, 1982, being Schedule B to the Canada Act 1982 (U.K.), 1982, c. 11 [Charter].
 Section 3 of the Charter guarantees to each Canadian citizen these rights:
3. Every citizen of Canada has the right to vote in an election of members of the House of Commons or of a legislative assembly and to be qualified for membership therein.
 The net effect of the challenged amendments is to require registered electors to prove their identity and residence by prescribed means before they are able to cast a ballot in a federal election. Such proof of identity and residence was not previously necessary.
 The gravamen of the plaintiffs’ challenge is that the new voter identification requirements impede or limit the exercise of the right to vote by those persons who do not have standard documentary proof of their identity and residence available to them.
 The plaintiffs seek a declaration that a number of provisions of the Act are contrary to s. 3 of the Charter, cannot be saved by s. 1 of the Charter and, by virtue of s. 52(1) of the Constitution Act, 1982 [Constitution Act, 1982], are void and of no effect. The plaintiffs also seek an order that the defendant Chief Electoral Officer of Canada (the “CEO”) not apply the voter identification rules set out in the impugned amendments in elections for the Parliament of Canada.
 The Attorney General of Canada (“Canada”) responds that the identification requirements do not limit or interfere with the right to vote as the plaintiffs assert. Canada’s position is that, by strengthening the basic fairness and integrity of the electoral process, the amendments protect and enhance the right to vote. In the alternative, Canada says, the provisions are reasonable limits on the s. 3 rights within the meaning of s. 1 of the Charter.
 The CEO does not take a position on the disposition of the plaintiffs’ constitutional challenge.
 The intervenor, the British Columbia Civil Liberties Association (the “BCCLA”), took a limited part in the proceedings and supported the plaintiffs’ challenge.
 The trial was heard summarily, under Rule 18A, on affidavit evidence.
 Prior to the legislative amendments at issue in these proceedings, registered electors did not have to produce identification in order to vote. A registered elector needed only to state his or her name and address to the clerk at the polling station. The clerk would confirm the elector’s name on the list of electors and cross it off, and the elector would then be permitted to vote. This procedure was set out in s. 143 of the Act prior to amendment (S.C. 2000, c. 9).
 Subsection 143(2) was subject to s. 144, which provided that in the event there were doubts about an elector’s identity, poll staff could request proof of identity and residence or have the elector swear an oath.
 If an individual was not on the list of electors, he or she could register on polling day with proof of identity and residence.
 Section 535 of the Act requires that the CEO, as soon as possible after a general election, submit a report to the House of Commons setting out any amendments that are seen as desirable for the better administration of the Act. In September 2005, the CEO’s report with respect to the June 2004 general election was tabled in the House of Commons. The CEO’s recommendations addressed operational issues related to electors, broadcasting, financial matters and technical amendments. The House of Commons did not have an opportunity to review the report and its recommendations before the next general election was called.
 In 2006, the House of Commons Standing Committee on Procedure and House Affairs (the “Committee”) considered the CEO’s report. During its study, the all-party Committee heard testimony from the CEO, the Privacy Commissioner and representatives from the four major federal political parties. In June of that year, the Committee tabled its own report entitled “Improving the Integrity of the Electoral Process: Recommendations for a Legislative Change” (the “Thirteenth Report”), in which it recommended various administrative and operational improvements. The Committee also expressed its firm view that electoral fraud and the means for reducing the potential for such fraud were issues that had to be addressed to preserve the integrity of the electoral process. At 25-27 of its report, the Committee wrote:
Many Canadians have expressed concern about the potential for fraud and misrepresentation in voting. Members of the Committee share this concern. While we have no means of knowing how widespread this problem is, the fact that it exists undermines the integrity of the electoral process. ...
At present, there is no requirement that voters show any identification before being able to vote, so long as their names are on the list. In our society, most important activities require that an individual be able to furnish some form of proper identification, often with a photograph. In the case of voting, we do not believe that it would be unreasonable to impose a similar requirement. Moreover, it would bring home to voters the seriousness and public importance of what they are about to do: exercise a valued and fundamental democratic right.
Traditionally, Canada has tried to make voting as easy as possible, but if confidence in the system is undermined, it becomes necessary to make changes. Obviously, it is not our intention to impose any measures that would discourage voting, nor do we want to make voting more difficult than necessary. The credibility and legitimacy of the system, however, require that procedures be adopted to ensure that only those persons who are entitled to vote do so, and that they are who they say they are. This is essential to preserve the integrity in the electoral system.
The Committee’s concerns about adequate proof of identification and residency, and proof of a person’s eligibility to vote were shared by the witnesses who appeared before the Committee and who made written submissions. All of the parties currently represented in the House of Commons support a more effective method of ensuring voter identification, including photo identification, with alternatives available for persons who are unable to furnish the required identification. The Committee wishes to make it clear that voter information cards should not entitle a person to cast a ballot. This was never the intent of these cards, although, in practice, they appear to often be used for this purpose. Given the problems with how they are delivered or disposed of, this must be clarified. Members of the Committee are also disturbed that voters seem to be able to use magazine subscription labels and utility bills to establish their identity. Although such documents may be useful in establishing one’s address, they are no substitute for adequate identification.
 The Committee also referred to the model of Québec legislation requiring voter identification, which had been in effect since 1999.
 The Committee, in its unanimous report, proposed that potential voters be required to provide one piece of government-issued identification with a photograph and residential address before being permitted to vote. If an elector did not have such identification, then he or she would be required to provide two other pieces of acceptable identification to establish identity and residence. The Committee proposed a vouching process for prospective voters who did not have two pieces of acceptable identification, and expressed its concern about “serial vouching”. It also expressed its concern about the high number of election day registrations and noted that the CEO had agreed to conduct an audit in one riding (Trinity-Spadina) where there were approximately 10,000 voters who registered on voting day.
 In October 2006, the Government tabled its formal response to the Thirteenth Report. In that response, the Government indicated that it would be introducing a bill which would implement most of the Committee’s recommendations. With respect to voter identification and related issues in particular, it stated, in part:
The bill will implement a number of Committee recommendations that will assist in reducing the potential for fraud or error that could impair the integrity of the voting system by allowing voting by individuals who are not entitled to vote. The Committee heard evidence from political parties and Committee members of instances in which non-citizens have voted. Concerns were also raised about voter information cards that are sent by Elections Canada being left in bundles in apartment buildings, leaving open the opportunity for individuals seeking to defraud the system to use these cards as evidence of their entitlement to vote.
The nature of our voting system makes it difficult to conclusively determine how widespread voter fraud may be. However, each instance that comes to light affects public confidence in the integrity of our electoral process and even a small degree of fraud can affect the results of a close election. In addition, some individuals may vote out of a mistaken belief that they are eligible to do so rather than due to any malicious intent. The proposed legislative reforms will go a significant way in preventing these opportunities from arising[.]
 The government followed its formal response with the introduction on October 24, 2006, of Bill C-31, An Act to amend the Canada Elections Act and the Public Service Employment Act, S.C. 2007, c. 21 [Bill C-31]. Bill C-31 implemented the majority of the Committee’s recommendations for improving the integrity of the electoral process. Of relevance to the present proceedings, it amended the Act to require all voters to present proof of identity and residence or to swear an oath to be vouched for by another elector before receiving a ballot in a federal election.
 Bill C-31 was studied in depth by both House of Commons and Senate Parliamentary Committees. During Committee hearings, a number of interested parties, including representatives of the British Columbia Public Interest Advocacy Centre (counsel for the plaintiffs in the present action) and of the intervenor BCCLA made submissions opposing the amendments due to what they considered would be their impact on certain categories of persons.
 When Bill C-31 was introduced, the Committee member from the New Democratic Party proposed amendments that would have allowed an elector who did not have the required identification to vote with a statutory declaration of identity and residence alone. The Committee did not pass these proposed amendments. The New Democratic Party again unsuccessfully proposed amendments in the House of Commons, and when Bill C-31 came to a final vote in the House of Commons, members of that party voted against it. The Bill passed by a vote of 249 to 27 with the support of the three other parties and the independent Member of Parliament.
 Bill C-31 received Royal Assent on June 22, 2007.
 Shortly after the voter identification provisions came into force, Elections Canada notified the Government that approximately one million Canadians could have difficulty satisfying the proof of residence requirements because the areas in which they lived were not assigned residential addresses or had no mail delivery, and their identification documents were therefore likely to record a mailing, as opposed to residential, address. This was particularly a concern in rural and northern areas of the country. As a consequence, Bill C-18, An Act to amend the Canada Elections Act (verification of residence), S.C. 2007, c. 37, was enacted to remedy that matter [Bill C-18]. Bill C-18 received Royal Assent on December 14, 2007.
 The following sections of the current Act are said to be unconstitutional: ss. 143(2)‑(7), 148.1(1), 161(1), (6) and (7), and 169(2) (the “impugned provisions”). The text of these sections is as follows:
143. (2) If the poll clerk determines that the elector’s name and address appear on the list of electors or that the elector is allowed to vote under section 146, 147, 148 or 149, then, subject to subsection (3), the elector shall provide to the deputy returning officer and the poll clerk the following proof of his or her identity and residence:
(a) one piece of identification issued by a Canadian government, whether federal, provincial or local, or an agency of that government, that contains a photograph of the elector and his or her name and address; or
(b) two pieces of identification authorized by the Chief Electoral Officer each of which establish the elector’s name and at least one of which establishes the elector’s address.
(2.1) For greater certainty, the Chief Electoral Officer may authorize as a piece of identification for the purposes of paragraph (2)(b) any document, regardless of who issued it.
(2.2) For the purposes of paragraph (2)(b), a document issued by the Government of Canada that certifies that a person is registered as an Indian under the Indian Act constitutes an authorized piece of identification.
(3) An elector may instead prove his or her identity and residence by taking the prescribed oath if he or she is accompanied by an elector whose name appears on the list of electors for the same polling division and who
(a) provides to the deputy returning officer and the poll clerk the piece or pieces of identification referred to in paragraph (2)(a) or (b), respectively; and
(b) vouches for him or her on oath in the prescribed form.
(3.1) If the address contained in the piece or pieces of identification provided under subsection (2) or paragraph (3)(a) does not prove the elector’s residence but is consistent with information related to the elector that appears on the list of electors, the elector’s residence is deemed to have been proven.
(3.2) Despite subsection (3.1), a deputy returning officer, poll clerk, candidate or candidate’s representative who has reasonable doubts concerning the residence of an elector referred to in that subsection may request that the elector take the prescribed oath, in which case his or her residence is deemed to have been proven only if he or she takes that oath.
(4) If the deputy returning officer is satisfied that an elector’s identity and residence have been proven in accordance with subsection (2) or (3), the elector’s name shall be crossed off the list and, subject to section 144, the elector shall be immediately allowed to vote.
(5) No elector shall vouch for more than one elector at an election.
(6) An elector who has been vouched for at an election may not vouch for another elector at that election.
(7) The Chief Electoral Officer shall publish each year, and within three days after the issue of a writ, in a manner that he or she considers appropriate, a notice setting out the types of identification that are authorized for the purpose of paragraph (2)(b). The first annual notice shall be published no later than six months after the coming into force of this subsection.
148.1 (1) An elector who fails to prove his or her identity and residence in accordance with subsection 143(2) or (3) or to take an oath otherwise required by this Act shall not receive a ballot or be allowed to vote.
161. (1) An elector whose name is not on the list of electors may register in person on polling day if the elector
(a) provides as proof of his or her identity and residence the piece or pieces of identification referred to in paragraph 143(2)(a) or (b), respectively, which piece or one of which pieces must contain an address that proves his or her residence; or
(b) proves his or her identity and residence by taking the prescribed oath, and is accompanied by an elector whose name appears on the list of electors for the same polling division and who
(i) provides the piece or pieces of identification referred to in paragraph 143(2)(a) or (b), respectively, which piece or one of which pieces must contain either an address that proves his or her residence or an address that is consistent with information related to him or her that appears on the list of electors, and
(ii) vouches for him or her on oath in the prescribed form, which form must include a statement as to the residence of both electors.
(6) No elector shall vouch for more than one elector at an election.
(7) An elector who has been vouched for at an election may not vouch for another elector at that election.
169. (2) An elector shall not be registered unless he or she
(a) provides as proof of his or her identity and residence the piece or pieces of identification referred to in paragraph 143(2)(a) or (b), respectively, which piece or one of which pieces must contain an address that proves his or her residence; or
(b) proves his or her identity and residence by taking the prescribed oath, and is accompanied by an elector whose name appears on the list of electors for the same polling division and who
(i) provides the piece or pieces of identification referred to in paragraph 143(2)(a) or (b), respectively, which piece or one of which pieces must contain either an address that proves his or her residence or an address that is consistent with information related to him or her that appears on the list of electors, and
(ii) vouches for him or her on oath in the prescribed form, which form must include a statement as to the residence of both electors.
 In order to understand the impugned provisions and the plaintiffs’ challenges to them, it is necessary to look at the voter identification requirements in the context of the electoral system as a whole.
 Federal elections are conducted under the aegis of Elections Canada and the CEO.
 Entitlement to vote in a federal election in Canada is established through the creation of a list of electors for each electoral district. The lists of electors for the various electoral districts are generated, in part, from the Register of Electors. Under s. 44 of the Act, the CEO maintains a Register of Electors which contains the name, gender, date of birth, civic and mailing addresses, and a unique, randomly generated identifier for each elector on the list, who is registered to vote at his or her place of ordinary residence.
 Individuals who were eligible electors and who were enumerated in 1997 when the Register of Electors was created were placed on the Register of Electors at the outset. Information regarding potential electors is obtained from data received from provincial, territorial and federal sources with which Elections Canada has signed agreements, specifically, the Canada Revenue Agency, and most provincial and territorial registrars of motor vehicles. Potential electors must first confirm their Canadian citizenship and give their consent to Elections Canada before being added to the Register. Further, electors are added from lists of electors prepared for provincial or territorial elections, and from provincial registers of electors, regarding which Elections Canada has sharing agreements.
 Aside from these indirect registration methods, the Act provides five other methods by which electors may become registered: (1) an elector can request to be added to the Register of Electors outside of an election period (s. 49); (2) an elector can be added to the list of electors during the revision period of an election by contacting the returning officer (s. 101(1)); (3) an elector can be added to a list of electors during an election period through targeted revision (i.e., a process of door-to-door confirmation of the accuracy of the list of electors in areas identified by the returning officer) (s. 101(1)(d)); (4) if an elector is already registered to vote in an electoral district but has moved within the district, he or she can be added to the electoral list for the new polling division during the revision period of an election by contacting the returning officer (s. 101(6)); and (5) an elector can be added to a list of electors on polling day or at an advance poll (ss. 161, 169).
 The Act establishes a voting system in which Canada is divided geographically into electoral districts, each of which returns one Member of Parliament to the House of Commons. There are presently 308 electoral districts. Electoral districts are divided into polling divisions which, for the most part, contain a minimum of 250 electors. Electors are generally required to vote at the polling station designated for their polling division, whether at an advance poll or on election day. Certain exceptions exist to accommodate electors with disabilities. As well, a prescribed special voting procedure permits electors who cannot or do not wish to vote at a polling station (for example, Canadian citizens temporarily living away from their electoral districts, members of the Canadian Forces and incarcerated persons) to vote by mail or in person at the office of a returning officer.
 The basic scheme for voting qualification is set out in the following provisions of the Act:
2. (3) For the purposes of this Act, satisfactory proof of an elector’s identify and satisfactory proof of residence are established by the documentary proof of the elector’s identity and residence that is prescribed by the Chief Electoral Officer.
3. Every person who is a Canadian citizen and is 18 years of age or older on polling day is qualified as an elector.
4. The following persons are not entitled to vote at an election:
(a) the Chief Electoral Officer;
(b) the Assistant Chief Electoral Officer; and
(c) every person who is imprisoned in a correctional institution serving a sentence of two years or more. [no longer in effect]
5. No person may
(a) vote or attempt to vote at an election knowing that they are not qualified as an elector or not entitled to vote under section 4; or
(b) induce another person to vote at an election knowing that the other person is not qualified as an elector or not entitled to vote under section 4.
6. Subject to this Act, every person who is qualified as an elector is entitled to have his or her name included in the list of electors for the polling division in which he or she is ordinarily resident and to vote at the polling station for that polling division.
7. No elector who has voted at an election may request a second ballot at that election.
8. (1) The place of ordinary residence of a person is the place that has always been, or that has been adopted as, his or her dwelling place, and to which the person intends to return when away from it.
(2) A person can have only one place of ordinary residence and it cannot be lost until another is gained.
(3) Temporary absence from a place of ordinary residence does not cause a loss or change of place of ordinary residence.
(4) If a person usually sleeps in one place and has their meals or is employed in another place, their place or ordinary residence is where they sleep.
(5) Temporary residential quarters are considered to be a person’s place of ordinary residence only if the person has no other place that they consider to be their residence.
(6) A shelter, hostel or similar institution that provides food, lodging or other social services to a person who has no dwelling place is that person’s place of ordinary residence.
9. If the rules set out in section 8 are not sufficient to determine the place of ordinary residence, it shall be determined by the appropriate election officer by reference to all the facts of the case.
 Section 2(3) of the Act delegates to the CEO the power to prescribe what is satisfactory documentary proof of an elector’s identity and residence. Section 3 of the Act provides that every person who is a Canadian citizen and 18 years of age or older on polling day is qualified as an elector. The only persons whom s. 4 disqualifies from voting are the CEO and the Assistant CEO. Although s. 4 still includes reference to persons imprisoned in a correctional institution serving a sentence of two years or more, that part of the provision is inoperative as a result of Sauvé v. Canada (Chief Electoral Officer), 2002 SCC 68,  3 S.C.R. 519 [Sauvé No. 2], which held that the disqualification unjustifiably infringed s. 3 of the Charter.
 Section 6 of the Act entitles every elector “to have his or her name included in the list of electors for the polling division in which he or she is ordinarily resident and to vote at the polling station for that polling division”. This is meant to ensure that each elector can only be registered to vote at one place and to vote once. Section 7 affirms that no elector who has voted in an election may request a second ballot in the same election. A voter must vote only in his or her own name (s. 167(1)(a)).
 Section 8 of the Act contains comprehensive rules for establishing an elector’s place of ordinary residence for the purpose of voting. Place of ordinary residence determines not only the electoral district but also the polling division to which a person is assigned. As noted, assignment to a polling division is necessary in order to vote. Section 9 provides that an election officer will determine the place of ordinary residence on all the facts of the case if it is not determined by the rules under s. 8. A permanent or complete civic address is not required in order to vote (for example, a post office box or “general delivery” to a town’s post office can suffice).
 The impugned provisions modified the rules prescribed by the Act for proving identity and address when voting, and when registering to vote, either on polling day or at an advance poll. Their net effect is discussed next.
 As earlier mentioned, the Act previously required only that a registered elector give his or her name to the clerk at the polling station in order to vote. As a result of the amendments in question, an elector must now also prove his or her identity and residence before being entitled to receive a ballot.
 An elector may provide identity and residence in a number of ways. First, an elector may produce a piece of government-issued identification that contains the bearer’s photograph, name and address. Alternatively, an elector may provide two pieces of identification that have been authorized by the CEO, each of which establishes the elector’s name and at least one of which establishes the elector’s address. These requirements are set out in s. 143(2) of the Act, which I repeat for ease of reference:
143. (2) If the poll clerk determines that the elector’s name and address appear on the list of electors or that the elector is allowed to vote under section 146, 147, 148 or 149, then, subject to subsection (3), the elector shall provide to the deputy returning officer and the poll clerk the following proof of his identity and residence:
(a) one piece of identification issued by a Canadian government, whether federal, provincial or local, or an agency of that government, that contains a photograph of the elector and his or her name and address; or
(b) two pieces of identification authorized by the Chief Electoral Officer each of which establish the elector’s name and at least one of which establishes the elector’s address.
 Documents which satisfy s. 143(2)(a) in British Columbia, for example, are either a driver’s license or a British Columbia Identification (“BCID”) card. A study of identification documents undertaken by Elections Canada while Bill C-31 was before Parliament revealed that very few government-issued documents, apart from a driver’s licence, meet all three requirements of photograph, name and address. A passport, for instance, does not suffice since it does not include the bearer’s address.
 With respect to s. 143(2)(b) of the Act, the CEO is required to publish a list of acceptable identification documents within three days of the issue of an election writ. In practice, the CEO has published the list in a permanent fashion on-line, and makes changes to that list as needed from time to time. As of April 2009, the list of CEO-approved documents was as follows:
· Health Card
· Social Insurance Number Card
· Birth Certificate
· Driver’s Licence
· Canadian Passport
· Certificate of Indian Status
· Certificate of Canadian Citizenship or Citizenship Card
· Credit/Debit Card with elector name
· Canadian Forces Identity Card
· Veterans Affairs Canada Health Card
· Employee Card issued by employer
· Old Age Security Identification Card
· Public Transportation Card
· Student ID Card
· Library Card
· Liquor Identification Card
· Canadian Blood Services/Héma-Québec Card
· Hospital Card
· Fishing Licence
· Wildlife Identification Card
· Hunting Licence
· Firearm Acquisition Card/Firearm Possession Card
· Outdoors Card and Licences
· Provincial/Territorial Identification Card
· Local Community Service Centre Card (CLSC)
Original Documents (containing name and address)
· Credit Card Statement
· Bank Statement
· Utility Bill (residential telephone, cable TV, public utilities commission, hydro, gas or water)
· Attestation of Residence issued by the responsible authority of an Indian band or reserve
· Local Property Tax Assessment
· School, College or University Report Card or Transcript
· Residential Lease, Residential Mortgage Statement or Agreement
· Canada Child Tax Benefit Statement
· Income Tax Assessment Notice
· Insurance Policy
· Government Cheque or Government Cheque Stub with elector name
· Statement of Employment Insurance Benefits Paid (T4E)
· Canada Pension Plan Statement of Contributions/Québec Pension Plan Statement of Participation
· Statement of Old Age Security (T4A) or Statement of Canada Pension Plan Benefits (T4AP)
· Statement of Benefits from provincial workplace safety or insurance board
· Statement of Direct Deposit for provincial works or provincial disability support program
· Vehicle Ownership
· Vehicle Insurance
· Attestation of Residence issued by the responsible authorities (shelters, soup kitchens, student/senior residences, long-term care facilities)
· Letter from public curator
 In the case of electors without a dwelling place, s. 8(6) of the Act prescribes that a temporary address may be used as their place of ordinary residence:
8. (6) A shelter, hostel or similar institution that provides food, lodging or other social services to a person who has no dwelling place is that person’s place of ordinary residence.
 The “Attestation of Residence issued by the responsible authorities” that is referred to in the CEO’s list of approved documents flows from s. 8(6) of the Act, and allows the administrators of certain types of facilities to issue a letter of attestation as proof of the elector’s “residence” at the facility for the purposes of voting.
 Rennie Molnar, the Associate Deputy CEO, Electoral Events, at Elections Canada, deposed that the CEO, in selecting the list of authorized identification, sought to facilitate the exercise of the right to vote while protecting the integrity of the electoral process, recognizing the following imperatives: the statutory requirement for a closed list (and certainty as to who the electors are); the need to identify with sufficient precision the acceptable documents so that elections can be administered without disputes or controversies; the need for the list of documents to be of a size that is understandable and manageable for electors and poll workers; and, in the case of letters of attestation, the need for election workers to obtain sample signatures from responsible authorities at the facilities ahead of polling day.
 In the event that an elector lacks identification that satisfies either s. 143(2)(a) or (b) of the Act, he or she may take a prescribed oath, provided that he or she is vouched for by an elector who is listed on the list of electors in the same polling division as the elector in question, who produces the identification required by ss. 143(2)(a) or (b), and who vouches for the elector by taking an oath. Section 143(3) addresses this issue:
143. (3) An elector may instead prove his or her identity and residence by taking the prescribed oath if he or she is accompanied by an elector whose name appears on the list of electors for the same polling division and who
(a) provides to the deputy returning officer and the poll clerk the piece or pieces of identification referred to in paragraph 2(a) or (b), respectively; and
(b) vouches for him or her on oath in the prescribed form.
 An elector may vouch for only one elector: s. 143(5). An elector who has been vouched for may not vouch for another elector: s. 143(6). Thus, neither “multiple” nor “serial” vouching is permitted.
 An elector who fails to prove his or her identity and residence in accordance with these procedures will not be permitted to vote. This is a consequence of s. 148.1(1) of the Act:
148.1 (1) An elector who fails to prove his or her identity and residence in accordance with subsection 143(2) or (3) or to take an oath otherwise required by this Act shall not receive a ballot or be allowed to vote.
 These identification requirements and vouching process apply to voting at both an advance poll and on polling day by virtue of s. 171(1) of the Act, which provides that an advance poll shall be conducted in the same manner as the vote on polling day.
 Section 161 imposes the same identification requirements on electors seeking to register on polling day, and the same vouching procedure is also available. Section 169 imposes the same identification requirements on electors seeking to register at an advance poll, with the same vouching procedure made available.
 Elections Canada conducted an extensive public information campaign to inform electors about the new voter identification requirements. The campaign included efforts specifically directed at Aboriginal communities, ethnocultural communities, and blind or visually-impaired persons.
 The CEO has reported to Parliament on seven by-elections and one general election conducted since the voter identification requirements came into effect, following some surveys and research. I will refer later to those reports.
 Elections Canada plans to undertake a comprehensive review of the results of the consultations, surveys, research and recommendations that have come forward since the by-elections and the 40th general election in 2008, in order to consider changes to its procedures and practices. It will consider making adjustments to the list of approved identification documents, including the possibility of adding the voter information card as possible proof of residence.
 A number of objections were made regarding the admissibility of evidence during the course of these proceedings.
 Canada objected to the admissibility of significant portions of a February 2009 affidavit of Judy Graves, the Tenant Assistance Co-ordinator for the City of Vancouver, on the basis that those portions constituted both hearsay and opinion evidence not in compliance with Rule 40A of the Rules of Court. I ruled that the plaintiff would be permitted to tender Ms. Graves’s affidavit as an expert report, with some deletions, so long as Canada had an opportunity to challenge Ms. Graves’s qualifications and the basis for her opinion. Canada chose not to cross-examine Ms. Graves on her affidavit, though it did object to the admissibility of certain paragraphs. The plaintiffs addressed a number of issues regarding Ms. Graves’s affidavit in their submissions and in correspondence to the other parties. Accordingly, so that there would be a clear record before the Court, I ordered that the plaintiffs file a replacement affidavit setting out Ms. Graves’s qualifications and identifying her statements of opinion, and the facts and assumptions upon which they are based, in conformity with the requirements for expert opinion evidence. That was done.
 Another area of some contention was the evidence produced by the CEO.
 The CEO in response to interrogatories served by both the plaintiffs and the defendant Canada, provided a large body of documentary evidence relating to the history of the legislative amendments in question and describing the experience of Elections Canada to date with the implementation of those amendments. This evidence included reports that the CEO was statutorily required to present to Parliament, as well as studies commissioned by the CEO. Although the plaintiffs initially referred to this body of evidence in support of their claim that the legislation infringes s. 3 of the Charter, they changed their position after the opening of the hearing and sought to rely on various aspects of the CEO’s evidence only as proof of legislative or social facts bearing upon the purpose and effect of the impugned legislation, in connection with the s. 1 justificatory analysis.
 I will refer later, in my discussion of the section 1 issues, to this evidence and to my conclusions regarding its admissibility and weight. Similarly, I will refer at that later stage to the objections made by the plaintiffs to some of the evidence tendered by Canada, and will state my conclusions as to its admissibility and weight.
 The plaintiff Rose Henry is an anti-poverty advocate who currently resides in Victoria, British Columbia. She is a member of the Snuneymuxw First Nation and a Canadian citizen.
 For over 20 years, Ms. Henry has been an advocate for homeless persons in Victoria. In that capacity, she has been active in facilitating the ability of such persons to exercise their right to vote.
 Ms. Henry herself has been homeless at various times in the past. In early 2007, she lived at an address on Linden Avenue in Victoria. While at that address, she registered to vote and obtained a change of address for her driver’s licence. She was homeless for much of 2007 and 2008. For most of that period, Ms. Henry and her partner slept in their car or “sofa surfed”, staying with different friends. In the fall of 2008, she was able to return to the Linden Avenue address.
 Ms. Henry voted in the 2008 federal general election. At the polling station, she showed the poll clerk her Indian status card and her driver’s licence. The clerk initially questioned her identification because the address on the front of her driver’s licence did not correspond with her address on the list of electors. After a discussion with a supervisor regarding the change of address sticker on the back of Ms. Henry’s licence, the clerk gave Ms. Henry a ballot. Ms. Henry deposes that it was only “pure luck” that she was able to vote because she happened again to be living at the same address as was on her driver’s licence.
 Ms. Henry swears that she has not always had the kind of documentation required to prove her address and identification and that she fears she may not be able to vote in the future.
 Ms. Henry also deposes:
14. In my experience persons who are homeless are often without identification due to the fact that they have no place to keep such documents for safekeeping. Their identification is then lost or stolen and it is very difficult to replace.
15. In addition, obtaining government issued identification that contains a photo and an address is difficult for homeless persons. The most common such document is a driver’s license and most persons who are homeless in our City do not have a driver’s license.
16. Other common forms of government issued documentation are beyond the ability of homeless persons to afford. For example, the cost of a birth certificate is $35.00 and it does not have a photograph or address. A BC Identification Card costs $35.00. Many homeless persons are status Indians under the Indian Act. A “status card” contains a photo, but not an address. The cost of a status card varies from band to band.
17. Homeless persons cannot afford to pay for such identification. Sometimes the provincial social assistance program covers the cost of a BCID card, but they will only do so after the fact, so the recipient has to be in a position to pay the necessary funds, and even then it can take up to a month to get re-imbursement.
26. During the 2006 Federal Election I was the Deputy Returning Officer in the polling place at the Royal Canadian Legion on the Gorge in Victoria, which was the polling station where most of the people in our City who are homeless came to vote.
27. Part of my job was to ask whether people had any acceptable identification in order to register on the voters’ list. We accepted statutory declarations confirming the person’s identity and describing where they customarily lived. Other documents that were acceptable, for example, included prescriptions, a note from a drop in centre, or a letter from an income assistance worker.
28. In the 2008 federal election statutory declarations were not acceptable as documents. I expect that many or most of the persons from whom I accepted statutory declarations were not able to vote.
 At her examination for discovery, Mr. Henry testified that when she voted she had with her a number of other pieces of identification, and two documents showing her address (such that she could have voted even without the driver’s licence). She also testified that there were at least two people who could have vouched for her. She agreed that she had worked on a program that enables homeless people to obtain free identification documents, including BCID cards, health care cards, passports, and social insurance numbers.
 The plaintiff Clyde Wright is a member of the Gitanmaax Band and the Gitksan First Nation, and is a Canadian citizen. He has lived in the Downtown Eastside of Vancouver for the past nine years and has frequently been homeless.
 Mr. Wright is on the Board of Directors of the Vancouver Area Network of Drug Users. One of the organization’s activities is encouraging drug users in the area to participate in elections. Mr. Wright has played an active role in this program, which over the years has helped hundreds of people to vote.
 Mr. Wright gave evidence about the difficulties he has experienced obtaining and keeping identification since arriving in the Downtown Eastside. His Care Card, social insurance card, birth certificate and Indian status card have been stolen at various times. While the welfare office has copies of his social insurance and Indian status cards, original documents are required in order to vote. He does not have a passport, credit or banking card, or a driver’s licence. The cost and procedure of obtaining a replacement Indian status card have deterred him from getting one. The only forms of identification that Mr. Wright does possess are a membership card issued by the Life Skills Centre and a vendor authorization card issued by the Pivot Legal Society so that he is able to assist them in selling calendars for fundraising.
 Mr. Wright voted in the 2008 federal election, and appears to have done so on the basis of an attestation of residence issued by the Pivot Legal Society and a community service centre card.
 He deposes that he fears that he will be unable to vote in future elections because he does not have a place of habitation which could be called a “residential address”.
 Mr. Wright at his examination for discovery testified that there are several organizations that could have provided him with an attestation of residence, and a number of individuals who could have vouched for him.
 The plaintiff Helen Eddlestone is 86 years of age. She is visually impaired but, with assistance from her daughter, is independent and lives on her own in an apartment in Vancouver. She is mobile with the aid of a cane and walker.
 Ms. Eddlestone has been a conscientious voter throughout her adult life, and has taken an active interest in government and public policy issues. She views voting as both a civic duty and a right.
 The last federal general election was held on October 14, 2008. At 7:30 that morning, Ms. Eddlestone walked to the polling station approximately four blocks away from her residence. She had understood that all she needed to vote was the voter information card she had received in the mail from Elections Canada and had not realized that she would be required to present specific forms of identification.
 At the polling station, Ms. Eddlestone was asked by staff whether she had a driver’s licence. She responded that she did not, and then produced a large number of cards, including a photo identification card issued by the Canadian National Institute for the Blind. The only card which included her address was a parking permit from a program for people with disabilities. As Ms. Eddlestone did not have identification that complied with the Act, there was some discussion among the polling station staff about what could be done. One suggestion was that she return later with a bank statement. Ms. Eddlestone was not herself able to see whether someone in the queue could vouch for her as permitted by the Act; she deposes that her suggestion that staff ask whether anyone recognized her was not taken up.
 One of the staff indicated that he would make telephone inquiries to see what could be done. Ms. Eddlestone waited away from the polling booth for a period but the individual did not return. She left the polling station without voting and arrived home at approximately 8:30 a.m. Her evidence is that she “felt disgusted, angry, singled out, embarrassed and humiliated, as though I were not good enough to vote”.
 Ms. Eddlestone had documents at her home which contained her name and address, including:
· her 2004 property assessment;
· a credit card statement;
· a utility bill;
· a bank statement;
· an Old Age Security identification card and T4A slip;
· a Canada Pension Plan benefits statement;
· a Health Insurance BC document; and
· a Canada Revenue Agency document.
 Given her visual impairment, Ms. Eddlestone was not able to see these documents.
 Ms. Eddlestone’s evidence is that she did not subsequently return to the polling station with her bank statement or other document because she could not see to find the documents in the first place, and was too tired to walk back. Also, Ms. Eddlestone had an appointment to take her cat to the veterinarian. She did not contact her daughter for assistance in finding qualifying documents.
 Judy Graves, the Tenant Assistance Co-ordinator for the City of Vancouver since 1993, swore an affidavit tendered by the plaintiffs in evidence. She is the City’s responder to low income tenants and the homeless, and has recently been appointed to the Mayor’s Homeless Emergency Action Team. She works closely with homeless persons in Vancouver, and deposes that throughout her career she has been able to observe and become familiar with their living conditions, the services available to them and the obstacles they encounter, including with respect to voting. She was closely involved in the development of the City’s Homeless Action Plan dated April 26, 2005, and in the Greater Vancouver Homeless Count.
 Ms. Graves deposes that there were some 2,660 counted homeless people in Greater Vancouver as of March 11, 2008. She says that Aboriginal people are greatly over-represented among urban homeless. Her evidence is that although some of the homeless are temporarily sheltered in transition houses, safe houses, emergency shelters or similar facilities, most homeless people are unsheltered or “street homeless”.
 One of Ms. Graves’s functions is to assist homeless people to obtain registration on voters’ lists and to vote in federal, provincial and municipal elections. She deposes:
41. From my experience, in order to permit homeless people to register on the voters’ list, Elections Canada has accepted designations of people’s place of residence other than a civic home address, such as a location they tend to frequent and use as a “base” – for instance, a community centre, library or other neighbourhood facility, or perhaps a location where they tend to sleep or eat which confirms that they normally reside within the constituency where they seek to be registered and to vote.
42. From my experience, an even larger obstacle to getting homeless people registered on the voter’s list, has been the problem of producing any kind of identification documents.
44. It is very common that they have no identity documentation of any sort. Lack of documentation is one of the many chronic problems that make life difficult for many homeless people, and turn commonplace transactions into obstacles.
45. When people are living outside or living in residential hotels it is common that their identification has been stolen from them, often violently. “I.D.” is a valuable asset in neighbourhoods like the Downtown Eastside of Vancouver. They do not replace the stolen identification due to the cost, and because carrying identification leaves them vulnerable to repeated violent theft.
 Ms. Graves describes the pre-Bill C-31 days when statutory declarations were accepted on election day, and refers to the Elections Canada list of approved documentation that may permit voters to receive ballots. She states that based on her experience, she is concerned that many homeless people in Vancouver would not possess any of the indicated documents or, if they did, the documents would not indicate a residential address.
 Canada filed extensive opinion evidence, which I will refer to later in these Reasons since it mainly bears on the section 1 issues.
 Canada filed evidence from Mr. Dan McDougall, who has been Assistant Secretary to the Cabinet (Legislation and House Planning) since June 2005 and was involved throughout the development and implementation of the voter identification amendments to the Act. His affidavit has been the source of much of the preceding description of the background to the amendments. It sets out: some of the history of the development of the federal electoral process, particularly registration and voting processes; a detailed legislative history of Bill C-31; a lengthy discussion of the rationale for the amendments based on the parliamentary record; and what he described as an “explanation of how the voter identification provisions are designed to be broadly accessible to all voters”; and “a discussion of how voter identification enhances the objective of protecting the system from the existence of, perception of and potential for fraud and errors in order to improve the integrity of the electoral process and public confidence ... therein”, which includes a review of instances of known fraud.
 The plaintiffs submit that his affidavit is legal argument, not evidence.
 Although the affidavit contains elements of advocacy for the defendant Canada’s position, it also contains admissible factual evidence regarding the background to the voter identification requirements and their legislative history. I have taken the admissible evidence into account and have disregarded the advocacy.
 Canada also filed affidavits from a number of persons whose evidence bears on the level of difficulty and cost of obtaining identification documents.
 Raymond Fieltsch is the Director of Project Management and Implementation, Regional Services Division: Regional Operations at the Ministry of Housing and Social Development, in the British Columbia government. He deposes that there were (as of September 30, 2008) 107,621 Income Assistance cases in British Columbia, representing some 143,310 individuals. He describes the identification requirements for the receipt of assistance, and says that Ministry staff commonly and readily assist applicants who request assistance in obtaining identification. In addition, he deposes, under provincial legislation, the Ministry may provide a supplement (one-time only) to assist applicants and members of an applicant’s family unit in obtaining proof of identification. He describes a process of assistance in applying for a birth certificate, then for a Social Insurance Number (“SIN”) card, and finally a BCID card issued by the Insurance Corporation of British Columbia (“ICBC”). He says that the Ministry will cover the cost of obtaining the BCID card if the applicant has not previously received a supplement. He states that homeless people may be eligible for Income Assistance and there is no requirement for the provision of an address.
 Antoinette Zanda is the Manager of the Financial Administration Program and the Adult Guardianship Programs at St. James Community Service Society (the “Society”) in Vancouver. Her duties include overseeing the financial management of about 1,000 client accounts of people who receive Income Assistance. Under both programs, the Society assists clients to apply for identification documents, including birth certificates, BCID cards, personal health cards, and citizenship and immigration documents. She deposes that the Society, from funds provided by the Ministry of Housing and Social Development and by Human Resources Development Canada, provides the required fees for the identification applications of individuals involved with the program.
 Linda McLenachan is a Protest, Appeals and Litigation Officer at Indian and Northern Affairs Canada in Gatineau, Québec. She deposes that the Certificate of Indian Status Card is an identification card for status Indians and that no fee is charged for its issue.
 Jack Shewchuk is the Acting Chief Executive Officer of the Vital Statistics Agency of British Columbia. He deposes that the ordinary fee for obtaining a birth certificate is $27.00, but the fee can be waived upon demonstration of financial hardship (an annual income of less than $20,000 per year).
 Kuljinder Sarai is the Manager of Community Outreach Services of Pacific Community Resources in Vancouver and Surrey. He deposes that the Community Outreach Program assists individuals who are referred by the Ministry of Housing and Social Development, including assisting them in obtaining or replacing identification documents such as BCID cards, birth certificates and SIN cards, and covers the necessary fees for the documents.
 Krista Thompson is the Executive Director of Covenant House in Vancouver, which is a crisis intervention centre for street youth. Its services include helping young people in obtaining their birth certificates, SIN or BCID cards, and paying the necessary fees for the documents.
 Kara-Leigh Jameson is the Housing Services Manager of the Senior Services Society in New Westminster. The Senior Services Society from time to time has assisted individuals in obtaining or replacing their identification documents, and in covering the necessary fees on a case-by-case basis.
 Angela Schultz is the Manager of the Client Record Services Department of ICBC in Victoria. She deposes that ICBC issues both photo driver’s licences and non-drivers’ identification cards. She describes the process for obtaining a BCID card (the fee for an original application is $35.00) and states that the $15.00 renewal fee is waived for persons in receipt of income assistance, youth allowance or disability allowance (with proof).
 Chris Davidson is a Program Manger of the Lookout Emergency Aid Society (“Lookout”) in Vancouver, which offers 24-hour, year-round services to adult men and women who are destitute and require assistance. He says that Lookout assists homeless clients to obtain identification documents, such as birth certificates, SIN cards, BCID cards, occasionally passports, and bank cards. He deposes that when the agency assists clients in obtaining income assistance, the Ministry will provide a cheque to clients to pay fully for identification fees in order for them to obtain identification from provincial and federal bodies for income assistance qualification. Mr. Davidson further states that Lookout staff and management put in extra effort to assist homeless people in applying for and acquiring identification, by filling out identification forms and attestations of residence, in order for them to be able to vote.
 The evidence provided by the CEO will be discussed later in these Reasons, in the context of the section 1 analysis.
 The issues to be determined are:
a. Have the plaintiffs established that the impugned provisions of the Act, in their purpose or effect, infringe s. 3 of the Charter?
b. If so, has Canada established that the provisions constitute reasonable limits demonstrably justifiable in a free and democratic society, under s. 1 of the Charter?
c. If the provisions unjustifiably infringe s. 3 of the Charter, what remedy should be granted?
 The plaintiffs submit that the impugned provisions infringe s. 3 of the Charter on three alternative bases: (1) the provisions are inconsistent with s. 3 on their face; (2) they impede, in their effect, the exercise of the franchise; and (3) they are an incomplete realization of the state’s positive duty to enable all citizens to vote. The plaintiffs’ submissions address these latter two bases together, and I will do so as well.
 The plaintiffs do not rely on the evidence provided by the CEO (in the form of reports and surveys regarding experience with the new voter identification requirements) to meet their burden of showing an infringement under s. 3. Instead, they rely solely on the language of the impugned provisions and on the evidence they have tendered themselves.
 The plaintiffs point to statements in authorities such as Sauvé No. 2 that the universal franchise is an essential part of democracy, and that democratic societies have moved from the notion of only a few meritorious people being entitled to vote (defined along lines of gender, class and property ownership) to “the modern precept that all citizens are entitled to vote as members of a self-governing citizenry” (at para. 33, per McLachlin C.J. for the majority).
 The plaintiffs say that the language of s. 148.1(1) of the Act is facially inconsistent with s. 3 of the Charter, and that any asserted countervailing considerations must be justified under s. 1 of the Charter. Further, they argue, the inconsistency of a statute with the Charter does not turn on whether the statute has actually operated to interfere with or limit an individual’s rights; they say that because Canada has conceded that the plaintiffs have standing, it is not necessary for the plaintiffs to show that they have personally been unable to vote in an election.
 The plaintiffs’ position is that the purpose of s. 148.1(1) is to deprive electors of the right to vote in a federal election unless certain conditions precedent are met, and that the provision therefore constitutes a facial infringement of s. 3 of the Charter. They argue that it defines a class of persons who cannot vote: individuals who fail to meet the mandatory requirements for proof of identity and residence.
 Further, the plaintiffs say that the impugned provisions, by preventing citizens from voting unless they are able to confirm their identity and address through one of the means prescribed in the legislation, impede the right to vote in their effect. They point to Ms. Eddlestone’s experience in that regard.
 Mr. Quail and Ms. Allevato, for the plaintiffs, argue that s. 3 rights are “positive” rights, which can only be exercised through state action and which impose obligations on the state to make their exercise possible (referring to Haig v. Canada (Chief Electoral Officer),  2 S.C.R. 995, 105 D.L.R. (4th) 577 [Haig]). They submit that it follows that the government has a positive duty to create electoral processes which enable every eligible Canadian to vote and that “any shortcoming in relation to that proactive standard is, by nature, a limit on the exercise of section 3, and must therefore be rigorously justified under section 1”.
 The plaintiffs concede that Parliament can require voters to prove identity and residence, but take issue with the way that it has imposed that requirement. They argue that the question in the present case is whether the rules have been formulated in such a way as to ensure that no-one falls through the cracks.
 Counsel for the plaintiffs submit that the plaintiffs, among them, share characteristics with the following categories of voters: Aboriginal voters; Aboriginal voters from rural areas; homeless voters; low-income voters; voters with disabilities; elderly voters; and voters without government-issued identification bearing their names and addresses.
 The plaintiffs point to the potential impact of the impugned provisions on persons who do not have access to conventional means of identification (for example, those who are homeless and do not have a residential address). They say that homeless persons, rural voters and Aboriginal voters are particularly vulnerable to disenfranchisement as a result of the provisions.
 In particular, the plaintiffs’ position is that the federal government has failed to provide a failsafe mechanism to ensure that every citizen is able to vote, despite the voter identification requirements. They say that the vouching mechanism does not achieve that goal. They further say that the goal could be met by providing for a sworn declaration in writing that confirms a voter’s identity and place of ordinary residence, comparable to the previously available statutory declaration procedure.
 The plaintiffs argue that the reach of s. 3 is not limited to protecting a “reasonable opportunity” to participate in elections and that “any such relativistic evaluation belongs under s. 1”.
 It is the position of the plaintiffs that, in the context of a challenge to the constitutional validity of the legislation, it is unnecessary for them to prove that anyone was actually prevented from voting by the impugned provisions. They nevertheless submit that the operation of the impugned provisions resulted in the actual deprivation of or interference with Ms. Eddlestone’s right to vote in the last federal election. They say that her ability to vote was “impeded” within the ordinary meaning of that term.
 In Canada’s initial submissions, it urged that the plaintiffs could not assert claims on behalf of other third parties, described as “mythical plaintiffs”. Mr. Wruck referred to Benner v. Canada (Secretary of State),  1 S.C.R. 358 at para. 78, 143 D.L.R. (4th) 577, where Iacobucci J. for the Court stated:
It now appears to be settled law that a party cannot generally rely upon the violation of a third party’s Charter rights: R. v. Edwards,  1 S.C.R. 128, at p. 145; Borowski v. Canada (Attorney General),  1 S.C.R. 342, at p. 367. If the appellant were truly attempting to raise his mother’s s. 15 rights, he would not have the requisite standing.
 Canada subsequently conceded the question of standing. However, Mr. Wruck did not resile from the position that the plaintiffs must provide concrete evidence of an infringement of their own rights (as opposed to the rights of other unspecified persons) and that they have failed to do so.
 Canada points to the evidence that two of the plaintiffs did in fact vote in the last federal election, and that the third plaintiff, Ms. Eddlestone, could have voted in that she possessed the documentary proof of her identity but did not bring it to the polling station. Mr. Wruck also refers to evidence that plaintiffs’ counsel made extensive efforts to locate a disenfranchised voter but were unable to do so, and that the plaintiffs themselves are well-connected with communities in which disenfranchised individuals might be expected to be found.
 Canada submits that the plaintiffs have failed to provide concrete facts or evidence demonstrating any likelihood that they will not be able to vote in any future federal election, and that concrete facts are critical to effective judicial decision-making in constitutional cases. Mr. Wruck points to a number of authorities underlining the importance of a proper factual foundation and the careful analysis of adjudicative facts in constitutional cases: MacKay v. Manitoba,  2 S.C.R. 357 at 361, 61 D.L.R. (4th) 385 [MacKay]; Hy and Zel’s Inc. v. Ontario (Attorney General); Paul Madger Furs Ltd. v. Ontario (Attorney General),  3 S.C.R. 675 at 694, 107 D.L.R. (4th) 634; Danson v. Ontario (Attorney General),  2 S.C.R. 1086 at 1099, 73 D.L.R. (4th) 686 [Danson]; Kitkatla Band v. British Columbia (Minister of Small Business, Tourism and Culture), 2002 SCC 31,  2 S.C.R. 146; Operation Dismantle v. The Queen,  1 S.C.R. 441, 18 D.L.R. (4th) 481 [Operation Dismantle]; R. v. Morgentaler,  3 S.C.R. 463, 107 D.L.R. (4th) 537; R. v. Big M Drug Mart Ltd.,  1 S.C.R. 295, 18 D.L.R. (4th) 321 [Big M Drug Mart]; and The Canadian Bar Association v. HMTQ et al, 2006 BCSC 1342,  1 W.W.R. 331 [Canadian Bar Assn.], aff’d 2008 BCCA 92,  6 W.W.R. 262, leave to appeal ref’d  S.C.C.A. No. 185.
 Mr. Wruck also argues that s. 3 of the Charter must be interpreted in a manner consistent with the underlying purpose of the right to vote. Canada denies that the impugned provisions infringe s. 3. It says that the procedures to establish identity and residence are simply part of the administrative process that enables electors to exercise their right to vote, much in the same way that polling stations are set up at specific locations and during specific times. Canada’s position is that one of the key requirements of any fair electoral process is ensuring that persons seeking to exercise their right to vote are who they say they are and reside in the electoral district in which they seek to vote; the identification requirements are merely procedural mechanisms aimed at ensuring this is the case. Mr. Wruck argues that the identification requirements have not been shown to interfere with the right to vote in any substantive way.
 Canada submits that voting more than once or in an electoral district in which one does not reside, whether fraudulently or in error, erodes the right to vote by violating the fundamental principle of political equality of citizens in a democracy. Mr. Wruck argues that the requirement that those seeking to cast a ballot establish their identities and addresses both protects and enhances the right to vote of all Canadians by strengthening the basic fairness and integrity of the electoral process; in other words, he argues that voter identification contributes to the practical realization of s. 3 rights.
 Canada acknowledges that, because the electoral process is based on residence, homeless persons present a special situation. However, it says that situation has been addressed by Parliament and that, in fact, the provisions of ss. 8(5), 8(6) and 9 of the Act are enfranchising – they create a scheme in which everyone has a place of “residence” and election officials have discretion to facilitate voting in that respect. Canada’s position is that the voter identification scheme is flexible, recognizes a wide range of identification and, in addition, allows for vouching.
 Canada’s position is that there is no infringement of s. 3 of the Charter so long as the means of identification are accessible to potential voters. Section 3, it says, does not protect the right of each citizen to play an unlimited role in the electoral process; rather, voters must act reasonably in pursuit of the right to vote. Mr. Wruck refers in that regard to a South African decision, New National Party of South Africa v. Government of the Republic of South Africa,  2 A.C.C. 5 at para. 40,  ZACC 5 (S. Afr. Const. Ct.) [New National Party]. Mr. Wruck refers to the pervasiveness of identification requirements in today’s society, and argues that the importance of the right to vote is diminished in the eyes of the public if it takes less effort to vote than it does to rent a movie.
 This is not a case of facial infringement, in the view of Canada, because it is not plain and obvious that the impugned legislation is inconsistent with the express language of s. 3 (as was the case with the legislation disentitling prisoners imprisoned for two years or more from voting in Sauvé No. 2, where the s. 3 infringement was conceded). Canada’s position is that the analogy with cases such as Sauvé No. 2 is false because the impugned provisions do not arbitrarily prohibit anyone from voting.
 Mr. Wruck says that the plaintiffs have failed to meet the test in Figueroa v. Canada (Attorney General), 2003 SCC 37,  1 S.C.R. 912 [Figueroa], requiring that they establish on a balance of probabilities that their capacity to vote has been interfered with by the impugned legislation.
 Canada says that if the impugned provisions did in fact interfere in some real way with the capacity of electors to cast a ballot (which it says has not been established), there would be a conflict between core components of the s. 3 right. The Court would then have to make an assessment of the competing core components, namely, between the ability to place a ballot in the ballot box and the ability to cast a vote that is not nullified or diluted by fraudulent or mistaken votes. Its position is that electors can be expected to make reasonable efforts to obtain and bring documentation, and such an expectation does not amount to obstructing the right to vote.
 The CEO neither supports nor opposes the orders sought by the plaintiffs. However, Mr. Whitehall, for the CEO, makes submissions further to the CEO’s provision of documents and responses to interrogatories during the pre-trial stage, and with respect to the question of the nature and timing of a remedy in the event any is granted. He submits that, by virtue of s. 9 of the Act, if anyone is missed under s. 8, the electoral officer has discretion to determine the matter; thus, everyone has an ordinary residence. Mr. Whitehall submits that the CEO has allowed for the broadest possible range of identification documents in the circumstances, and suggests that it would be very exceptional for a person to be unable to prove identity and ordinary residence. He argues that even if there is a gap, the process of identifying appropriate documents is an ongoing one, and the CEO has the capacity and will to close that gap.
 The intervenor submits that the voter identification requirements constitute a step backward in the progress toward a universal right to vote because they impose a barrier to voting which disproportionately affects certain groups, such as homeless, Aboriginal, and rural voters who have greater difficulty meeting the new requirements. Citing Big M Drug Mart at 331 (cited to S.C.R.), Mr. Burnett points to the well-established principle that a law can infringe a Charter right either by purpose or effect. He submits that the impugned provisions have effects which constitute an infringement of s. 3 of the Charter, even if those effects only impact a tiny number of citizens.
 In the submissions of the intervenor, an infringement is established where a law places an impediment before the exercise of a right, and it is no answer to say that the impediment could have been overcome with a greater expenditure of effort or time by the rights-holder.
 The guarantee of the right to vote is a core element of the Canadian Charter and is foundational to democracy. Mr. Justice Cory expressed it in these terms in Haig at 1048-1049 (cited to S.C.R.) (and L’Heureux-Dubé J. for the majority agreed generally with Mr. Justice Cory’s Reasons at 1028 (cited to S.C.R.)):
All forms of democratic government are founded upon the right to vote. Without that right, democracy cannot exist. The marking of a ballot is the mark of distinction of citizens of a democracy. It is a proud badge of freedom. While the Canadian Charter of Rights and Freedoms guarantees certain electoral rights, the right to vote is generally granted and defined by statute. That statutory right is so fundamental that a broad and liberal interpretation must be given to it. Every reasonable effort should be made to enfranchise citizens. Conversely, every care should be taken to guard against disenfranchisement.
The principle was captured by J. P. Boyer in Election Law in Canada: The Law and Procedure of Federal, Provincial and Territorial Elections (1987), vol. 1, at p. 383:
Drawing two short lines to form an “X” is the simplest act imaginable. Yet the right to so mark a ballot is as profound as the act is simple. Such marks, systematically compiled, are transformed by our beliefs and our laws into the most eloquent voice the people have.
The right to cast a vote for those seeking public office is encircled by procedures and laws designed not to make the exercise of this right difficult (although someone frustrated at not being able to vote for a technical reason may feel this is the case), but rather to ensure that it cannot be easily swept away.
 The central importance of the electoral rights to democracy is underscored by the fact that s. 3 is not subject to constitutional override under s. 33 of the Charter: Sauvé No. 2 at para. 36.
 Canada is a signatory to the International Covenant on Civil and Political Rights, 19 December 1966, 999 U.N.T.S. 171, Can. T.S. 1976 No. 47 (entered into force 23 March 1976, accession by Canada 19 May 1976), which provides in Article 25:
Every citizen shall have the right and the opportunity, without any of the distinctions mentioned in article 2 and without unreasonable restrictions:
(a) To take part in the conduct of public affairs, directly or through freely chosen representatives;
(b) To vote and to be elected at genuine periodic elections which shall be by universal and equal suffrage and shall be held by secret ballot, guaranteeing the free expression of the will of the electors;
(c) To have access, on general terms of equality, to public service in his country.
 Although the s. 3 electoral rights are not subject to legislative override, they are, nevertheless, subject to reasonable limits that are prescribed by law if those limits are demonstrably justified in a free and democratic society under s. 1 of the Charter.
 In Figueroa, the Supreme Court of Canada struck down provisions in the Canada Elections Act, R.S.C. 1985, c. E‑2, which had the effect of undermining the ability of minor parties to participate in federal elections. Those provisions required that a political party nominate candidates in at least 50 electoral districts in order to obtain and retain registered party status. Such status entitled a party to issue tax receipts for donations received outside election periods, to transfer unspent funds to the party and to list candidates’ party affiliations on ballots. In the result, the majority concluded that the 50-candidate threshold did infringe s. 3 and that it could not be justified under s. 1 of the Charter.
 The fundamental purpose of s. 3 of the Charter was described by Iacobucci J., for the majority in Figueroa, at para. 30:
In the final analysis, I believe that the Court was correct in Haig, supra, to define s. 3 with reference to the right of each citizen to play a meaningful role in the electoral process. Democracy, of course, is a form of government in which sovereign power resides in the people as a whole. In our system of democracy, this means that each citizen must have a genuine opportunity to take part in the governance of the country through participation in the selection of elected representatives. The fundamental purpose of s. 3, in my view, is to promote and protect the right of each citizen to play a meaningful role in the political life of the country. Absent such a right, ours would not be a true democracy.
 Section 3 grants to each citizen the right to vote and to run for federal and provincial office. The central focus of the provision is therefore, on its very face, the right of each citizen to participate in the electoral process: Figueroa at para. 26.
 In Canada, it is accepted that, at a minimum, “each citizen must have the right to vote, to cast that vote in private, and to have that vote honestly counted and recorded”: Re Prov. Electoral Boundaries at 165 (cited to S.C.R.) per Cory J., dissenting but not on this point. In an earlier s. 3 decision, Dixon v. British Columbia (Attorney General) (1989), 59 D.L.R. (4th) 247,  4 W.W.R. 393 (B.C.S.C.) [Dixon], McLachlin C.J.B.C. (as she then was) set out a broader list of core values or rights that form part of the s. 3 right as follows at 258-259 (cited to D.L.R.):
The Attorney General, relying on Boyer, Political Rights: The Legal Framework of Elections in Canada (1981), at pp. 81 et seq., concedes that the following core values or rights form part of the s. 3 guarantee of the right to vote.
1. The right not to be denied the franchise on the grounds of race, sex, educational qualification or other unjustifiable criteria;
2. The right to be presented with a choice of candidates or parties;
3. The right to a secret ballot;
4. The right to have one’s vote counted;
5. The right to have one’s vote count for the same as other valid votes cast in a district;
6. The right to sufficient information about public policies to permit an informed decision;
7. The right to be represented by a candidate with at least a plurality of votes in a district;
8. The right to vote in periodic elections; and
9. The right to cast one’s vote in an electoral system which has not been “gerrymandered” -- that is, deliberately engineered so as to favour one political party over another.
I would add to this list a tenth precept. It cannot be denied that equality of voting power is fundamental to the Canadian concept of democracy. The claim of our forefathers to representation by populations -- “rep by pop” -- preceded Confederation and was confirmed by it.
 While the wording of s. 3 is quite narrow, courts have looked beyond its literal words in determining the contours of the right to vote and have afforded it a generous interpretation. In Dixon, McLachlin C.J.B.C. commented at 256 (cited to D.L.R.):
The maxim that a full and generous construction must be given to Charter rights and freedoms precludes a narrow, technical view of the right to vote. More is intended than the bare right to place a ballot in a box. Just as “fundamental justice” in s. 7 of the Charter has been interpreted as possessing a substantive as well as a procedural content, so too must s. 3 and its guarantee of the right to vote comprehend substantive rights.
 Just what “more” is intended than the bare right to place a ballot in a box has been the subject of discussion in subsequent Supreme Court authorities. Effective representation, for instance, has been held to be one of the purposes of the right to vote: Reference Re Prov. Electoral Boundaries (Sask.),  2 S.C.R. 158, 81 D.L.R. (4th) 16 [Re Prov. Electoral Boundaries]; Haig; Harvey v. New Brunswick (Attorney General),  2 S.C.R. 876, 137 D.L.R. (4th) 142 [Harvey]; and Thomson Newspapers Co. v. Canada (Attorney General),  1 S.C.R. 877, 159 D.L.R. (4th) 385 [Thomson Newspapers]. Effective representation contemplates more than the right to have and to vote for a representative in Parliament or the legislative assembly; it also implies the right of each citizen to play a meaningful role in the electoral process: Figueroa at para. 25.
 Iacobucci J. for the majority in Figueroa discussed the analytical approach to be adopted under s. 3 of the Charter, at paras. 33-36:
With respect, I do not agree with LeBel J. that the proper analytical approach varies with the nature of the alleged breach. The only difference, in my view, is one of proof. As discussed throughout, the purpose of s. 3 is to protect the right of each citizen to play a meaningful role in the electoral process. Where the impugned legislation is inconsistent with the express language of s. 3, it is unnecessary to consider the broader social or political context in order to determine whether the legislation interferes with the right of each citizen to play a meaningful role in the electoral process. It is plain and obvious that the legislation has this effect. But where the legislation affects the conditions in which citizens exercise those rights it may not be so obvious whether the legislation has this effect. Consequently, it may be necessary to consider a broad range of factors, such as social or physical geography, in order to determine whether the legislation infringes the right of each citizen to play a meaningful role in the electoral process. In neither instance, however, is the right of each citizen to play a meaningful role in the electoral process subject to countervailing collective interests. These interests fall to be considered under s. 1.
As this suggests, I do not believe that the right to play a meaningful role in the electoral process is a “qualified” right, in the same sense as the right to life, liberty and security of the person and the right not to be deprived thereof except in accordance with the principles of fundamental justice (s. 7) or the right to be secure against unreasonable search and seizure (s. 8). It should be noted that the language of s. 7 and s. 8 contains balancing language within the provisions themselves. Accordingly, it is not only appropriate but obligatory to recognize this in interpreting their meaning: see for example R. v. S. (R.J.),  1 S.C.R. 451; and Hunter v. Southam Inc.,  2 S.C.R. 145, which are referenced by my colleague.
According to LeBel J., the fact that we identify the implicit content of s. 3 with reference to qualified phrases such as the right a voter to be “reasonably informed of all the possible choices”, or the right of parties and candidates to have “a reasonable opportunity to present their positions” (Libman v. Québec (Attorney General),  3 S.C.R. 569, at para. 47 (emphasis added)) indicates that the balancing of individual and collective interests that is appropriate in the context of the expressly “qualified” rights also applies under s. 3, except when literal disqualifications are at issue.
In my view, the use of such language does not indicate that the right of each citizen to play a meaningful role in the electoral process is to be balanced against countervailing values, such as the collective interest in the aggregation of political preferences. Rather, the use of such phrases reflects that the purpose of s. 3 is not to protect the right of each citizen to play an unlimited role in the electoral process, but to protect the right of each citizen to play a meaningful role in the electoral process; the mere fact that the legislation departs from absolute voter equality or restricts the capacity of a citizen to participate in the electoral process is an insufficient basis on which to conclude that it interferes with the right of each citizen to play a meaningful role in the electoral process. But if the legislation does, in fact, interfere with the capacity of each citizen to play a meaningful role in the electoral process, it is inconsistent with s. 3. Any corresponding benefits related to democratic values other than the right of each citizen to play a meaningful role must be considered under s. 1.
[Italicized emphasis in original; underline emphasis added.]
 With respect to the role of contextual factors in the s. 3 analysis, Mr. Justice Iacobucci emphasized at para. 33 of Figueroa that it is unnecessary to consider the broader social or political context where the impugned legislation is inconsistent with the express language of s. 3 because in such cases it is plain and obvious that the legislation interferes with the right of each citizen to play a meaningful role.
 The plaintiffs rely on Weremchuk v. Jacobsen (1986), 35 D.L.R. (4th) 278, 10 B.C.L.R. (2d) 205 (C.A.) [Weremchuk], where the Court of Appeal for this province dismissed a challenge to the requirement that ballots be marked with an “X”, not a tick mark, to be valid. After setting out the text of s. 3, Seaton J.A., on behalf of the Court, wrote as follows at 281 (cited to D.L.R.):
The first question must be whether the statutory requirement interferes with or impedes or hinders or restricts the right to vote. Section 101(1)(a) [of the British Columbia Election Act] provides what the voter is to do upon receiving his ballot paper and entering the polling booth. It provides:
“(a) In the case of an ordinary ballot paper in the form prescribed under section 86 [those are the ballot papers with which we are concerned] by making a cross in the white square on the ballot paper opposite to the division, or, if there is more than one member to be elected, opposite to each of the divisions containing the name of a candidate for whom he desires to vote ...”
Does that requirement impede the rights of citizens to vote? I think not. There must be a scheme organized by someone to provide how people are to exercise the right to vote. I think that that someone must be the Legislature; the Charter does not shift responsibility to the courts. So long as the scheme devised by the Legislature does not impede the right to vote, so long as it simply creates a scheme to give effect to the right to vote, the Charter is not offended.
 Based on Weremchuk, the plaintiffs submit that the essential test for infringement is whether the impugned provision impedes the right to vote.
 In my view, the test in Weremchuk is consistent with what the Supreme Court of Canada said in Figueroa: that the question is whether the legislation interferes with the right or capacity of each citizen to play a meaningful (but not unlimited) role in the electoral process, with reference to the purpose of s. 3, which is that “each citizen must have a genuine opportunity to take part in the governance of the country through participation in the selection of elected representatives” (Figueroa at para. 30).
 While the constitutional guarantee of electoral rights in s. 3 of the Charter is a fundamentally important provision, not subject to constitutional override, and must be given the most liberal and generous interpretation consistent with its purpose, some limitations on the right to vote are inherent in the language of s. 3 itself. As was stated in Figueroa, the electoral rights guarantee meaningful participation, not unlimited participation. Section 3 provides that citizens have the right to vote “in an election of members of the House of Commons or of a legislative assembly” (emphasis added). The right to vote in an “election” means that citizens have the right to vote when elections are called. The right to vote in the election of “members of the House of Commons” reflects Canada’s constitutional character as a Parliamentary democracy. It means that Canadian citizens do not have the right to vote at large or to vote anywhere in the country; rather, they have the right to vote in a specific electoral district, choosing among various candidates who wish to be the Member of Parliament for that district. Section 3 thus defines certain parameters on the exercise of the rights it protects.
 Section 3 rights are “participatory in nature” (Figueroa at para. 26) and the Charter creates a positive obligation on the state to put in place appropriate arrangements for the effective exercise of the right to vote: Haig; Weremchuk. In Hoogbruin v. Attorney General of British Columbia (1985), 24 D.L.R. (4th) 718, 70 B.C.L.R. 1 (C.A.) [Hoogbruin], for instance, the British Columbia Election Act, R.S.B.C. 1979, c. 103, entitled absentee voters to vote but failed to provide a mechanism for the exercise of this particular right, namely, an absentee ballot. Mr. Justice Nemetz (as he then was), for the Court, held that by depriving absentee voters of the substance of their right, the Election Act infringed s. 3 of the Charter.
 To give substance to the right to vote, the government must create an apparatus to permit its exercise, which includes rules and procedures for the conduct of fair elections. This point is reflected in the passage from J. Patrick Boyer, Election Law in Canada: The Law and Procedure of Federal, Provincial and Territorial Elections, vol. 1 (Toronto: Butterworths, 1987) at 383, cited with approval by Cory J. in Haig at para. 105:
The right to cast a vote for those seeking public office is encircled by procedures and laws designed not to make the exercise of this right difficult (although someone frustrated at not being able to vote for a technical reason may feel this is the case), but rather to ensure that it cannot be easily swept away.
 The electoral process is necessarily a heavily regulated one, reflected in the fact that the present Act governing federal elections runs to 558 sections. Some aspects of regulation may have restrictive implications for the ease with which certain citizens are able to vote, such as the location and hours of polling stations or the requirement that an elector be on the list of electors for the polling division in which he or she is ordinarily resident.
 In creating the election apparatus, Elections Canada must ensure that the process of voting is as easy and straightforward as possible for all voters. The Charter value of equality (set out in s. 15 of the Charter and recognized in a number of cases: Health Services and Support – Facilities Subsector Bargaining Assn. v. British Columbia, 2007 SCC 27,  2 S.C.R. 391 at paras. 80-81, 84, 86; R. v. Mills,  3 S.C.R. 668, 180 D.L.R. (4th) 1 [Mills]; R. v. Keegsta,  3 S.C.R. 697 at 755-758,  2 W.W.R. 1; Reference Re Public Service Employee Relations Act (Alta.),  1 S.C.R. 313 at 334-335, 38 D.L.R. (4th) 161; Big M Drug Mart at 336 (cited to S.C.R.)) comes into play in ensuring that s. 3 of the Charter is understood and interpreted in a way that maintains the Charter’s underlying values and internal coherence. No group or category of voters should be disproportionately burdened by the requirements imposed for voting, even if the requirements are, on their face, neutral. The government would not be meeting its obligations to conduct fair elections if it failed to take steps to ensure equal access to polling stations and to accommodate Canadian citizens, in all of their diversity, in becoming registered electors and exercising their right to vote.
 By analogy, the courts have clearly stated that relative equality of voting power is fundamental to the principle of representation by population: Dixon; Re Prov. Electoral Boundaries.
 I also note what was said, in a different context, in Harper v. Canada (Attorney General), 2004 SCC 33,  1 S.C.R. 827 [Harper], about the role of egalitarian principles in assessing legislation governing the conduct of elections. In Harper, the plaintiff challenged provisions in the Act that limited election advertising by third parties. The plaintiff relied primarily on s. 2 of the Charter (freedom of expression and freedom of association, but also alleged that his s. 3 rights were infringed. The majority upheld the legislation, accepting that it infringed s. 2(b) (as was conceded by the appellant) but finding that it was saved by s. 1 of the Charter. The Court was unanimous in finding no infringement of s. 3.
 In Harper, Bastarache J., writing for the majority, stated (at paras. 62-63) that Canada employs an “egalitarian” election model:
The Court’s conception of electoral fairness as reflected in the foregoing principles is consistent with the egalitarian model of elections adopted by Parliament as an essential component of our democratic society. This model is premised on the notion that individuals should have an equal opportunity to participate in the electoral process. Under this model, wealth is the main obstacle to equal participation; see C. Feasby, “Libman v. Québec (A.G.) and the Administration of the Process of Democracy under the Charter: The Emerging Egalitarian Model” (1999), 44 McGill L.J. 5. Thus, the egalitarian model promotes an electoral process that requires the wealthy to be prevented from controlling the electoral process to the detriment of others with less economic power. The state can equalize participation in the electoral process in two ways; see O. M. Fiss, The Irony of Free Speech (1996), at p. 4. First, the State can provide a voice to those who might otherwise not be heard. The Act does so by reimbursing candidates and political parties and by providing broadcast time to political parties. Second, the State can restrict the voices which dominate the political discourse so that others may be heard as well. In Canada, electoral regulation has focussed on the latter by regulating electoral spending through comprehensive election finance provisions. These provisions seek to create a level playing field for those who wish to engage in the electoral discourse. This, in turn, enables voters to be better informed; no one voice is overwhelmed by another. In contrast, the libertarian model of elections favours an electoral process subject to as few restrictions as possible.
The current third party election advertising regime is Parliament’s response to this Court’s decision in Libman. The regime is clearly structured on the egalitarian model of elections. The overarching objective of the regime is to promote electoral fairness by creating equality in the political discourse.
 In A History of the Vote in Canada, 2nd ed. (Ottawa: Office of the Chief Electoral Officer of Canada, 2007), the following observation is made at xiv-xv:
Barriers to voting are not only legal or constitutional – they can be procedural or administrative. If citizens have the right to vote but are unable to exercise it because of obstacles inherent in the electoral rules or the way they are implemented, these barriers constitute a restriction of the franchise – one not intended by legislators. The steps taken to overcome such barriers – some of them taken before the advent of the Charter and some of them since – include proxy voting, advance voting, special mail-in ballots, polling-day registration, use of multiple languages in election information, a ballot template for people with visual impairments and level access at polling stations, among many others. In short, the Charter not only guaranteed the right to vote but also highlighted the need to ensure that the right can be exercised.
 Canada urges that the Court must consider whether there is more than a trivial infringement of the plaintiffs’ electoral rights in this case. Mr. Wruck refers to Alberta v. Hutterian Brethren of Wilson Colony, 2009 SCC 37,  2 S.C.R. 567 [Hutterian Brethren], where it was alleged that the imposition of a requirement to have one’s photograph taken for a driver’s licence infringed the freedom of religion of Hutterites, who believe that the Second Commandment prohibits having their photographs willingly taken.
 In Hutterian Brethren, the Supreme Court of Canada majority stated at para. 34 that in assessing an infringement, it was necessary to consider whether the claimants’ religious beliefs or conduct might reasonably or actually be threatened by the universal photo requirement, referring to R. v. Edwards Books and Art Ltd.,  2 S.C.R. 713, 35 D.L.R. (4th) 1, at 759, where the Court stated:
...legislative or administrative action which increases the cost of practicing or otherwise manifesting religious beliefs is not prohibited if the burden is trivial or insubstantial[.]
 Chief Justice McLachlin for the majority in Hutterian Brethren said at para. 34 that “[e]vidence of a state-imposed cost or burden would not suffice”; there would have to be evidence that such a burden was “capable of interfering with religious belief or practice”.
 Mr. Wruck for Canada submits that the test for infringement of s. 3 rights is whether there is a non-trivial interference with the plaintiffs’ capacity to play a meaningful role in the electoral process.
 Mr. Quail for the plaintiffs, on the other hand, says that Hutterian Brethren concerns an infringement of the right to freedom of religion and arose in a very different context than the one at issue here.
 I do not read Hutterian Brethren as changing the existing law regarding the burden on plaintiffs in Charter cases in general. In any event, in my view, imposing preconditions on the right to vote that may have the impact of preventing electors from casting a ballot would clearly constitute more than a trivial interference with their rights.
 Canada’s position is that considerations underlying the decision to pass the impugned legislation, such as the need to prevent voting fraud or voter personation, should enter into the analysis of whether s. 3 of the Charter has been infringed. The plaintiffs say, on the other hand, that those considerations belong under the analysis regarding possible justification pursuant to s. 1 of the Charter.
 In my view, that question is settled by what the Supreme Court of Canada stated in Harvey, finding that the possible justifications for an infringement should be considered in the context of s. 1.
 In Harvey, the issue was the constitutionality of s. 119(c) of New Brunswick’s Elections Act, R.S.N.B. 1973, c. E-3, which prescribed that a member of the legislature convicted of an illegal or corrupt practice pursuant to the Elections Act would lose his or her seat and be disqualified from running as a candidate for five years. The plaintiff in that case asserted that any restriction on the rights set out in s. 3 must be justified under s. 1 of the Charter, while the defendant argued that s. 3 contains its own inherent limitations that need not be justified under s. 1 of the Charter, particularly when a contextual approach is taken to its interpretation. Mr. Justice La Forest for the majority wrote at paras. 25 and 28-31 (cited to S.C.R.):
In contrast to this approach is that used by this Court and others in dealing with particular statutory disqualifications of voters. In Sauvé v. Canada (Attorney General),  2 S.C.R. 438, the Court found that the voting disqualification for inmates found in the Canada Elections Act, R.S.C., 1985, c. E-2, violated s. 3 and could only be justified under s. 1 of the Charter. Similarly, the federal disqualifications of mentally incompetent persons and federally appointed judges, and a provincial disqualification of absentee citizens have all been found to be prima facie unconstitutional; see Canadian Disability Rights Council v. Canada,  3 F.C. 622 (T.D.); Muldoon v. Canada,  3 F.C. 628 (T.D.); and Re Hoogbruin and Attorney-General of British Columbia (1985), 24 D.L.R. (4th) 718 (B.C.C.A.).
While these arguments may initially appear persuasive, I agree with the appellant that the provisions of s. 119(c) are prima facie unconstitutional as violating his rights under s. 3 of the Charter. My reasons are twofold. First, there is the language of s. 3. …. In short, while the English version is somewhat lacking in clarity, the French version is straightforward and indicates that the right to be a candidate and to sit as a member of Parliament or a legislative assembly should be read in a broad manner.
Secondly, and in my view this is decisive, to accept the respondents’ position would be to remove the balancing of interests from s. 1 and incorporate it in s. 3 of the Charter. In their oral submissions counsel for both the respondents and the Attorney General of Canada argued that any given qualification or limitation should first be weighed against the interests represented by s. 3 to determine if there was a violation of that section. Such an approach runs counter to the recent practice of this Court.
In interpreting the right to vote under s. 3 this Court, and Canadian courts in general, have taken the approach that the justification for limitations on the right must be grounded in s. 1 of the Charter. As I have earlier noted, I do not believe the wording in the second part of s. 3 justifies taking a different approach to the right to stand for election and become a member of Parliament or a legislative assembly. This is in accord with this Court’s well established approach of reading Charter rights broadly and putting the burden of justifying limitations upon the state. ...
That is not to say that there can never be limitations or qualifications on the right to stand for election that do not violate s. 3 of the Charter.
 Constitutional issues should be determined in the context of a concrete set of facts, not in the abstract or hypothetically. The Supreme Court stated this firmly in MacKay v. Manitoba at 361-362 (cited to S.C.R).
 Subsequently, in Danson, Mr. Justice Sopinka for the Supreme Court of Canada reiterated the importance of concrete facts, stating at 1099-1100 (cited to S.C.R.):
This Court has been vigilant to ensure that a proper factual foundation exists before measuring legislation against the provisions of the Charter, particularly where the effects of impugned legislation are the subject of the attack. For example, in R. v. Edwards Books and Art Ltd.,  2 S.C.R. 713, at pp. 767-68, this Court declined to hold that the Retail Business Holidays Act, R.S.O. 1980, c. 453, infringed the s. 2(a) Charter rights of Hindus or Moslems in the absence of evidence about the details of their respective religious observance. Similarly, in Rio Hotel Ltd. v. New Brunswick (Liquor Licensing Board),  2 S.C.R. 59, at p. 83, this Court declined to consider a s. 2(b) Charter challenge to certain provisions of the Liquor Control Act, R.S.N.B. 1973, c. L-10, in the absence of evidence on the nature of the conduct that was claimed to constitute “expression” within the meaning of s. 2(b).
It is necessary to draw a distinction at the outset between two categories of facts in constitutional litigation: “adjudicative facts” and “legislative facts”. These terms derive from Davis, Administrative Law Treatise (1958), vol. 2, para. 15.03, p. 353. (See also Morgan, “Proof of Facts in Charter Litigation”, in Sharpe, ed., Charter Litigation (1987).) Adjudicative facts are those that concern the immediate parties: in Davis’ words, “who did what, where, when, how and with what motive or intent ....” Such facts are specific, and must be proved by admissible evidence. Legislative facts are those that establish the purpose and background of legislation, including its social, economic and cultural context. Such facts are of a more general nature, and are subject to less stringent admissibility requirements: see e.g., Re Anti-Inflation Act,  2 S.C.R. 373, per Laskin C.J., at p. 391; Re Residential Tenancies Act, 1979,  1 S.C.R. 714, per Dickson J. (as he then was), at p. 723; and Reference re Upper Churchill Water Rights Reversion Act,  1 S.C.R. 297, per McIntyre J., at p. 318.
In the present case, the appellant contends that he ought to be entitled to proceed with his application under Rule 14.05(3)(h) in the complete absence of adjudicative facts, and, moreover, that it is sufficient that he present in argument (but not prove by affidavit or otherwise) legislative “facts”, in the form of textbooks and academic material about the prevailing understanding of the concept of the independence of the bar, and material concerning the legislative history of the impugned rules. In the view I take of this matter, the appellant is not entitled to proceed with the application as presently constituted.
 The defendant Canada, in its submissions, also relied on Canadian Bar Assn. In that case, the plaintiff sought declarations that the defendants (the British Columbia and Canadian governments and the Legal Services Society) were providing inadequately for civil legal aid, and that they were thereby in breach of the Charter, the Constitution Act, 1982, and international human rights instruments. The Court found that public interest standing is reserved for specific challenges to legislation or government decisions and that the plaintiff did not have standing to bring the claim. Further, Chief Justice Brenner found that the statement of claim failed to show a reasonable cause of action. He stated at para. 114:
The requirement that Charter breaches be pleaded for particular individuals in particular circumstances is not merely a formal requirement arising from the wording of s. 24(1). Without a pleading of individual circumstances, there is no basis on which to make the required causal connection between the government conduct and the alleged breach: see Operation Dismantle at paras. 9-10 and 37-38.
 The Court of Appeal agreed that the Charter challenges failed to raise a reasonable claim (Canadian Bar Assn. v. British Columbia, 2008 BCCA 92, cited previously). Madam Justice Saunders, having referred to British Columbia (Attorney General) v. Christie, 2007 SCC 21,  1 S.C.R. 873, and noting that the law does not support a general right to legal assistance whenever a matter of rights and obligations is before a court or tribunal, stated at paras. 49–50:
In other words, a s. 7 Charter challenge in respect to legal services must be brought in the context of specific facts of an individual’s case because not every legal proceeding affecting a person’s rights requires counsel....
This statement of claim, devoid of particulars of individuals, their cases, and their jeopardy, does not raise a justiciable issue on s. 7. The pleading is simply too general to permit the enquiry sought or the relief contended for.
 A recent case addressing sufficiency of facts was Shapray v. British Columbia (Securities Commission), 2009 BCCA 322, 308 D.L.R. (4th) 704 [Shapray]. In Shapray, the petitioner, a solicitor often involved in proceedings under the Securities Act, R.S.B.C. 1996, c. 418, attacked a provision prohibiting the disclosure of any information or evidence obtained or sought to be obtained, or the name of any witness examined, except to person’s counsel, unless the British Columbia Securities Commission consented to such disclosure. The challenge was based on alleged infringement of freedom of expression. The petitioner contended that the provision had on numerous occasions constrained his ability to communicate lawfully with potential witnesses, discuss matters and cooperate with counsel for other witnesses, and fully engage in normal preparation as counsel. He did not produce evidence that he had sought and been refused consent by the Securities Commission in any particular instance. The trial judge found that s. 2(b) of the Charter was infringed, but that the legislation constituted a reasonable limit under s. 1 of the Charter.
 While allowing the appeal on the merits and finding the provision unconstitutional, the Court of Appeal in Shapray agreed with the decision of the trial judge to address the constitutional issue on the evidentiary record that had been produced.
 A different but closely related issue is whether a challenge is premature. This is sometimes called the question of “ripeness”. Canada emphasizes that the plaintiffs have failed to show that any specific voter has been unable to vote because of the identification requirements. It contends, in effect, that the plaintiffs’ challenge is premature for that reason.
 Canada refers to Operation Dismantle for the proposition that the burden on the plaintiffs is to show that there is a high degree of probability that the alleged harm will in fact occur. Operation Dismantle arose in unique circumstances – an application for an order declaring that proposed cruise missile testing, if it took place, would infringe the s. 7 Charter rights to life and security of the person of Canadian citizens.
 The Shapray case shows that a constitutional challenge to legislation may proceed even where the evidence does not disclose that specific harm has occurred in a specific case, so long as the evidentiary record is sufficient to permit a determination. As another example, see Reform Party of Canada v. Canada (Attorney General), 123 D.L.R. (4th) 366,  4 W.W.R. 609 (Alta. C.A.), where the plaintiff claimed that provisions of the Canada Elections Act limiting the sale of broadcast time during elections infringed the guarantee of freedom of expression, particularly for smaller and newly emerging political parties. The Court of Appeal held that it was not necessary for the plaintiff to prove that it both intended and was able to purchase more time than the scheme would have allowed, in order to proceed.
 As well, proof that the harm feared from an alleged Charter infringement has actually materialized has not been required in cases such as Mills. The issue in Mills was the constitutional validity of ss. 278.1 to 278.91 of the Criminal Code, which dealt with the production of records in sexual offence proceedings. The accused challenged the provisions on the basis that they violated ss. 7 and 11(d) of the Charter. The Court held that it was appropriate to proceed on the basis of an assessment of the effects of the legislation under reasonable hypothetical circumstances. The majority’s reasoning on this issue is set out by McLachlin and Iacobucci JJ. in paras. 35-37 (cited to S.C.R.):
The appellant the Attorney General for Alberta (“Alberta”) submitted that Belzil J.’s finding of constitutional invalidity was premature and lacked an adequate factual foundation. Alberta argued that as no application for records had been made by the respondent under the new provisions, it is unclear what records, if any, would be denied to the respondent. Several lower courts have endorsed this reasoning….
The mere fact that it is not clear whether the respondent will in fact be denied access to records potentially necessary for full answer and defence does not make the claim premature. The respondent need not prove that the impugned legislation would probably violate his right to make full answer and defence. Establishing that the legislation is unconstitutional in its general effects would suffice, as s. 52 of the Constitution Act, 1982, declares a law to be of no force or effect to the extent that it is inconsistent with the Constitution.
However, accepting that the respondent may challenge the general constitutionality of the impugned legislation does not answer the question of whether the respondent must first apply for, and be denied, the production of third party records before bringing a constitutional challenge. The question to answer is whether the appeal record provides sufficient facts to permit the Court to adjudicate properly the issues raised
[Italic emphasis in original; underline emphasis added.]
 These authorities support the conclusion that it is not always necessary for a claimant to show pre-existing harm or damage in order to establish that legislation is inconsistent with the Charter, so long as the factual record is sufficient to permit a determination.
(i) Concreteness and maturity
 I turn now to the present case, and will begin with the questions of whether the factual record is sufficient and the challenge is “ripe” for determination.
 Is there a sufficient factual record?
 The plaintiffs seek a declaration under s. 52 of the Constitution Act that portions of the Act are unconstitutional. The provisions are in effect, and have governed the conduct of seven by-elections and one general election to date. The plaintiffs have provided uncontradicted evidence that they are registered electors, directly affected by the legislation. They have also provided evidence about their experience in voting in past elections, including in the 2008 general election, and about the possible impact of the voter identification provisions on themselves and on others in the future. I find that the plaintiffs have led sufficient evidence of concrete facts to enable the determination of their claim.
 Has the challenge been brought prematurely? The plaintiffs’ position is that the legislation creates a substantial potential obstacle to any voter seeking to exercise the franchise and that they, as citizens, are not required to wait until they have actually lost the opportunity to vote in a specific election before they may call on the government to justify rules which put their votes at risk.
 Canada’s position is that the plaintiffs have not established an infringement and that if they wish to rely on a possible future infringement they must meet the test set out in Operation Dismantle.
 The violation of s. 7 of the Charter alleged by the appellants in Operation Dismantle turned on a hypothesis—namely, that permitting cruise missile testing would cause an actual increase in the risk of nuclear war—that was incapable of proof and contingent on some future harm that was merely speculative. The Court refused to grant the declaratory or injunctive relief sought by the appellants for that reason, among others.
 However, the infringement alleged here does not turn on a hypothesis that is incapable of proof and contingent on some speculative future event. The infringement alleged in this case pertains to legislation that has in fact been promulgated. The plaintiffs say that it is likely that the effect of the legislation will be to prevent some qualified electors from exercising the right to vote. The evidence they have produced (including the evidence that Ms. Eddlestone was refused a ballot because she had not brought with her the requisite identification) bears on the likelihood that such an effect may materialize.
 In Canadian Bar Assn., the challenge was brought by an organization, the Canadian Bar Association, rather than by any individuals, and was “systemic”, being directed at the overall legal aid scheme on the basis that it fails to provide adequate legal aid in civil cases, rather than being directed at the denial of legal aid to specific persons in particular circumstances. In contrast, the challenge here is brought by three individual Canadian citizens who are registered electors, and it is directed at specific legislation that they say affects them.
 The Canadian Bar Assn. case does not stand for the proposition that plaintiffs who challenge specific legislation under the Charter and seek declaratory relief pursuant to s. 52 of the Constitution Act must show that they have already been personally harmed by the legislation (for example, by being denied a ballot in a past election).
 I am not persuaded that the absence of evidence that a specific individual has been unable to vote in a federal election or by-election solely because of the identification requirements is fatal to the plaintiffs’ case. (Ms. Eddelstone was unable to vote because of a combination of the identification requirements, her blindness and mobility impairment, as well as her schedule for the day. The identification requirements were not the sole cause of her inability to vote, but the additional burden they created on her ability to vote significantly contributed to her eventual failure to do so.)
 As Canada correctly concedes, the plaintiffs have standing, as citizens affected by the voter identification requirements in the Act, to challenge the legislation. To then say, as Canada does, that the plaintiffs must fail because they have not established that they or specific other persons have been unable to vote in a specific election because of the identification requirements is, in effect, to say that they must wait until they have actually lost the opportunity to vote before they can complain. That would greatly attenuate the electoral rights. Being denied a ballot is an irretrievable deprivation.
 I decline to accede to Canada’s submissions on this point, which are inconsistent with the purposive and generous approach to Charter interpretation laid down by Dickson J. (as he then was) for the majority of the Supreme Court of Canada in Big M Drug Mart at 344 (cited to S.C.R.):
...In Hunter v. Southam Inc.,  2 S.C.R. 145, this Court expressed the view that the proper approach to the definition of the rights and freedoms guaranteed by the Charter was a purposive one. The meaning of a right or freedom guaranteed by the Charter was to be ascertained by an analysis of the purpose of such a guarantee; it was to be understood, in other words, in the light of the interests it was meant to protect.
In my view this analysis is to be undertaken, and the purpose of the right or freedom in question is to be sought by reference to the character and the larger objects of the Charter itself, to the language chosen to articulate the specific right or freedom, to the historical origins of the concepts enshrined, and where applicable, to the meaning and purpose of the other specific rights and freedoms with which it is associated within the text of the Charter. The interpretation should be, as the judgment in Southam emphasizes, a generous rather than a legalistic one, aimed at fulfilling the purpose of the guarantee and securing for individuals the full benefit of the Charter’s protection. At the same time it is important not to overshoot the actual purpose of the right or freedom in question, but to recall that the Charter was not enacted in a vacuum, and must therefore, as the Court’s decision in Law Society of Upper Canada v. Skapinker,  1 S.C.R. 357, illustrates, be placed in its proper linguistic, philosophic and historical contexts.
[Emphasis in original.]
 Courts should not decide hypothetical issues. However, this issue is not hypothetical; the legislation has been passed, and it affects the rights of the plaintiffs.
(ii) Inconsistency between s. 148.1(1) of the Act and s. 3 of the Charter
 Having concluded that it is appropriate to consider this case, the next question is whether one of the impugned provisions, s. 148.1(1) of the Act, is, on its face, inconsistent with the express language of s. 3 of the Charter. I repeat s. 148.1(1) for convenience:
148.1 (1) An elector who fails to prove his or her identity and residence in accordance with subsection 143(2) or (3) or to take an oath otherwise required by this Act shall not receive a ballot or be allowed to vote.
 A number of the s. 3 cases have dealt with legislation expressly denying specifically defined groups of people the right to vote: prison inmates serving sentences of two years or more in Sauvé No. 2; certain people with mental disabilities in Canadian Disability Rights Council v. Canada,  3 F.C. 622, 21 F.T.R. 268 (T.D.); federally appointed judges in Muldoon v. Canada (Attorney General), 2004 FC 380, 249 F.T.R. 42 (T.D.); and persons convicted of offences involving corrupt electoral practices in Harvey.
 The language of the challenged provisions in those cases was literally inconsistent with s. 3. Because the provisions expressly disenfranchised defined categories of existing persons, their interference with the right of each citizen to play a meaningful role in the electoral process was “plain and obvious”, to use the language in Figueroa at para. 33. As a result, the analysis in each case focused on whether the limitations could be justified under s. 1.
 On the other hand, in this case, s. 148.1(1) of the Act does not expressly disenfranchise a category of citizens. Instead, it precludes persons from voting who are unable to prove their identity and “residence” in the polling district (as broadly defined in s. 8 and s. 9 of the Act). This is an important distinction because it is indisputable that the very purpose of the provisions in other cases, such as Sauvé No. 2, was to disenfranchise a category of persons for a substantive reason, based on their identity, not for their failure to comply with a procedural requirement such as the production of identification.
 Canada points to the language in Figueroa at para. 33 and suggests that any inconsistency with s. 3 must be plain and obvious on the face of the legislation and that it is necessary to go beyond the literal words and consider the context of the legislation, including the fact that the voter identification scheme is designed to enhance the integrity of the vote and ensure that the value of one person’s vote is not eroded through the fraudulent, mistaken or improper voting of others.
 The plaintiffs’ position is that there was no literal provision preventing anyone from running for office in Figueroa and, therefore, it was necessary in Figueroa to consider the implied rights needed to give full effect to the intent of s. 3. In contrast, they say, here a literal provision prevents persons from voting, in that s. 148.1(1) expressly prohibits voting by those who fail to prove their identity and residence.
 I do not understand Figueroa to create a “plain and obvious” test for inconsistency or to approve the importation of justificatory considerations into the analysis of whether the legislation infringes s. 3. Instead, what the Court’s majority stated in Figueroa was that where the impugned legislation is inconsistent with the express language of s. 3, it is unnecessary to go further and look at the broader context because it is plain and obvious that the legislation has the effect of interfering with the right of each citizen to play a meaningful role in the electoral process.
 Hoogbruin bears some similarity to the instant case, in that it too involved an attack on a part of the mechanism for voting. The alleged infringement was in the failure to provide a means for voters to cast ballots if they were absent from the province at election time. There was no explicit disenfranchisement of those who failed to appear in person at the polling station on election day. Nevertheless, the British Columbia Court of Appeal held that s. 3 was infringed, stating at 720-721 (cited to D.L.R.):
We also agree with what Mr. Justice Sirois said in Re Maltby et al. and A.‑G. Sask. et al. (1982), 143 D.L.R. (3d) 649 at p. 670, 2 C.C.C. (3d) 153, 20 Sask. R. 366, where the Legislature had failed to provide a mechanism for certain persons to vote: “To the extent that the statute makes no exception in their case, their rights as guaranteed by s. 3 of the Canadian Charter of Rights and Freedoms to my mind have been violated.” In our opinion, although s. 2(1) of the Election Act of British Columbia states that these appellants have an entitlement to vote, by failing to provide a mechanism to implement that right, the statute has deprived them of the substance of that right and thus infringed their Charter right to vote.
 The Hoogbruin case, to which the Supreme Court of Canada referred in Harvey with seeming approval, shows that the s. 3 rights have both negative and positive characteristics. Section 3 both protects against legislation removing the franchise from a category of citizens and requires governments to create mechanisms permitting the exercise of the franchise.
 Section 148.1(1) denies a ballot to a registered elector who is unable, on election day, to prove his or her identity and residence to the satisfaction of Elections Canada officials, pursuant to the Act. On its face, it limits access to the right to vote to those who are able to satisfy the identification requirements, and its evident purpose is to create that limit. The question is whether this provision is inconsistent with the guarantee of electoral rights under s. 3 of the Charter.
 Prior to the enactment of s. 148.1(1) of the Act, any citizen on the Register of Electors who appeared at the appropriate polling station and identified herself or himself was able to vote. Since the enactment of s. 148.1(1), only those registered electors who provide the required documentary identification or a person to vouch for them are able to vote; other registered electors are disenfranchised. The purpose of the legislation is to create a new condition precedent to voting. In Hoogbruin, the condition precedent of personal attendance at the polling station was held to be a limitation on the right to vote that contravened s. 3 of the Charter. Similarly, in this case, the condition precedent that registered electors must provide identification is inconsistent with the s. 3 Charter guarantee.
 Paraphrasing the words of Iacobucci J., in Figueroa at para. 36, if the legislation interferes with the capacity of each citizen to play a meaningful role in the electoral process, it is inconsistent with s. 3, and any benefits of the legislation (in furtherance of countervailing collective interests) must be considered under s. 1. Section 148.1(1) does interfere with the entitlement to cast a vote for those citizens who are unable to produce the required identification, and thus interferes with their capacity to play a meaningful role in the electoral process.
 I find that s. 148.1(1) of the Act is on its face inconsistent with s. 3 of the Charter.
(iii) Effects of the Legislation
 In the alternative, the plaintiffs submit that the effect of the impugned legislation is to place an increased burden on themselves and on other members of our society – those without a conventional residential address, credit card, driver’s license, or other such documents – who are among our most vulnerable. They say that the effect of the legislation is to put the exercise of the franchise by such persons at risk.
 It is well established in constitutional law that an infringement may occur either through the purpose or effects of legislation: Big M Drug Mart at 331 (cited to S.C.R.); Reference re Same-Sex Marriage, 2004 SCC 79,  3 S.C.R. 698 at para. 40.
 In this regard, the plaintiffs have the burden of establishing that the impugned provisions in effect interfere with the right of each citizen to play a meaningful role in the electoral process, and, in doing so, the plaintiffs may refer to the broader social or political context: Figueroa at para. 33.
 The question is whether the plaintiffs have met that burden. Does the evidence establish that the effect of the impugned provisions is to infringe their rights under s. 3 of the Charter?
 Ms. Henry’s evidence is that she was able to vote in the 2008 general election by mere happenstance, in that by coincidence she had returned to live at the same address as was on her driver’s licence. She deposes that she has not always had the kind of documentation that would permit her to vote, and fears that she will be in that position again in the future. She also says that homeless persons are often without documentary identification and that obtaining common forms of government-issued documentation is beyond the means of homeless persons.
 Mr. Wright’s evidence is that he does not have most of the forms of documentary identification contemplated by the legislation, and does not have a residential address. He was able to vote in the 2008 general election on the basis of (apparently) an attestation of residence issued by the Pivot Legal Society. He, too, deposes that he fears that he will be unable to vote in future elections.
 Ms. Eddlestone was refused a ballot at the polling station and, ultimately, did not vote in the last federal election. That outcome was not the result of an inability on her part to satisfy the new identification requirements; she had in her possession documents that would have entitled her to receive and cast a ballot. Rather, she was not aware of the new identification requirements when she attended at the polling station, and, consequently, did not have appropriate documents with her at the time. Polling station staff suggested to her that she return with a bank statement – a document that she had at her home. Ms. Eddlestone also had an appointment with her daughter later that day and could have sought her daughter’s assistance in gathering that or other documents that would have satisfied the identification requirements. However, given her circumstances, it would have taken much greater effort for her to return to the polling station than for someone more mobile.
 Mr. Wright and Ms. Henry’s evidence shows that they have a reasonably-founded belief that they may have more difficulty voting in the future because of the voter identification provisions. Ms. Eddlestone’s evidence shows that her ability to vote in the 2008 general election was interfered with or impeded by the voter identification requirements, even though those requirements were not the sole cause of her failure to vote.
 I turn to Ms. Graves’s evidence. She deposes that, of the some 2,660 homeless people in Greater Vancouver as of March 11, 2008, Aboriginal people are greatly over-represented (as they are, in general, among urban homeless). Her evidence is that most homeless people are unsheltered or “street homeless” – 58% of the total in Metro Vancouver in 2008.
 Ms. Graves deposes that, commonly, homeless people have no documentation of their identity of any sort, referring to the common problem of theft of such documents and the difficulty of replacing them. She states that based on her experience, she is concerned that many homeless people in Vancouver would not possess any of the indicated documents or, if they did, the documents would not indicate a residential address.
 I find that the plaintiffs’ evidence goes beyond showing that there is some remote or theoretical possibility that the impugned provisions will have the effect of interfering with the exercise of the franchise by members of particularly vulnerable communities.
 I have no doubt that citizens must act reasonably in pursuit of the right to vote, as was stated by Mr. Justice Yacoob of the Constitutional Court of South Africa in New National Party at para. 21. However, the standard of reasonableness must be informed by the actual circumstances of all citizens, not solely by those of the citizens who live in urban communities, are not currently disabled, have a residential address and hold a driver’s licence.
 The existence of economically disadvantaged and homeless persons is not theoretical. I could take judicial notice of the notorious fact that, unfortunately, in our society there are more than a few persons who are economically at the margins, some of whom are homeless; however, the plaintiffs have led evidence to that effect. Specific difficulties that the identification requirements impose upon economically disadvantaged and homeless people are established through the evidence of the plaintiffs and of Ms. Graves.
 In addition, the existence of citizens who live in rural or remote communities, where a journey to a polling station may be long, is not theoretical. For persons in such communities, an error in assembling documentation of their identity before leaving for the polling station may result in denial of a ballot for that election.
 Similarly, the existence of citizens such as Ms. Eddlestone, who are blind or otherwise disabled, who are less likely to have driver’s licences, and for whom return trips to the polling station are inordinately difficult, is not theoretical.
 I find that the evidence shows that the identification requirements imposed by the Act will have the effect of increasing the burden on citizens – particularly economically disadvantaged and homeless citizens, citizens who live in rural or remote communities, and citizens who are disabled – in exercising their right to vote in elections for Members of Parliament. I find that a possible consequence is that some eligible citizens (though likely few in number, given the extensive measures Elections Canada has taken to facilitate voting) may be unable to cast a vote in future elections.
 In my view, the plaintiffs have established that the impugned provisions, in their effect, interfere with the right of each citizen to play a meaningful role in the electoral process.
 I conclude that the plaintiffs have proved an infringement of s. 3 of the Charter on that basis as well as on the basis of a facial inconsistency.
 Having found that s. 148.1(1) of the Act on its face is inconsistent with s. 3, and that the Act’s voter identification scheme in its effects infringes s. 3, the next question is whether the legislation constitutes a reasonable and demonstrably justifiable limitation on the s. 3 right to vote.
(1) Positions of the Parties
 I will briefly outline the positions of the parties at this point, but their submissions are discussed in greater detail later in the s. 1 analysis.
 Canada’s position is that the impugned provisions are justified under s. 1 of the Charter. Applying the analysis mandated by R. v. Oakes,  1 S.C.R. 103, 26 D.L.R. (4th) 200 [Oakes], Canada identifies the objectives of the impugned provisions as protecting the integrity of the vote and maintaining public confidence in the integrity of the electoral system. Both objectives, it says, are pressing and substantial.
 Canada says that the established test for deference, the relative expertise of Parliamentarians in electoral matters, and the weight of the jurisprudence all support a deferential approach to the s. 1 analysis.
 Mr. Wruck submits that the voter identification requirements operate to verify that voters are who they say they are, are voting in the polling division in which they reside, and are the persons on the voters list whom they claim to be – making it more difficult for a person to vote fraudulently or in error. Canada argues that since guarding against fraudulent or mistaken votes protects the integrity of the vote and public confidence in the integrity of the electoral system, the impugned provisions are rationally connected to their objectives.
 Canada submits that the identification regime offers a flexible and accessible system for verifying the identity and residence of voters and does not compromise the universality or accessibility of an effective electoral process. Consequently, Canada says, the provisions are minimally impairing.
 Finally, Canada submits that there is appropriate proportionality between the effects and objectives of the impugned provisions. It says that the plaintiffs have not put before the Court any evidence of deleterious effects flowing from the identification requirements and that the closest the plaintiffs come in this regard are opinion poll surveys that are not reliable for the purposes for which the plaintiffs purport to use them. Canada submits that the impugned provisions occasion few deleterious effects since the vast majority of the population is able to vote with identification already in hand and individuals who do not currently possess suitable documents have only to make some effort to obtain them, causing little more than a minimal inconvenience given the broad and flexible list of documents that will satisfy the requirements. It says that the vouching option further reduces the existence of any deleterious effects. Thus, Canada submits that this final balancing exercise tilts overwhelmingly in favour of the constitutionality of the impugned provisions.
 The plaintiffs stress that the foundational nature of the right to vote mandates an exceptionally rigorous analysis under s. 1 of the Charter. Canada, they say, has failed to justify its infringement.
 The plaintiffs concede that protecting the integrity of the vote and maintaining public confidence in the integrity of the electoral system are both, in principle, pressing and substantial objectives. They further concede that the impugned provisions are rationally connected to those objectives. Where they part company with Canada is with respect to the minimal impairment and proportionate effects stages of the justificatory analysis.
 With respect to minimal impairment, the plaintiffs’ principal submission is that there are alternatives that achieve Parliament’s objectives but are less impairing of the right to vote. In this regard, they say that permitting electors to confirm their identity and residence within the polling division by means of a written declaration or oath administered at the polling station is an alternative measure that actually provides a superior realization of Parliament’s objectives.
 As for proportionate effects, the plaintiffs submit that the deleterious impact of the impugned provisions is disproportionate to the marginal benefits the provisions achieve. They say that the evidence demonstrates that Canada does not have a problem with electoral fraud beyond an absolutely trivial level; on the other hand, the impugned provisions infringe a core democratic right. The plaintiffs further argue that impacts that do not amount to the absolute prevention of voting but that nevertheless discourage participation in the electoral process are detriments that the Court is entitled to take into account when considering proportionate effects. The plaintiffs’ position is that when the full ramifications of the impugned provisions for citizens and the health of Canadian democracy are taken into account, it is apparent that maximum harm has been inflicted upon voters to address a minimal problem.
 The BCCLA submits that the significant contextual factors with respect to the impugned legislation are: the poor and declining participation rates in elections; the absence of any significant public concern over election fraud in Canada; the absence of any history of abuse of the procedures for qualifying to vote; the particular vulnerability of those most affected by the legislation; and the vital nature of the infringed right. Mr. Burnett submits that the contextual factors suggest a heavy burden on Canada to justify the impugned measures, and that deference is inappropriate.
 The position of the BCCLA is that the measures are not minimally impairing and that, for example, Parliament should have retained the statutory declaration option. Further, it submits that there was no actual evidence of problems with voter fraud that needed to be addressed, yet the deleterious effects of the legislation are serious, not only for individuals but for our democratic system.
 Section 1 of the Charter provides:
1. The Canadian Charter of Rights and Freedoms guarantees the rights and freedoms set out in it subject only to such reasonable limits prescribed by law as can be demonstrably justified in a free and democratic society.
 In the present case, the burden is on Canada to establish on a balance of probabilities that the impugned provisions constitute a reasonable limit on the s. 3 rights that can be demonstrably justified in a free and democratic society.
 The relevant analytical framework was set out in Oakes, as recently summarized in Hutterian Brethren. Broadly speaking, the analysis proceeds through these steps:
(a) Is the purpose for which the limit is imposed pressing and substantial?
(b) Are the means by which the goal is furthered proportionate?
(i) Is the limit rationally connected to the purpose?
(ii) Does the limit minimally impair the Charter right?
(iii) Is the law proportionate in its effect?
 Section 1 also requires the Court to satisfy itself that the measures in question are “prescribed by law” before this proportionality analysis is undertaken. As the impugned provisions were enacted by Parliament, I am satisfied that they are unquestionably prescribed by law.
 In Thomson Newspapers at para. 87, the Supreme Court described context as the “indispensable handmaiden” to the proper characterization of the objective of the impugned provision and the proportionality analysis:
The analysis under s. 1 of the Charter must be undertaken with a close attention to context. This is inevitable as the test devised in R. v. Oakes,  1 S.C.R. 103, requires a court to establish the objective of the impugned provision, which can only be accomplished by canvassing the nature of the social problem which it addresses. Similarly, the proportionality of the means used to fulfil the pressing and substantial objective can only be evaluated through a close attention to detail and factual setting. In essence, context is the indispensable handmaiden to the proper characterization of the objective of the impugned provision, to determining whether that objective is justified, and to weighing whether the means used are sufficiently closely related to the valid objective so as to justify an infringement of a Charter right.
 The Supreme Court has found relevant contextual factors to include: (a) the nature of the harm and the inability to measure it; (b) the vulnerability of the group protected; (c) subjective fears and apprehension of harm; and (d) the nature of the infringed activity (Thomson Newspapers at paras. 90-91; Harper; R. v. Bryan, 2007 SCC 12,  1 S.C.R. 527 at para. 10 [Bryan]). Assessment of these factors in a given case will determine whether the court takes a more or less deferential posture with respect to the means chosen by Parliament to implement its objectives. The concept of deference in this context is best understood as being about the nature and sufficiency of evidence required for the government to justify its measures under s. 1: Bryan at para. 28.
 In general, deference will be given to the legislature where the state is balancing competing social interests, where a vulnerable group is being protected, where the effectiveness of the legislation is difficult to measure with scientific accuracy, or where the activity affected by the legislation has a very low social or moral value.
 In Harper, for example, a majority in the Supreme Court of Canada concluded that there was no infringement of s. 3 electoral rights by legislative restrictions on third party election advertising. An infringement of freedom of expression was conceded, and in conducting the analysis under s. 1 with respect to that infringement, the majority held that the contextual factors favoured a deferential approach to Parliament. It found that, given the inherent difficulties in measuring the harm of electoral unfairness brought about by unlimited third party advertising, a reasoned apprehension that the absence of third party advertising limits would lead to electoral unfairness was sufficient.
 Although the parties do not dispute the importance of context in the section 1 analysis, they have quite different views about the appropriate degree of deference that the Court should show Parliament regarding its legislative choices in this case.
 Canada places some emphasis upon the conclusions in Harper described above, and on certain further comments made in Bryan regarding deference.
 The issue in Bryan was the constitutionality of s. 329 of the Canada Elections Act, S.C. 2000, c. 9, which prohibited the broadcasting of election results from one electoral district to another electoral district before all polling stations have closed in that other district. The legislation was challenged under s. 2(b) of the Charter (freedom of expression). The Supreme Court of Canada dismissed an appeal from the British Columbia Court of Appeal decision upholding the legislation, with McLachlin C.J. and Binnie, LeBel and Abella JJ. dissenting. Both Mr. Justice Bastarache and Mr. Justice Fish wrote reasons for the majority.
 The comments upon which the defendant Canada relies in this case were those of Mr. Justice Bastarache at paras. 9-10 of Bryan:
This Court's decision in Harper contains two important principles that are applicable to this case. First, it establishes that courts ought to take a natural attitude of deference toward Parliament when dealing with election laws: "Given the right of Parliament to choose Canada's electoral model and the nuances inherent in implementing this model, the Court must approach the justification analysis with deference" (Harper, at para. 87).
Second, it reaffirms that, in determining the nature and sufficiency of evidence required for the Attorney General to establish that a violation of s. 2(b) is saved by s. 1, the impugned provision must be viewed in its context: see Harper, at paras 75-76, and Thomson Newspapers Co. v. Canada (Attorney General),  1 S.C.R. 877, at para. 88 . This context can be best established by reference to the four factors which this Court set out in Thomson Newspapers and Harper: (i) the nature of the harm and the inability to measure it, (ii) the vulnerability of the group protected, (iii) subjective fears and apprehension of harm, and (iv) the nature of the infringed activity.
 Mr. Wruck for Canada argues that there should be a “natural attitude of deference” toward the impugned election legislation in this case, and that the four factors referred to in Thomson Newspapers indicate that significant deference is appropriate.
 Canada submits that Parliament’s choices in electoral issues should be respected because Parliament is in a far better position than the courts to determine the best course of action.
 The plaintiffs, on the other hand, rely on what was said about deference in Sauvé No. 2. At para. 9, Chief Justice McLachlin wrote:
... The right to vote is fundamental to our democracy and the rule of law and cannot be lightly set aside. Limits on it require not deference, but careful examination. This is not a matter of substituting the Court’s philosophical preference for that of the legislature, but of ensuring that the legislature’s proffered justification is supported by logic and common sense.
 In further explaining why judicial deference to Parliament is not appropriate where the right to vote has been infringed, McLachlin C.J. wrote as follows at paras. 13-15 of Sauvé No. 2:
The core democratic rights of Canadians do not fall within a “range of acceptable alternatives” among which Parliament may pick and choose at its discretion. Deference may be appropriate on a decision involving competing social and political policies. It is not appropriate, however, on a decision to limit fundamental rights. This case is not merely a competition between competing social philosophies. It represents a conflict between the right of citizens to vote – one of the most fundamental rights guaranteed by the Charter – and Parliament’s denial of that right. Public debate on an issue does not transform it into a matter of “social philosophy”, shielding it from full judicial scrutiny. It is for the courts, unaffected by the shifting winds of public opinion and electoral interests, to safeguard the right to vote guaranteed by s. 3 of the Charter.
Charter rights are not a matter of privilege or merit, but a function of membership in the Canadian polity that cannot lightly be cast aside. This is manifestly true of the right to vote, the cornerstone of democracy, exempt from the incursion permitted on other rights through s. 33 override. Thus, courts considering denials of voting rights have applied a stringent justification standard: Sauvé v. Canada (Attorney General) (1992), 7 O.R. (3d) 481 (C.A.) (“Sauvé No. 1”), and Belczowski v. Canada,  2 F.C. 440 (C.A.).
The Charter charges courts with upholding and maintaining an inclusive, participatory democratic framework within which citizens can explore and pursue different conceptions of the good. While a posture of judicial deference to legislative decisions about social policy may be appropriate in some cases, the legislation at issue does not fall into this category. To the contrary, it is precisely when legislative choices threaten to undermine the foundations of the participatory democracy guaranteed by the Charter that courts must be vigilant in fulfilling their constitutional duty to protect the integrity of this system.
 Notwithstanding this stringent approach to the justificatory analysis where s. 3 has been infringed, Canada is not required to justify the infringement with empirical or mathematical precision. As McLachlin C.J. explained in Sauvé No. 2 at para. 18:
While deference to the legislature is not appropriate in this case, legislative justification does not require empirical proof in a scientific sense. While some matters can be proved with empirical or mathematical precision, others, involving philosophical, political and social considerations, cannot. In this case, it is enough that the justification be convincing, in the sense that it is sufficient to satisfy the reasonable person looking at all the evidence and relevant considerations, that the state is justified in infringing the right at stake to the degree it has: see RJR-MacDonald, supra, at para. 154, per McLachlin J.; R. v. Butler,  1 S.C.R. 452, at pp. 502-3, per Sopinka J. What is required is “rational, reasoned defensibility”: RJR-MacDonald, at para. 127. Common sense and inferential reasoning may supplement the evidence: R. v. Sharpe,  1 S.C.R. 45, 2001 SCC 2, at para. 78, per McLachlin C.J. However, one must be wary of stereotypes cloaked as common sense, and of substituting deference for the reasoned demonstration required by s. 1.
 This approach was reaffirmed by Iacobucci J. for the majority in Figueroa at para. 60:
Before beginning this analysis, I note this Court's prior conclusion that limits on s. 3 require not deference, but careful examination: Sauvé, supra, at para. 9. As the Court observed in that case, s. 3 is one of the Charter rights that cannot be overridden by the invocation of s. 33 of the Charter. This highlights the extent to which s. 3 is fundamental to our system of democracy and indicates that great care must be exercised in determining whether or not the government has justified a violation of s. 3.
 Mr. Quail for the plaintiffs submits that Bryan did not overrule Sauvé No. 2 and Figueroa regarding deference. He distinguishes Bryan on the basis that it was not a s. 3 case; rather, it involved an infringement of s. 2(b) of the Charter. He also submits that, like Bryan, Harper was not about the right to vote in the sense of access to the universal franchise; rather, it was about electoral rights that might be described as “derived or incidental” and as “penumbral” in comparison with those regarding the core provisions of s. 3 protecting the literal right to vote and to run for office.
 Mr. Quail submits that the concurring reasons of Mr. Justice Fish in Bryan attracted as much support as did those of Mr. Justice Bastarache, and that in his reasons, Fish J. formulates the deference issue somewhat differently, at paras. 58‑59:
In addressing this issue, I find it important to bear in mind from the outset that we are dealing here with one element of a comprehensive electoral system that temporarily restricts various forms of expression, including exit polls and the ban on election day advertising upheld unanimously by the Court only recently in Harper v. Canada (Attorney General),  1 S.C.R. 827, 2004 SCC 33. In this context, we must be particularly careful not to usurp Parliament's role in determining the rules of the electoral game most appropriate for Canada as a whole. And we must avoid any temptation driven by mere preference, even for objective reasons carefully explained, to tamper with those rules unless they run afoul of Canada's constitutional requirements.
For electoral arrangements of this kind, when Parliament prefers, the courts defer - except where the Constitution otherwise dictates. Judicial deference, however, should not be mistaken for diminished constitutional vigilance, still less for judicial approval or entrenchment. Our role is simply to decide whether Parliament's impugned preference passes constitutional muster. In finding that it does here, I take care to add that Parliament can of course change its mind. Within constitutional bounds, policy preferences of this sort remain the prerogative of Parliament, not of the courts.
 In sum, the plaintiffs’ position is that while a wide variety of policy options may be relevant when considering (as in Harper) whether electoral rules give full meaning and effect to the right to vote and therefore deference may be warranted, that is not the case when considering a limit on the universal franchise, as in Sauvé No. 2. The plaintiffs’ position is that deference is neither required nor appropriate in this case.
 In Hutterian Brethren, Chief Justice McLachlin, for the majority of four, made these general observations about the approach courts should take to s. 1 analysis, at paras. 35‑37:
This Court has recognized that a measure of leeway must be accorded to governments in determining whether limits on rights in public programs that regulate social and commercial interactions are justified under s. 1 of the Charter. Often, a particular problem or area of activity can reasonably be remedied or regulated in a variety of ways. The schemes are typically complex, and reflect a multitude of overlapping and conflicting interests and legislative concerns. They may involve the expenditure of government funds, or complex goals like reducing antisocial behaviour. The primary responsibility for making the difficult choices involved in public governance falls on the elected legislature and those it appoints to carry out its policies. Some of these choices may trench on constitutional rights.
Freedom of religion presents a particular challenge in this respect because of the broad scope of the Charter guarantee. Much of the regulation of a modern state could be claimed by various individuals to have a more than trivial impact on a sincerely held religious belief. Giving effect to each of their religious claims could seriously undermine the universality of many regulatory programs, including the attempt to reduce abuse of driver's licences at issue here, to the overall detriment of the community.
If the choice the legislature has made is challenged as unconstitutional, it falls to the courts to determine whether the choice falls within a range of reasonable alternatives. Section 1 of the Charter does not demand that the limit on the right be perfectly calibrated, judged in hindsight, but only that it be "reasonable" and "demonstrably justified". Where a complex regulatory response to a social problem is challenged, courts will generally take a more deferential posture throughout the s. 1 analysis than they will when the impugned measure is a penal statute directly threatening the liberty of the accused. Courts recognize that the issue of identity theft is a social problem that has grown exponentially in terms of cost to the community since photo licences were introduced in Alberta in 1974, as reflected in the government's attempt to tighten the scheme when it discontinued the religious exemption in 2003. The bar of constitutionality must not be set so high that responsible, creative solutions to difficult problems would be threatened. A degree of deference is therefore appropriate: Edwards Books, at pp. 781-82, per Dickson C.J., and Canada (Attorney General) v. JTI-Macdonald Corp., 2007 SCC 30,  2 S.C.R. 610, at para. 43, per McLachlin C.J.
 In a dissenting judgment in Hutterian Brethren with which Fish J. concurred, LeBel J. commented on the threshold question of deference at para. 187:
It has also been said, at times, that context should be considered at the outset of the analysis in order to determine the scope of the deference of courts to government when applying the Oakes test (Thomson Newspapers Co. v. Canada (Attorney General),  1 S.C.R. 877). One part of this context should not be forgotten: the constitutional context itself. The Charter is designed to uphold and protect constitutional rights. The justification process under s. 1 is not designed to sidestep constitutional rights on every occasion. Rather, it seeks to define and reconcile these rights with other legitimate interests or even between themselves. The burden of justification rests on the state, although I will not attempt, within the limited scope of these reasons, to delve any further into the vexed question of what is sufficient evidence or demonstration of justification. The justification process also reflects the democratic life of a state like Canada, which operates under the rule of law, in the tradition of a parliamentary government, within the framework of a federal form of government. Section 1 and the Oakes test are designed to reach a proper equilibrium between the rule of law, the roles of courts, Parliament or legislatures, and executives, and the democratic life of our country. In the end, when conflict does arise and cannot be resolved, courts must try to strike a proper balance between competing demands, always mindful of their place within the constitutional and political sphere.
 Mr. Justice Bastarache’s comments in Bryan about deference must be read in context. Bryan was a freedom of expression case and did not involve legislation that directly limited the exercise of the franchise. The Court in Bryan did not refer to Sauvé No. 2 or to its non-deferential approach to infringement of s. 3 of the Charter involving the actual exercise of the franchise. Similarly, the comments in Harper must be read in context: a s. 2(b) infringement and legislation that did not limit the exercise of the franchise. Further, the issue in Harper required consideration of the nuances inherent in the choice of a particular regulatory scheme.
 I do not believe that what the Supreme Court said in Sauvé No. 2 has been displaced by a general requirement to defer to Parliament in all challenges to electoral laws. Assessing the constitutionality of schemes for regulating electoral broadcasting, funding of political parties, access to all-party debates, and the like entails reviewing multiple competing considerations in an area where Parliament might bring to bear superior expertise and capacity to make nuanced judgments. On the other hand, the constitutionality of provisions limiting the individual exercise of the franchise – particularly qualification and disqualification from voting – is not such an area.
 The legislation at issue in this case includes a provision explicitly limiting the exercise of the franchise (s. 148.1(1)), but mainly consists in a detailed scheme allowing for a number of different ways to establish identity and residence when voting. Thus, this challenge to the voter identification requirements bears characteristics of both the issue in Sauvé No. 2 and the issue in Harper. It is comparable to Sauvé No. 2 in that one provision in the legislation denies the ballot to persons who fail to provide the required identification. It is comparable to Harper in that the other impugned provisions set up a detailed voter identification scheme involving nuances and the balancing of competing logistical and administrative considerations.
 Bearing those general considerations in mind, I turn to an assessment of the contextual factors specified in Thomson Newspapers: (a) the nature of the harm and the inability to measure it; (b) the vulnerability of the group protected; (c) subjective fears and apprehension of harm; and (d) the nature of the infringed activity.
 In support of deference, Canada’s position regarding the first factor is that the harm in this case – voter fraud and error, actual, potential, and perceived – is by its nature, and due to the secrecy of the vote, very difficult to measure. Turning to the second factor, Mr. Wruck says that the group protected is legitimate voters, who are vulnerable to feeling that their democratic voice is not being respected and who must rely on government action to protect the integrity of the vote. He says with respect to the third factor that, as in Bryan, public confidence in the electoral process is an objective of the impugned provisions and that “the subjective perceptions of Canadian voters that the electoral system is fair is a vital element in the value of the system” (at para. 25). Finally, relating to the fourth factor, he characterizes the nature of the “infringed activity” as the ability of a voter to cast a ballot without any verification of the voter’s identity and residence. He argues that Parliament requires some deference to its arbitration between the democratic values of accessibility and integrity in designing an electoral process that reinforces the effective representation of all electors by making every legitimate vote meaningful.
 The plaintiffs emphasize the fourth factor and the fact that it is a right at the core of the s. 3 rights – the very ability to cast a ballot in an election – that is infringed.
 I agree that the first and third factors indicate that some deference is warranted. I particularly note that it may be difficult to prove that the public’s faith in the integrity of the electoral system will be enhanced by the voter identification provisions, or to prove that the voter identification requirements are a superior deterrent to fraudulent or mistaken voting. The fourth factor, the nature of the infringed activity (i.e., the exercise of the franchise), weighs heavily on the other side of the scale.
 The foundational importance of the right to vote has already been discussed in these Reasons. The Court must exercise great care in determining whether the state has justified denying the right to vote to those who would otherwise be entitled to cast a ballot. However, when assessing whether Parliament has devised mechanisms for voting that maximize the opportunity for all qualified electors to cast a ballot while maintaining the integrity of the system, there must be a level of deference.
 I will now consider whether Canada has established by reasoned demonstration that the impugned provisions constitute a reasonable limit on the electoral rights that is demonstrably justified in a free and democratic society.
 As briefly described in the opening section of these Reasons, both the plaintiffs and Canada made some objections to the admissibility of evidence tendered by the other party; I made some rulings on these objections (Henry v. British Columbia (Attorney General), 2009 BCSC 977). I will not repeat what was said in those rulings, but will summarize the issues and my conclusions.
(i) Evidence Provided by the CEO Regarding Experience
to Date with the Voter Identification Requirements
 I begin with certain evidence originating with the CEO, provided in response to interrogatories from both parties or pursuant to the disclosure requirements on a party to litigation.
 The most pointed dispute concerns certain surveys and reports commissioned by the CEO regarding experience with the voter identification requirements. Those surveys and reports formed the basis of two reports of the CEO to Parliament (Report of the Chief Electoral Officer of Canada following the September 17, 2007, By-elections Held in Outremont, Roberval–Lac-Saint-Jean and Saint-Hyacinthe–Bagot (Ottawa: Chief Electoral Officer of Canada, 2008); and Report of the Chief Electoral Officer of Canada on the 40th General Election of October 14, 2008 (Ottawa: Chief Electoral Officer of Canada, 2009)) [Reports of the CEO].
 Early in this litigation, Canada asked that the surveys and reports themselves be made available, even in draft form, so that Canada’s experts could review them; the CEO co-operated with that request. As I have said, the plaintiffs seek to rely upon portions of this evidence in support of their section 1 submissions.
 Although Canada tendered expert opinion evidence regarding the section 1 issues, the plaintiffs did not (with the exception of Ms. Graves’s affidavit regarding the nature and extent of homelessness in Vancouver). Instead, the plaintiffs wish to rely upon the evidence provided by the CEO as establishing certain legislative and social facts. They say that the evidence is reliable, that it is properly before the Court, and that there is more latitude with respect to evidence tendered for this purpose than there is with respect to evidence tendered to prove adjudicative facts. They refer to Reference re: Upper Churchill Water Rights Reversion Act 1980 (Newfoundland),  1 S.C.R. 297, 8 D.L.R. (4th) 1; Daum v. Schroeder,  8 W.W.R. 432 (Sask. Q.B.); and M. (E.R.) v. Clarke, 2000 BCSC 1695, 105 A.C.W.S. (3d) 435.
 They argue that the Reports of the CEO are admissible for the truth of their contents as public documents, or as legislative facts, and that the underlying surveys and reports are incorporated in the Reports of the CEO by reference.
 Mr. Quail in his submissions also referred to the financial circumstances of the individual plaintiffs and of the organization, the Public Interest Advocacy Centre (“PIAC”), which is representing them in these proceedings. He stated, and I accept, that the plaintiffs and PIAC do not have the means to engage experts, let alone to commission customized surveys of electors and reports on such surveys. He also stated, and this was undisputed, that the federal Court Challenges Program (which, in the past, provided some financial support to individuals or organizations engaged in Charter challenges) no longer exists.
 Canada takes the position that the Reports of the CEO to Parliament (regarding experience with the voter identification requirements to date) are admissible, but that the plaintiffs cannot rely on them for the truth of their contents. This is because, Canada says, the various surveys and reports upon which the CEO’s reports to Parliament are based are inadmissible as hearsay and opinion.
 Mr. Wruck for Canada emphasizes what was said by Mr. Justice Binnie in Public School Boards Assn. of Alberta v. Alberta (Attorney General), 2000 SCC 2,  1 S.C.R. 44 at para. 5, and in R. v. Spence, 2005 SCC 71,  3 S.C.R. 458, regarding the desirability of leading evidence to establish legislative facts if the proposed material does not meet the test for appropriate judicial notice.
 My ruling (subject to further rulings at the end of the day) was that insofar as the survey reports were, in substance, expert opinion reports, the plaintiffs could rely on them for the purpose of proving legislative or social facts (but not adjudicative facts), provided that Canada had an opportunity to challenge the credentials of the researchers and their findings, to cross-examine them and to provide expert opinion evidence in rebuttal. Canada chose not to go to the expense of conducting cross-examinations of the researchers. It did apply to adduce further expert opinion evidence about the reliability and validity of the survey findings, and I allowed that application.
 The CEO subsequently provided extensive further information about the survey reports, including the underlying data and survey questions, as requested by Canada. The CEO also facilitated access between Dr. Corbin, the consultant retained by Canada, and the researchers who had prepared the reports in question for Elections Canada.
 I requested that the CEO, in addition, provide information as to the qualifications of the researchers where that information had not already been provided, and the CEO did so. To ensure that the record is complete, I have ordered that the package of information be marked as Exhibit 1 in this trial.
 The Reports of the CEO are relevant to the issues in this case, which include the effectiveness of the implementation of the new legislation and, in particular, the extent to which the voter identification requirements may have interfered with or impeded qualified electors in the voting process. The Reports of the CEO are prepared pursuant to a statutory duty (ss. 534 and 535 of the Act). The CEO is an independent Officer of Parliament who is responsible to conduct federal elections and by-elections in Canada, to inform himself regarding the conduct of elections and to make recommendations for legislative change if necessary. As I have said, all parties agree that the Reports of the CEO are admissible per se, as relevant public documents, to prove legislative facts.
 The Reports of the CEO are admissible in evidence, for the purpose of proving what the CEO, who is responsible for Canadian federal elections, has concluded regarding experience to date with the voter identification requirements. However, it does not follow from the admissibility of the Reports of the CEO for that purpose that the underlying surveys and reports are, without more, admissible for the truth of their contents.
 I turn to the particular survey reports upon which the plaintiffs wish to rely, which are a subset of the numerous such reports produced by the CEO in response to interrogatories.
 The plaintiffs place particular reliance on the following: Environics Research Group, “Application of the New Voter Identification Procedures under Bill C-31 for the September 17 Federal By-elections in Québec: Evaluation Synthesis”, January 15, 2008; Environics Research Group, “Evaluation of New Voter Identification Requirements, By-elections – March 17, 2008”, June 2008; The Strategic Counsel “Survey of electors following the 40th General Election”, March 2009; and Leger Marketing, “Survey of Election Officers – 40th General Election”, March 2009.
 The plaintiffs also ask me to take into account these reports: Les Études de Marché Créatec, “Evaluation of New Voter Identification Requirements at the September 17, 2007 Federal By-Elections (Electors and Election Officers)”, October 2007; “Homeless Elector Research Project”, I 23, 2008; and Apathy is Boring, “Implementation of the Identification Requirements in the Canadian North”, October 7, 2008.
 None of the reports is attached to an affidavit from its author swearing that its contents are true and that it sets out the author’s opinion. Thus, the reports themselves are hearsay. Insofar as the reports set out what the authors have been told by others, whether through surveys or otherwise, they are double (sometimes triple) hearsay.
 With respect to the reports themselves and the first level of hearsay, although there is no evidence from the authors under oath or affirmation, the parties could have taken advantage of the opportunity to examine or cross-examine the authors as to their credentials and to challenge their evidence.
 Further, the circumstances in which the reports were prepared (at the request of Elections Canada, to assist the CEO in his statutory mandate, and with a view to public availability) provide some guarantee of reliability.
 Given the purpose for which the plaintiffs tender the evidence (to assist the Court in assessing the “big picture” of salutary and deleterious effects of the legislation, but not as proof of any adjudicative facts), and the reasons for viewing the evidence as prima facie facie reliable, I have concluded that it is not necessary to insist that the reports be attached to affidavits from the authors.
 In my view, the more significant concern with these reports is the fact that the reports in many instances set out what others (sometimes anonymously) told the authors or the authors’ associates, rather than the authors’ direct observations. As counsel for Canada pointed out, evidence of survey results can normally be admitted only as a foundation for an expert opinion and not for the truth of its contents.
 I take into account the particular circumstances of this case: the nature of the evidence and the limited purpose for which it is tendered; the opportunities afforded to Canada to review the credentials of the authors of the reports, to cross-examine them, to review the underlying data and methodology, and to call evidence in rebuttal; the provenance of the evidence; the experience and background of the authors of the reports; and the lack of resources of the plaintiffs. In these circumstances, and as was indicated in my earlier ruling, I accept the reports in evidence as if they were expert opinion reports regarding the results of the surveys they describe. In assessing the weight of this evidence, I take into account the contents of the reports, the credentials of their authors, and the evidence of Dr. Corbin and Dr. Archer.
 There were other items, provided by the CEO as part of his disclosure obligations or downloaded from the CEO’s website, that the plaintiffs say I should consider. I indicated that I would assess the admissibility and weight of those items after I had reviewed the evidence as a whole.
 The items provided by the CEO include three internal documents: Electoral Events Sector of Elections Canada, “Analysis of Elector Complaints Instigated by the New Identification Rules” by Stéphane Desrocher, James Lehman & Dominique Nicol, I 16, 2009; Office of the Chief Electoral Officer, “Summary of Political Parties Representatives’ Comments on the Implementation of New Voter Identification Rules Following the 40th General Election”; and Elections Canada, “RO post-mortems – comments on new ID rules (regional consultations)”. In each case, counsel for the Defendant CEO stated that the document was prepared in the ordinary course of business. The analyses in these documents do not reflect an attempt to conduct scientifically valid surveys, but instead to collect and summarize anecdotes and commentary. I will accept these documents in evidence for the limited purpose of showing that the complaints or comments were made and are before Elections Canada in its future planning process.
 Finally, the plaintiffs wish me to consider a document published by the CEO in 2007: Michael J. Prince, “The Electoral Participation of Persons with Special Needs”, which forms part of a “Working Paper Series on Electoral Participation and Outreach Practices”. In the Foreword to this paper, the CEO states that it was commissioned as a concept paper to help Elections Canada refine its outreach strategy and initiatives. The CEO describes this paper as analyzing recent voter participation literature, and reviewing “best practices” in elector outreach in various jurisdictions; he states that the observations and conclusions are those of the author, Michael J. Prince, who is Lansdowne Professor of Social Policy, University of Victoria. Published in 2007, the paper does not discuss the new voter identification requirements or their effects, but does touch on the challenges faced by people with disabilities, homeless people and people with low literacy skills in registering to vote and casting ballots.
 I will take that paper into account only as evidence that the analysis and recommendations it contains were placed before Elections Canada, but not for the truth of its contents.
(ii) Evidence Tendered by Canada
 The plaintiffs objected to some of the evidence tendered by Canada.
 I will begin with the evidence of Dr. Ruth Corbin and of Dr. Keith Archer, which bears on the weight that can be given to the survey evidence provided by the CEO.
 Dr. Corbin holds a doctoral degree in psychology from McGill University. She is the managing partner of CorbinPartners Inc., a marketing science company which conducts survey research and provides market analysis for business and policy decisions. She is also an adjunct professor at Osgoode Hall Law School, where she teaches courses in intellectual property law, trademarks and cognitive science. Throughout her career, Dr. Corbin has been personally involved in the design, execution and analysis of at least 1500 survey research studies, including voting polls, advertising testing, competitive positioning studies, media content analysis and brand audits for a variety of national and international corporations, law firms, governments and public institutions. Her publications include co-authorship of the following books: Ruth M. Corbin, A. Kelly Gill & R. Scott Jolliffe, Trial by Survey: Survey Evidence and the Law (Toronto: Carswell, 2000), and Ruth M. Corbin & A. Kelly Gill, Survey Evidence and the Law Worldwide (Toronto: LexisNexis Canada, 2008).
 Dr. Corbin provided a report regarding the interpretation of the surveys commissioned by Elections Canada.
 In the course of preparing her report, Dr. Corbin obtained the original questionnaires, data tables, verbatim responses to open-ended questions and the raw numerical data, where such supporting details were available.
 The plaintiffs objected to the receipt of the evidence of Dr. Corbin on the basis that it came late in the day when Canada was attempting to “back-stop” its case and that it duplicated evidence already provided from Dr. Archer. I ruled that the evidence of Dr. Corbin would be admitted because it bore on the weight to be given to the survey reports when considered as, in substance, expert opinion evidence. The possibility that the survey reports would be so considered arose as a result of my earlier evidentiary ruling, where I allowed the survey reports in evidence so long as Canada had an appropriate and reasonable opportunity to further challenge the credentials of the researchers and their findings. The plaintiffs did not seek to cross-examine Dr. Corbin or to tender evidence in response to hers.
 The plaintiffs say that much of Dr. Corbin’s evidence should be disregarded as it is based on a mistaken assumption as to the meaning of “impedance of the right to vote” (she defines it as “the inability to reasonably access acceptable identification documents, or to find a person to vouch for the elector”) and on a mistaken belief that the plaintiffs rely on the survey evidence to prove a breach of s. 3 of the Charter. They argue that the survey reports shed some light on the experience in recent elections, and should be considered and given some weight even though undoubtedly the researchers might have asked some different questions or used somewhat different methods if the reports had been designed for the very purpose of this litigation.
 They also submit that Dr. Corbin’s evidence in fact supports the admissibility of the survey evidence in that she confirms that the surveys were prepared with reasonable competence and her critique largely relates to the inferences to be drawn from them.
 I find that Dr. Corbin’s report is of assistance to the court, but take the plaintiffs’ submissions into account in weighing her evidence. I do not find that her definition of “impedance” is tantamount to “absolute prevention”, as counsel for the plaintiffs suggested. While the report strays from time to time into areas appropriate for legal argument rather than expert opinion, it is by and large admissible expert opinion evidence which this witness is qualified to give.
 Dr. Archer is a Professor in the Department of Political Science at the University of Calgary. One of his major areas of research has been election administration. Another has been survey research and data analysis. He prepared two reports at the request of counsel for Canada.
 In the first report, Dr. Archer does the following: describes voter identification requirements in Canada at the federal and provincial level; describes the background to the federal voter identification requirements; and analyzes “whether the amendments are well designed to enhance the integrity of, and public confidence in, the electoral process while maintaining its accessibility” (including a discussion of trends in voter turnout in Canada and the determinants of voter turnout). He concludes that “the weight of evidence leads to the conclusion that no causal relationship has been established between voter identification requirements and voter turnout.” He notes, however, that he has not directly examined the experience with the new federal requirements either in the by-elections of 2007 and 2008 or in the federal general election of 2008.
 Much of Dr. Archer’s first report is essentially legal argument and fails to relate to matters upon which expert opinion evidence is necessary. I give little weight to those portions of his report. However, his opinion as to the absence of a causal relationship between voter identification requirements and voter turnout is relevant and of assistance, as is his evidence regarding the literature on electoral fraud and public perceptions of electoral fraud, and his evidence regarding voter identification provisions in other Canadian jurisdictions.
 In his second report, Dr. Archer addresses the validity of certain conclusions that the plaintiffs, in their original submissions (now withdrawn), said should be drawn from the survey evidence provided by the CEO. I accept that he is qualified to give expert opinion evidence regarding inferences to be drawn from the surveys commissioned by Elections Canada, and that the evidence in his second report is admissible.
 The plaintiffs also objected to certain other evidence submitted by the defendant Canada. Although I considered some portions of the affidavits and reports in question to be argumentative, I declined to rule them wholly inadmissible and indicated that I would address the plaintiffs’ concerns as questions of weight.
 Dr. Pastor is a Professor at American University, and the Founder and Co‑Director of the Center for Democracy and Election Management. His major field of research is elections and democracy and he has consulted and done research in that field in many parts of the world. In his 40-page report, he addresses a number of matters, including: voter identification requirements in the United States (both state and federal); the U.S. Supreme Court decision in Crawford v. Marion County Election Board, 128 S.Ct. 1610 (2008), 553 U.S. 181 (2008) [Crawford], upholding Indiana’s voter identification statute; his opinion as to the rationale for voter identification requirements; the development of voter identification requirements in other Western democracies including the United Kingdom, Northern Ireland, and Australia; and a comparison between the Canadian voter identification requirements and those in other Western democracies. The plaintiffs submit that portions of his report are simply legal argument in the guise of opinion.
 In a number of respects Dr. Pastor’s report is more like legal argument than it is evidence; in other respects it is appropriate expert opinion evidence.
 I have reviewed Dr. Pastor’s report and give it weight only insofar as it is admissible expert opinion evidence within the parameters described by the Supreme Court of Canada in R. v. Mohan,  2 S.C.R. 9, 114 D.L.R. (4th) 419. In general, it is admissible expert opinion evidence where it provides information about voter identification requirements in other countries, including the United States. I accept Dr. Pastor as qualified to provide such evidence.
 The plaintiffs also argued that Dr. Pastor went beyond his expertise in a number of areas; in particular, they argued that he was not qualified to comment on Canadian electoral practices. I am satisfied that he does have expertise in Canadian electoral practices and I decline to give effect to that objection.
 Finally, Canada objects to a law review article upon which the plaintiffs rely: Stephen Ansolabeher & Nathaniel Persily, “Voter Fraud in the Eye of the Beholder: The Role of Public Opinion in the Challenge to Voter Identification Requirements” (2008) 121 Harv. L. Rev. 1737. It was published just before the U.S. Supreme Court decision in Crawford was rendered, and forms part of an enormous body of academic literature on the issues raised in that case. The plaintiffs submit that it shows how “common sense” understandings may be misplaced regarding the tendency of voter identification rules to enhance public confidence in the electoral process.
 The article is inadmissible as evidence of fact, but I will consider it for the assistance it gives in understanding Crawford in its context.
 The first stage in the Oakes analysis requires Canada to establish that the right to vote was limited in pursuit of an objective “of sufficient importance to warrant overriding a constitutionally protected right or freedom”: Big M Drug Mart at 352 (cited to S.C.R.).
 Canada asserts two discrete but related objectives for the impugned provisions, each of which it contends is pressing and substantial: (a) protecting the integrity of the vote and (b) maintaining public confidence in the integrity of the electoral system. The former, it says, relates to protecting the right to vote in the electoral process and ensuring the legitimacy of that process. The latter relates to the public’s perceptions about, and level of trust in, democratic institutions.
 Canada further particularizes these fairly broad objectives as follows:
(a) to address the harm of fraud;
(b) to address the potential for fraud;
(c) to address the public perception of fraud and the potential for fraud;
(d) to address mistakes and inaccuracies in the electoral system and in the conduct of elections thereby ensuring that the electoral system and the conduct of elections is, and is perceived to be, fair, secure and effective; and
(e) to ensure that only eligible voters cast their votes.
 The plaintiffs concede that, in principle, protecting the integrity of the vote and maintaining public confidence in the integrity of the electoral system are both pressing and substantial objectives.
 I agree that Canada’s asserted objectives are capable of justifying limits on the right to vote. As McLachlin C.J. and Major J. wrote in Harper, dissenting but not on this point, at para. 26:
Common sense dictates that promoting electoral fairness is a pressing and substantial objective in our liberal democracy, even in the absence of evidence that past elections have been unfair; see Harvey v. New Brunswick (Attorney General),  2 S.C.R. 876, at para. 38. A theoretical objective asserted as pressing and substantial is sufficient for purposes of the s. 1 justification analysis: see Thomson Newspapers, supra, at para. 38; Harvey, supra, at para. 38; R. v. Wholesale Travel Group Inc.,  3 S.C.R. 154, at p. 191; McKinney v. University of Guelph,  3 S.C.R. 229, at p. 281; Edmonton Journal, supra, at pp. 1343-45.
 With respect to the importance of promoting confidence in the electoral process, I observe the comments of Bastarache J. for the majority in Harper, at para. 103:
Maintaining confidence in the electoral process is essential to preserve the integrity of the electoral system which is the cornerstone of Canadian democracy. In R. v. Oakes,  1 S.C.R. 103, at p. 136, Dickson C.J. concluded that faith in social and political institutions, which enhance the participation of individuals and groups in society, is of central importance in a free and democratic society. If Canadians lack confidence in the electoral system, they will be discouraged from participating in a meaningful way in the electoral process. More importantly, they will lack faith in their elected representatives. Confidence in the electoral process is, therefore, a pressing and substantial objective.
 Having concluded that the asserted objectives of the impugned provisions are pressing and substantial, I turn to consider proportionality.
i. Is the Limit Rationally Connected to the Purpose?
 To establish a rational connection, the government “must show a causal connection between the infringement and the benefit sought on the basis of reason or logic”: RJR-MacDonald Inc. v. Canada (Attorney General),  3 S.C.R. 199 at para. 153, 127 D.L.R. (4th) [RJR-MacDonald]. The requirement that the government prove a rational connection is aimed at preventing the arbitrary imposition of limits on Charter rights. In Hutterian Brethren, McLachlin C.J. for the majority referred to the rational connection requirement, stating at para. 48 that:
...The rational connection requirement is aimed at preventing limits being imposed on rights arbitrarily. The government must show that it is reasonable to suppose that the limit may further the goal, not that it will do so.
 Canada submits that the impugned provisions further the two objectives of protecting the integrity of the vote and maintaining confidence in the integrity of the electoral system. The voter identification requirements verify that voters are the persons they claim to be on the list of electors and are voting in the polling division in which they reside. As a result, it is more difficult for a person to vote fraudulently or in error. Guarding against fraudulent or mistaken votes, Canada says, protects the integrity of the vote and enhances public confidence in the integrity of the electoral system.
 The plaintiffs concede that the majority of the impugned provisions are rationally connected to those objectives. However, the plaintiffs’ position is that, while requiring proof of residence and identity is rationally connected to the objectives, requiring an address is not. They also say that there is no rational connection between the objectives and the following aspects of the legislation as enacted or applied: refusing to accept Indian Status cards as identification; refusing to permit multiple vouching or serial vouching; and requiring that the person vouching for an elector be registered to vote in the same polling district as the elector.
 I will return to these specific issues when discussing the question of minimal impairment.
 Mr. Quail submits, in general, that the existence of a rational connection between the measures and their objectives requires the existence of a real problem to be met by the objectives. He says that there is no real problem of voter fraud and for that reason a rational connection is lacking. He also observes that the voter identification requirements do not actually verify qualification to vote (that is, being an adult citizen of Canada) or prevent mistaken multiple voting by persons who own property in more than one polling district, and that there is no evidence that the rules will bolster public confidence in the electoral system.
 The standard for establishing a rational connection is low. It is reasonable to suppose that this legislation will prevent certain kinds of fraudulent or mistaken voting, and that it will assist in maintaining public confidence in the electoral system. I find that Canada has established a causal connection between the infringement and the benefit sought on the basis of reason and logic.
 The legislation is rationally related to each of Canada’s asserted objectives of protecting the integrity of the vote and maintaining confidence in the integrity of the electoral system.
ii. Does the Limit Minimally Impair the Right?
 The question at this stage of the proportionality analysis is whether the limit on the right to vote is reasonably tailored to the objectives put forth to justify it.
 In RJR-MacDonald, the minimal impairment analysis was explained in these terms by McLachlin J. (as she then was), at para. 160 (and quoted in Hutterian Brethren at para. 54):
As the second step in the proportionality analysis, the government must show that the measures at issue impair the right of free expression as little as reasonably possible in order to achieve the legislative objective. The impairment must be “minimal”, that is, the law must be carefully tailored so that rights are impaired no more than necessary. The tailoring process seldom admits of perfection and the courts must accord some leeway to the legislator. If the law falls within a range of reasonable alternatives, the courts will not find it overbroad merely because they can conceive of an alternative which might better tailor objective to infringement. ... On the other hand, if the government fails to explain why a significantly less intrusive and equally effective measure was not chosen, the law may fail. [Citations omitted.]
 Chief Justice McLachlin discussed minimal impairment further in Hutterian Brethren. After quoting the above passage from RJR-MacDonald, she continued, at paras. 54‑55:
... In this manner, the legislative goal, which has been found to be pressing and substantial, grounds the minimum impairment analysis. As Aharon Barak, former President of the Supreme Court of Israel, puts it, “the rational connection test and the least harmful measure [minimum impairment] test are essentially determined against the background of the proper objective, and are derived from the need to realize it”: “Proportional Effect: The Israeli Experience” (2007), 57 U.T.L.J. 369, at p. 374. President Barak describes this as the “internal limitation” in the minimum impairment test, which “prevents it [standing alone] from granting protection to human rights” (p. 373). The internal limitation arises from the fact that the minimum impairment test requires only that the government choose the least drastic means of achieving its objective. Less drastic means which do not actually achieve the government’s objective are not considered at this stage.
I hasten to add that in considering whether the government’s objective could be achieved by other less drastic means, the court need not be satisfied that the alternative would satisfy the objective to exactly the same extent or degree as the impugned measure. In other words, the court should not accept an unrealistically exacting or precise formulation of the government’s objective which would effectively immunize the law from scrutiny at the minimal impairment stage. The requirement for an “equally effective” alternative measure in the passage from RJR-MacDonald, quoted above, should not be taken to an impractical extreme. It includes alternative measures that give sufficient protection, in all the circumstances, to the government’s goal: Charkaoui v. Canada (Citizenship and Immigration), 2007 SCC 9,  1 S.C.R. 350. While the government is entitled to deference in formulating its objective, that deference is not blind or absolute. The test at the minimum impairment stage is whether there is an alternative, less drastic means of achieving the objective in a real and substantial manner.
[Emphasis in original.]
 As I have said earlier in these Reasons, in calibrating the appropriate level of deference, I bear in mind that although the legislation affects the ability of Canadian citizens to vote in federal elections, crafting the legislative scheme for voter identification when they vote entails carefully balancing competing interests and objectives.
 Canada’s position is that the voter identification regime is minimally impairing. It says that, in giving three possible options for proving identity and residence and in authorizing a broad number of acceptable documents, the identification scheme is specifically designed to be accessible and minimally intrusive for all voters, while still capable of meeting Parliament’s dual objectives. The plaintiffs counter that minimal impairment presents a formidable hurdle to Canada’s case, if for no other reason than there are reasonable alternatives available that would eliminate the infringement of s. 3 while still meeting the legislative objectives.
 In arguing that the impugned provisions are minimally impairing, Canada submits that its identification regime is consistent with international principles of electoral best practices. Canada relies on the evidence of Dr. Robert Pastor, who has written extensively about and is a noted authority on elections and democracy around the world.
 Dr. Pastor states that Canada’s voter identification requirements fall roughly in the middle of the spectrum when compared with similar western democracies. They are not as stringent as those found in certain American states, Northern Ireland or Sweden. On the other hand, they are not as flexible as those in American federal elections, where only first-time voters who registered to vote by mail are required to present identification at the polling station.
 Dr. Pastor further explains that the United States has a highly decentralized election administration system. Under the Constitution, states have the principal responsibility for administering elections, but in reality, almost all states have delegated that authority to the county or municipality level. There is almost no uniformity of procedures or standards. More than half the states have mandated more stringent requirements, many of which require more of voters and offer fewer alternatives than the Canadian identification requirements. Dr. Pastor says that Canada’s identification requirements constitute the middle ground of identification requirements in place in the United States.
 Canada also refers to the evidence of Dr. Keith Archer that British Columbia, Ontario and Québec have voter identification systems similar to Canada’s.
 Dr. Archer’s descriptions of the voter identification systems in the three largest Canadian provinces show both similarities and differences to the system under the Canadian Act. He says that in all three of the jurisdictions, the categories of documents that can be used to establish one’s identification are broad and seemingly designed to be inclusive. He states that he is unaware of any court challenges to those provisions.
 Mr. Pierre-F. Côté, Q.C., the Chief Electoral Officer for the Province of Québec from 1978 to 1997, provided an affidavit regarding the Québec experience with voter identification.
 Mr. Côté describes the Québec system for voter identification and gives some background to the legislation, referring to incidents in 1998 when several voters were solicited by an organized entity to act as impersonators in a particular riding, and a journalist demonstrated that he was able to vote five different times, using five different names, in an election.
 Québec was the first province to pass voter identification requirements in 1999. Its legislation, the Election Act, R.S.Q. c. E‑3.3, as amended in 1999, provides that a voter, to obtain a ballot, must show his or her face and produce one of the following documents: health insurance card, driver’s licence, Canadian passport, Certificate of Indian Status or Canadian Armed Forces identification card. Failing that, the voter must swear an oath as to his or her identity and either produce two documents (one of which contains a photograph) with the elector’s name, or at least two documents which indicate the elector’s name, date of birth and the address appearing on the voters’ list. If the voter is unable to provide documentary identification, he or she may be vouched for by a registered voter with photo identification. Multiple vouching is prohibited, except with respect to one’s spouse and relatives. Voters cannot be registered to vote on polling day. To be registered to vote, a voter must provide two documents that together indicate the voter’s name, date of birth and address.
 Mr. Côté states the opinion that the Québec legislation has been effective in preventing fraud and reinforcing the integrity of the electoral process, without compromising access to the voting booth by legitimate electors.
 The British Columbia legislation, the Election Act, R.S.B.C. 1996, c. 106, as amended on May 29, 2008, requires proof of identity and residential address before a voter can receive a ballot, with three documentary options: (1) government-issued identification showing name, photograph and residential address; (2) Certificate of Indian Status; or (3) two documents with the voter’s name, at least one of which also has the residential address. In addition, a voter may be vouched for by a registered voter. Multiple vouching is prohibited, except that one person can vouch for more than one family member or for voters for whom the voucher has authority under the common law or an enactment to make “personal care decisions”.
 The Ontario legislation, the Election Act, R.S.O. 1990, c. E.6, subsequent to amendments on June 4, 2007, provides two options for entry on the voters’ list or obtaining a certificate to vote: producing a document that includes a name, residential address and signature of the voter; or two other documents, one of which includes the voter’s name and signature and the other of which includes the voter’s name and address. No vouching is permitted. As for obtaining a ballot, voters who do not produce requisite identification are permitted to make a statutory declaration to confirm their eligibility.
 Canada further refers to comments made in Hutterian Brethren at para. 43, where McLachlin C.J. noted:
The chambers judge found that the universal photo requirement was also aimed at harmonization of international and interprovincial standards for photo identification. The evidence supports the Province’s contention that other provinces and nations are moving toward harmonization, and that a feature of this harmonization is likely to be a universal photo requirement for all licence holders. While the fact that other provinces have not yet moved to this requirement arguably undercuts the position that a universal photo requirement is necessary in Alberta now, governments are entitled to act in the present with a view to future developments. Accordingly, harmonization may be considered as a factor relevant to the Province’s goal of ensuring the integrity of the licensing system by reducing identity theft associated with the system.
 The plaintiffs’ position, however, is that the legislation in other jurisdictions is irrelevant, referring to McLachlin C.J.’s statement for the majority in Sauvé No. 2 at para. 41:
... The government’s novel political theory that would permit elected representatives to disenfranchise a segment of the population finds no place in a democracy built upon principles of inclusiveness, equality, and citizen participation. That not all self-proclaimed democracies adhere to this conclusion says little about what the Canadian vision of democracy embodied in the Charter permits.
 The plaintiffs say that, even if what goes on elsewhere is relevant, it is only to set a baseline or minimum standard, referring to Health Services and Support – Facilities Subsector Bargaining Assn. v. British Columbia, 2007 SCC 27,  2 S.C.R. 391 at para. 79:
In summary, international conventions to which Canada is a party recognize the right of the members of unions to engage in collective bargaining, as part of the protection for freedom of association. It is reasonable to infer that s. 2(d) of the Charter should be interpreted as recognizing at least the same level of protection[.]
 Mr. Justice Bastarache for the majority of the Supreme Court of Canada, in Thomson Newspapers at para. 121, stated:
The respondent pointed to the presence of similar blackout periods on public opinion surveys in other democratic countries to support its argument that this measure fell within the permissible range of alternatives. I do not find this evidence to be highly persuasive. Although a number of countries do have such provisions, most democratic countries have minimal or no restrictions on polling information; Lachapelle Study, supra, at pp. 52-62. This may be contrasted with the evidence before the Court in Butler, supra, at p. 497, to the effect that most free and democratic countries had legislation similar to that under scrutiny. Moreover, some of the bans on publication of polling information in other countries extend for very long periods, even the entire election campaign. This suggests that the purpose of those bans may be something other than ensuring that polls are as accurate as possible, which is the only permissible objective under our Charter. Where the approach in other countries is variable, or is in some relevant way different from the legislation under scrutiny in Canada, then those legislative measures must be examined more closely to determine their precise purpose and whether those purposes are of persuasive force here. Not only may the social context be quite different from that in Canada, but also the legal context within which measures restricting the freedom of speech are evaluated may be dissimilar. In the absence of some consensus in the international context, or of evidence explaining why the provisions adopted in some other free and democratic countries are compelling given the situation in Canada, the experience of some other countries as a justification under s. 1 should not be accorded great weight. This is no more than to say that the example of those countries which do not have such provisions is of as much weight in evaluating whether the legislation is justified as those which do. The key question, once the divergence of approach in the international community is established, is whether the values of Canadian society – of which the Charter itself forms a part – are more in accord with one approach rather than the other. The respondent has not taken this extra step in his analysis of those countries which do have publication bans, and therefore, I find this evidence neutral to the outcome of this case.
[Italic emphasis in original; underline emphasis added.]
 It does not appear that there is an international consensus about voter identification, or a consensus within Canada (although it is relevant to note that the three largest provinces, with the great majority of Canada’s population, all have some form of voter identification requirements). There was no evidence presented to suggest that the need for harmonization, flagged in Hutterian Brethren, is a factor. As well, it does not appear that the voter identification requirements in other Canadian jurisdictions have been tested in court.
 Although I have taken note of the approaches to voter identification in other jurisdictions as set out in Dr. Pastor’s, Dr. Archer’s, Mr. McDougall’s and Mr. Côté’s affidavits, for the reasons adverted to in Thomson Newspapers, the evidence about other jurisdictions is ultimately neutral in my analysis.
 As well, the two cases from other jurisdictions that Canada has brought to my attention where voter identification schemes were upheld (Crawford and New National Party) are enlightening in that they illustrate analysis of issues similar to those before me in this case. Nevertheless, they both arose in different social and constitutional contexts than pertain to the issues here.
 Crawford involved a facial challenge to the constitutionality of an Indiana statute requiring citizens to present government-issued photo identification when voting in person on, or casting a ballot prior to, election day. The complainants in Crawford alleged that the new law substantially burdened the right to vote (in violation of the Fourteenth Amendment to the U.S. Constitution), was not an appropriate method for avoiding electoral fraud, would arbitrarily disenfranchise qualified voters who lacked the required identification, and would place an unjustified burden on those who cannot readily obtain such identification, disproportionately affecting elderly, disabled or poor voters, religious objectors, and other minority voters. Although the Court split four ways, a majority of the United States Supreme Court held that requiring non-expired, state-issued photographic identification to vote did not unjustly burden Indiana voters, in large part because the majority found the complainants failed to demonstrate the burdens imposed by the requirements.
 In a strongly worded dissent in Crawford, Mr. Justice Souter found the impugned law unconstitutional since the state interests had failed to justify the practical limitations placed on the right to vote, and the law imposed an unreasonable and irrelevant burden on poor and elderly voters. In coming to this conclusion, Souter J. noted that despite the State’s asserted interest in modernizing elections and combating electoral fraud, Indiana had little support for its voter fraud justification, having not come across one instance of in-person voter impersonation fraud in its history. As well, Souter J. found that the statute would likely burden individual voters given that an estimated 43,000 Indiana voters lacked the requisite identification, and that both the fees and travel costs for obtaining such identification were disproportionately heavy for, and thus disproportionately likely to deter, the poor, the old and the immobile.
 In the New National Party case, the appellant political party, intent on contesting the 1999 national and provincial elections in South Africa, brought a challenge to the constitutionality of certain provisions requiring otherwise qualified voters to possess certain identification documentation in order to register and vote. The impugned provisions prescribed that pursuant to the Electoral Act, No. 73 of 1998, South African citizens otherwise entitled to vote could only participate in the 1999 elections if they possessed and produced one of two identification documents when voting: either a bar-coded identification card or a temporary identification card. Ultimately, a majority of the Constitutional Court of South Africa held that the bar-coded identification scheme prescribed by Parliament did not infringe the right to vote. In coming to this conclusion, the Court noted that it was not unconstitutional for Parliament to expect citizens to act reasonably in pursuit of the right to vote, including taking reasonable steps to meet the documentary requirements.
 The plaintiff argues, and I accept, that New National Party must be understood in the context of South Africa as an emerging democracy attempting to distance itself from its Apartheid past. Voter identification documentation per se was not the central issue in the case. Rather, New National Party involved the implementation of a universal regime of government-issued identity cards, which were to replace the pass cards of the Apartheid era. The pass cards contained codes to register the bearer’s racial classification. The appellant political party sought to have the pass cards, previously used for voter identification, included on the list of acceptable voting documentation. Mr. Justice Yacoob noted that such identity documents were reminiscent of a “shameful past characterised by racial discrimination, oppression and exploitation, untold misery and suffering and the denial to the majority of South African citizens not merely of their right to vote but also of their essential humanity... [and it] would be highly embarrassing if not positively offensive” (at para. 35) to have them included, as the appellants proposed, in the list of acceptable electoral documentation.
 Before moving to the analysis of the specifics in this case, I note that the majority decision in Hutterian Brethren makes clear that the minimal impairment analysis should not be informed by the principles of reasonable accommodation now familiar in human rights cases. At paras. 68-69, McLachlin C.J. wrote:
Minimal impairment and reasonable accommodation are conceptually distinct. Reasonable accommodation is a concept drawn from human rights statutes and jurisprudence. It envisions a dynamic process whereby the parties – most commonly an employer and employee – adjust the terms of their relationship in conformity with the requirements of human rights legislation, up to the point at which accommodation would mean undue hardship for the accommodating party. In Multani, Deschamps and Abella JJ. explained:
The process required by the duty of reasonable accommodation takes into account the specific details of the circumstances of the parties and allows for dialogue between them. This dialogue enables them to reconcile their positions and find common ground tailored to their own needs. [para. 131]
A very different kind of relationship exists between a legislature and the people subject to its laws. By their very nature, laws of general application are not tailored to the unique needs of individual claimants. The legislature has no capacity or legal obligation to engage in such an individualized determination, and in many cases would have no advance notice of a law’s potential to infringe Charter rights. It cannot be expected to tailor a law to every possible future contingency, or every sincerely held religious belief. Laws of general application affect the general public, not just the claimants before the court. The broader societal context in which the law operates must inform the s. 1 justification analysis. A law’s constitutionality under s. 1 of the Charter is determined, not by whether it is responsive to the unique needs of every individual claimant, but rather by whether its infringement of Charter rights is directed at an important objective and is proportionate in its overall impact. While the law’s impact on the individual claimants is undoubtedly a significant factor for the court to consider in determining whether the infringement is justified, the court’s ultimate perspective is societal. The question the court must answer is whether the Charter infringement is justifiable in a free and democratic society, not whether a more advantageous arrangement for a particular claimant could be envisioned.
 I turn to the application of the principles regarding minimal impairment to this case.
 As described earlier, the first option for satisfying the identification requirements under the Act is a single piece of government-issued identification which includes the bearer’s name, address and photograph. A driver’s licence meets these requirements. The evidence of the former CEO presented to the Committee in December 2006 was that 85% of electors hold driver’s licenses.
 Canada refers to the body of evidence it provided showing that “option 1” identification documents such as the BCID card (government-issued, bearing the elector’s name, address and photograph) are available, sometimes without charge, to persons of limited means, and that some assistance is available from governmental and non-governmental organizations to persons who need to obtain identification.
 Two pieces of identification authorized by the CEO, both with the voter’s name and at least one with the voter’s address, is the second option for satisfying the identification requirements. The list of authorized “option 2” documents, current as of I 2009, was set out earlier in these Reasons. The CEO assesses and updates this list as required. It is the evidence of a number of witnesses, including Mr. Molnar, that the list of approved documents is crafted to ensure broad accessibility. This is indeed apparent from a review of the extensive list, which includes such common documents as a health card, credit card and library card. Approved documents which also bear the voter’s address include such generally accessible documents as residential lease, bank statement and utility bill.
 With respect to particular groups of electors who might be less likely to have such documents in their possession, the list includes some documents that may facilitate their ability to satisfy this second option. Thus, seniors are able to rely on an Old Age Security Identification Card, Statement of Old Age Security (T4A) or Statement of Canada Pension Plan Benefits (T4AP). For students, the list includes student identification cards, report cards and transcripts. Aboriginal Canadians are able to use a Certificate of Indian Status, while homeless people can use a local community service centre card.
 Attestations of residence are also available. An attestation of residence can act as proof of residence for students, seniors in residences, people in long term care facilities, Aboriginal persons on reserves and those without a residence. The form is completed by the administrator of the relevant institution to confirm the identities of electors “residing” there. Canada submits that this document is included precisely for the purpose of facilitating the capacity of voters without a fixed address to establish a temporary address in order to vote.
 The third option that satisfies the identification requirements is vouching. Canada argues that this procedure provides yet greater accessibility and a “failsafe” measure because it enables those without acceptable identification to cast a ballot if vouched for by another elector who is on the list at the same polling division and who does have acceptable identification. Prior to the 2007 amendments, vouching was permitted for registration purposes pursuant to s. 161(1)(b) of the Act. It is now permitted for voting, as well.
 The plaintiffs emphasize specific aspects of the legislation as failing to minimally impair the electoral rights: requiring electors to have an address; refusing to permit multiple or serial vouching; requiring that a person vouching for an elector be registered in the same polling district as that elector; and refusing to accept Indian status cards.
 The plaintiffs’ position is that to the extent that the legislation requires an elector to have an address, it is not minimally impairing.
 However, the extent to which the legislation requires an elector to have an address is a matter of some controversy among the parties.
 Both Canada and the CEO submit that the legislation does not in fact require a registered elector to establish an address in order to receive a ballot; instead, it requires electors to prove their residence. Mr. Whitehall for the CEO submits that, under the legislative scheme, people are entitled to vote where they are ordinarily resident and everyone has a single ordinary residence by definition (referring to s. 8 of the Act) that can be proved either by showing proof of address (which need not be a civic address), or through a letter of attestation pursuant to s. 8(6). The CEO’s position is that everyone has an ordinary residence and virtually everyone will be able to prove their place of ordinary residence. Mr. Whitehall points as well to s. 9 of the Act, which he describes as the “gap” rule, allowing the election officer to determine residence if the rules in s. 8 are not sufficient to do so.
 Is it necessary, however, to provide an address in order to become a registered elector? The Register of Electors (s. 44(2) of the Act) is to contain, for each elector, his or her surname, given names, sex, date of birth, civic address and mailing address. Further, s. 49(1)(b) of the Act, which sets out the procedure for seeking inclusion on the Register of Electors, requires the applicant to provide that same information, including mailing and civic addresses. It requires proof of identity, but does not explicitly require proof of residence (although, in practice, it appears that proof of residence is required because s. 6 of the Act requires that the elector be added to the list for the polling division where he or she is ordinarily resident, and s. 2(3) of the Act allows the CEO to require documentary proof of the elector’s identity and residence.)
 When a person who does not have a civic address or who does not have identification documents with a civic address wishes to register as an elector on voting day (for example, an elector who has only a Post Office box address), that can only be done by vouching. The person vouching must have acceptable identification, but need not have a civic address if the address in his or her identification documents is consistent with that found on the list of electors (s. 161(1)(b)(i)). The same applies for registering to vote at an advance poll.
 The plaintiffs submit that the legislation could be read to require the possession of an address in order to be on the Register of Electors. However, Canada and the CEO submit that the legislation permits persons without civic or mailing addresses to be placed on the list of electors, whether before voting day or on voting day, for the polling district in which they ordinarily reside. The CEO submits that s. 6 of the Act shows what the requirement of “address” is for – it is to show which polling division contains the elector’s place of ordinary residence, and thus the place where he or she can vote and that the legislation should be read accordingly. The evidence of Ms. Graves confirms that the practice of Elections Canada is to permit homeless people to register as voters on the basis of a location that they frequent or use as a “base”.
 I agree with the interpretation urged by the CEO and Canada. Voters may be placed on the Register of Electors, and vote, based on ordinary residence; having a civic or mailing address is not a pre-condition.
 Electoral legislation, where it is ambiguous, should be interpreted in a way that is enfranchising (consistent with what was said in Haig at 1048 (cited to S.C.R.)). In general, where there is genuine ambiguity in that legislation is amenable to two possible interpretations, a court should choose the interpretation that upholds the legislation as constitutional: Bell Express Vu Ltd. Partnership v. Rex, 2002 SCC 42,  2 S.C.R. 559 at paras. 62-63. Because an interpretation requiring all electors to have a civic or mailing address would be disenfranchising, and possibly unconstitutional, I accept the interpretation argued by Canada and the CEO.
 The plaintiffs dispute Canada’s characterization of vouching as a “failsafe” alternative that makes sure no-one falls through the cracks, arguing that the vouching provisions do not go sufficiently far. Vouching is only available to an elector who has an address and personally knows someone who: (1) is an eligible voter; (2) has the requisite identification documents; (3) is registered to vote in the same polling division; (4) has not vouched for anyone else in the election; (5) attends in person at the same time as the elector; and (6) is willing to provide this voluntary assistance. The effect of these requirements, they say, is to create only a small pool of potential vouchers. To illustrate this point, counsel for the plaintiff argues that, for example, a parent with two voting-age children who came to the polling station would only be able to vouch for one of them. The plaintiffs say that a scheme permitting multiple and serial vouching would be just as effective in meeting the objectives because there is no evidence that multiple or serial vouching has led to wrongful voting in the past.
 Canada responds that more permissive vouching provisions, such as multiple or serial vouching or vouching by persons outside the polling division, would undermine the achievement of the legislative objectives and would not represent less impairing means of ensuring the same degree of security or confidence in the voting process. Serial vouching (the situation in which an individual without identification who has been vouched for then vouches for someone else without identification), it says, undermines the goal of electoral integrity since it would permit persons with no direct evidence of their identity, and therefore no direct paper trail, to enable another in the same circumstances to vote. In explaining before the House of Commons in January 2007 the rationale for the prohibition on multiple vouching, the Government House Leader and Minister for Democratic Reform stated that “it limits the ability of one bad actor to commit large scale fraud and that is a positive thing for the integrity of the system” (i.e., one person with bad intentions would not be able to vouch for 250 people). Canada says that as a practical matter, the list of electors at a polling station relates only to that particular polling division, so the individual who vouches can only be verified as a qualified elector in that division.
 Further, the plaintiffs say that the availability of reasonable alternatives which eliminate the Charter infringement deprives the impugned measures of any s. 1 justification. They offer a hypothetical amendment to s. 148.1(1) that they say would render it facially consistent with s. 3:
An elector who
fails to prove his or her identity and residence in accordance with subsection
143(2) or (3)
or to take an oath otherwise required by this Act shall not
receive a ballot or and be allowed to vote upon making the
declaration provided in Schedule X.
The plaintiffs suggest that where an elector arrived to vote without the requisite documentation, this written declaration or oath would be administered by election officials at polling stations. The plaintiffs say that such a process, which is similar to that under the Ontario Election Act, would provide a better means of satisfying the legislative objectives than the impugned provisions. They argue that, by any measure, an oath is a superior guarantor of an elector’s identity than a fishing licence or gas bill, the production of which would satisfy the present requirements. The plaintiffs say it is noteworthy that Parliament considers an oath to be an adequate guarantee in a number of relevant circumstances, including: (a) where poll staff have “reasonable doubts concerning whether a person intending to vote is qualified as an elector” (s. 144); (b) there is a discrepancy between the name and address on the voters list and the elector seeking to vote (s. 146); and (c) a voter’s name has already been crossed off the list, suggesting either an administrative error or that the individual has already voted (s. 147).
 A proposed amendment to permit a statutory declaration alone as sufficient voter identification was raised during the Committee consideration of Bill C-31 and was ultimately rejected by the Committee.
 Canada counters that the plaintiffs’ proposal would not address the objectives of the impugned provisions. Any person could fabricate a name and address, and swear a statutory declaration. The fact that he or she left behind a piece of paper with a signature would be meaningless in that situation. If the person had used another person’s name and address or simply made up a name and address, they could not be tracked down. Canada argues that the deficiency in the plaintiffs’ proposed alternative is that there is no counterpart to the cross-referencing protection contained in the attestation process. This latter process requires two documents: a list of authorized signatures form, and the attestation of residence form. The administrator verifies in the attestation of residence form that the elector is a resident or receives services at the institution. The list of authorized signatures, signed by the administrator, is sent to the deputy returning officer. At the polling station, signatures are compared between the two forms.
 The question at this stage is whether Canada has established that there are no alternative, less drastic means of achieving both of Canada’s legislative objectives in a real and substantial manner.
 The plaintiffs submit that a “failsafe” provision of the nature they suggest (allowing an elector who arrives without the requisite documentation, and without another elector to vouch for him/her, to vote upon swearing a statutory declaration) is a less impairing but equally effective means of achieving the purposes of the legislation. I am not convinced that they are correct. The objectives of preventing fraud and enhancing confidence in the election system would not be as well met by a system with the suggested provision because such a provision would not prevent personation nearly as effectively. As well, I do not think that a scheme that permitted serial or multiple vouching, or vouching by persons from outside the polling district, would meet the objectives of preventing fraud and enhancing confidence in the election system as well as the more restrictive provisions for vouching that are now in effect.
 No doubt the list of documents authorized by the CEO could be further expanded or the vouching provisions could be made more flexible without completely destroying the effectiveness of the scheme, but the question is whether the scheme falls within a range of reasonable alternatives. As was stated in RJR-MacDonald, while the law must be carefully tailored so that the rights are impaired no more than necessary, the tailoring process seldom admits of perfection and the courts must accord some leeway to the legislator. Looking at the scheme as a whole, taking into account the broad range of options available for proof of identity and residence, and the delegation to the CEO of the ability to continue to fine-tune and expand that range of options, I am satisfied that there are no alternative, less drastic means of achieving both of Canada’s legislative objectives in a real and substantial manner. I conclude, therefore, that Canada has established that the provisions are minimally impairing.
iii. Are the Impugned Provisions Proportionate in their Effect?
 The final step in the proportionality analysis is to determine the proportionality of effects. In Oakes at 139-40 (cited to S.C.R.), Dickson C.J. explained the function of this step as follows:
Some limits on rights and freedoms protected by the Charter will be more serious than others in terms of the nature of the right or freedom violated, the extent of the violation, and the degree to which the measures which impose the limit trench upon the integral principles of a free and democratic society. Even if an objective is of sufficient importance, and the first two elements of the proportionality test are satisfied, it is still possible that, because of the severity of the deleterious effects of a measure on individuals or groups, the measure will not be justified by the purposes it is intended to serve. The more severe the deleterious effects of a measure, the more important the objective must be if the measure is to be reasonable and demonstrably justified in a free and democratic society.
 At paras. 76–78 of Hutterian Brethren, McLachlin C.J. discussed this stage of the analysis further:
It may be questioned how a law which has passed the rigours of the first three stages of the proportionality analysis – pressing goal, rational connection, and minimum impairment – could fail at the final inquiry of proportionality of effects. The answer lies in the fact that the first three stages of Oakes are anchored in an assessment of the law’s purpose. Only the fourth branch takes full account of the “severity of the deleterious effects of a measure on individuals or groups”. As President Barak explains:
Whereas the rational connection test and the least harmful measure test are essentially determined against the background of the proper objective, and are derived from the need to realize it, the test of proportionality (stricto sensu) examines whether the realization of this proper objective is commensurate with the deleterious effect upon the human right. ... It requires placing colliding values and interests side by side and balancing them according to their weight. [p. 374]
In my view, the distinction drawn by Barak is a salutary one, though it has not always been strictly followed by Canadian courts. Because the minimal impairment and proportionality of effects analyses involve different kinds of balancing, analytical clarity and transparency are well served by distinguishing between them. Where no alternative means are reasonably capable of satisfying the government’s objective, the real issue is whether the impact of the rights infringement is disproportionate to the likely benefits of the impugned law. Rather than reading down the government’s objective within the minimal impairment analysis, the court should acknowledge that no less drastic means are available and proceed to the final stage of Oakes.
The final stage of Oakes allows for a broader assessment of whether the benefits of the impugned law are worth the cost of the rights limitation. In Thomson Newspapers Co. v. Canada (Attorney General),  1 S.C.R. 877, Bastarache J. explained:
The third stage of the proportionality analysis performs a fundamentally distinct role. ... The focus of the first and second steps of the proportionality analysis is not the relationship between the measures and the Charter right in question, but rather the relationship between the ends of the legislation and the means employed. Although the minimal impairment stage of the proportionality test necessarily takes into account the extent to which a Charter value is infringed, the ultimate standard is whether the Charter right is impaired as little as possible given the validity of the legislative purpose. The third stage of the proportionality analysis provides an opportunity to assess, in light of the practical and contextual details which are elucidated in the first and second stages, whether the benefits which accrue from the limitation are proportional to its deleterious effects as measured by the values underlying the Charter. [Emphasis in original; para. 125.]
In my view, this is a case where the decisive analysis falls to be done at the final stage of Oakes. The first two elements of the proportionality test – rational connection and minimum impairment – are satisfied, and the matter stands to be resolved on whether the “deleterious effects of a measure on individuals or groups” outweigh the public benefit that may be gained from the measure. In cases such as this, where the demand is that the right be fully respected without compromise, the justification of the law imposing the limit will often turn on whether the deleterious effects are out of proportion to the public good achieved by the infringing measure.
 Thus, at this final stage, the task is to identify the salutary and deleterious effects of the legislation, and to assess whether the legislation’s deleterious effects are out of proportion to the public good it achieves or promises to achieve.
 Canada says that the impugned provisions address the issues of voter fraud and lack of confidence in the electoral system stemming from a perception by the electorate that fraud is prevalent or could affect the outcome of elections. Canada’s position is that the existence or perception of electoral fraud undermines many of the benefits that elections bring to democratic governance. It argues that the legislation’s salutary effects are that it ensures access to the electoral process while maintaining the integrity of that process.
 The plaintiffs, however, submit that Canada does not, on the available record, have an actual problem with voter fraud, except on an absolutely trivial level. They say that isolated instances of fraudulent voting under the pre-existing rules are cogent evidence that the system has sufficiently robust disincentives and protections. They also point to the existence of criminal sanctions for contraventions of the Act (see s. 483 of the Act) and to the competitive and adversarial nature of the electoral process, in which competing candidates and parties expend considerable effort to monitor possible efforts by their opponents to cheat or otherwise subvert the process.
 Thus, the plaintiffs say, the positive impact of the provisions is tenuous. They additionally submit that the fact that the provisions are not comprehensive and only apply to one aspect of the voting process, contradicts the allegedly pressing nature of the harms: for example, the legislative provisions are aimed only at one type of fraud – namely, personation – and they do not apply to special voting.
 Canada says that electoral fraud and abuse, while not widespread, do exist in the present system, but their extent is unknown because of the inherent secrecy involved in voting. Its position is that Parliament is justified in taking preventive measures, referring to Hutterian Brethren at para. 43, where McLachlin C.J. for a majority of the Supreme Court said that “governments are entitled to act in the present with a view to future developments”.
 I will briefly review some of the evidence that relates to the existence of electoral fraud or voting error in Canadian elections.
 There have been a few criminal convictions for electoral fraud over the years. In a letter to the Committee on Procedure and House Affairs in June 2006, the former CEO outlined the convictions on record between 1979 and 2006: one conviction for personation; four convictions for voting when not qualified or entitled; one conviction for inducing or procuring an unqualified voter to vote; and five convictions for voting twice.
 Section 517 of the Act permits the Commissioner to enter into a compliance agreement, in lieu of prosecution, where the Commissioner believes an offence under the Act has been, is about to be, or is likely to be committed. A compliance agreement is essentially a voluntary agreement between the Commissioner and the person shown to have contravened the Act in a situation where it is not thought necessary to prosecute. This option was added to the Act in 2000. Between that date and 2006, 42 compliance agreements were entered into with electors who voted twice.
 Following the 2000 election, Mr. Molnar deposed, the CEO initiated a small pilot project to investigate potential cases of alleged personation. For the purposes of the project, alleged personation was defined as instances in which an elector was required to take an oath because someone else had already voted in his or her name (according to the list of electors), or because the elector’s name had been inadvertently crossed off the list of electors. The CEO examined a sample of 786 poll books chosen randomly from a random sample of ridings. Alleged personation was observed in the case of 27 voters, representing 0.01% of all electors registered to vote in those polling divisions. These 27 cases did not lead to any enforcement measures, and remain “alleged” personations because it was impossible to identify on the basis of information available who, if anyone, voted under the elector’s name or whether administrative error by poll staff was the cause. The investigation would only have identified personation in cases where the legitimate voter went to the poll to vote.
 A similar review of 1033 randomly selected poll books from various electoral districts following the 2006 general election identified 84 potential cases of voter personation, representing 0.02% of the roughly 350,000 electors who would have been eligible to vote in those polling divisions.
 A few voting irregularities have been found on the rare occasions when audits or investigations have been done. For instance, Members of Parliament made a number of complaints regarding illegal voting in the Trinity-Spadina electoral district in the 2006 federal election. The CEO conducted an independent audit into the unusually large number of polling day registrations, which was 10,738. The audit, which looked only at polling day registrants and not at all 62,728 ballots cast in the electoral district, revealed: one instance from a sample of 344 in which a polling day registrant may have also voted in another electoral district; four instances out of 991 in which a polling day registrant may have voted more than once within the riding; and five instances in which a polling day registrant may have also voted at an advance poll. The high number of polling day registrations was probably explained by the presence of numerous post-secondary students in the riding, which encompasses the University of Toronto.
 In another example, complaints regarding the presence of business addresses on the list of electors in the district of Edmonton Centre during the 2006 general election led the Commissioner to investigate. While the Commissioner concluded that there was not “any hard evidence” of voter fraud or organized voter fraud, he did find a “systemic problem” with the lists of electors and individual cases of people voting in the wrong place.
 Elections Canada conducted internal studies to identify possible cases of double voting after both the 2000 and 2004 general elections. In 2000, there were 5,107 instances of potential double or multiple voting. The study sampled 408 of those cases from six electoral districts across the country. In the vast majority of cases, the evidence was inconclusive. The researchers found 15 cases that could probably be successfully prosecuted: three persons voted twice at an ordinary poll, five voted at an advance poll and an ordinary poll, and seven voted by special ballot as well as at an ordinary poll.
 With respect to the 2004 election, 299 cases of potential double voters identified by Elections Canada were referred to the Commissioner and a further 23 complaints were received from the public. The Commissioner reviewed all of them and the result was seven compliance agreements. There were no prosecutions.
 Canada says that not only fraud but also error can undermine the integrity of the vote. For instance, in 2006, the Commissioner entered into compliance agreements following the 39th general election with electors who held the mistaken belief that an elector who was registered to vote in one polling division and had a residence in another was entitled to vote twice.
 There have been some high profile instances involving journalists and students who voted fraudulently in order to demonstrate and publicize weaknesses in the voting system. In one instance, a freelance writer communicated to the media about securing three separate ballots during the 2004 general election. In a similar case that was prominent in the French-language media, a journalist revealed that he had voted twice in the 2004 general election, again to expose the potential for abuse in the system. The plaintiffs point out, however, that the new voter identification requirements would not have prevented the kind of fraudulent voting that was demonstrated, which involved identity theft or the false presentation of documents.
 Canada stresses that statistics regarding convictions for voter fraud do not reflect undetected instances of illegal or multiple voting. Comprehensive audits or verifications are rarely, if ever, done following elections, and the system of investigations for prosecutions is primarily a complaint-based one. According to Mr. McDougall, this impossibility of knowing the extent of the problem with any certainty is one reason Parliament chose voter identification as a means to reduce the potential or risk for voting fraud or irregularities, rather than relying solely on after-the-fact enforcement.
 Canada points, for example, to a statement by the current Commissioner of Canada Elections in his testimony before the Standing Committee on Procedure and House Affairs on February 8, 2007. The Commissioner, Mr. William Corbett, had been asked (in effect) whether the perception by elected Members of Parliament that there is a problem with voter fraud was inaccurate. He responded, referring to the investigation of particular complaints regarding voting in Desnethé-Missinippi-Churchill River, in Edmonton Centre, and in Spadina, as follows (at page 9 of the Proceedings):
On voter fraud, my conclusion wasn’t that we didn’t have any hard evidence. My conclusion was that there wasn’t voter fraud. There was no organized voter fraud, that’s for sure. There wasn’t some organization moving voters into one district who shouldn’t be voting there. There were individual cases of people voting in the wrong place, if you will. We found no evidence of double votes, which is important. People weren’t voting in one place and voting somewhere else, voting twice – none of that. It wasn’t just an absence of evidence, in my opinion; it was an indication that there wasn’t voter fraud.
The voter fraud we do find is in individual cases of double voting, wilful and otherwise, but not organized voter fraud. Those two investigations are examples of it’s not being there. The Spadina one is still ongoing, as you know, and I can’t comment on that. It hasn’t been resolved yet.
I’m not saying I’m investigating it, but the CEO’s inquiry is still ongoing. I don’t want to talk about investigations anyway, but I talked about the two cases I did because they are done. Is there voter fraud? Yes, there is. It’s individual, though. It’s people who are not qualified to vote voting, and some people voting twice, or trying to vote twice.
How much of a problem is it? I don’t see a major problem, frankly, but then I haven’t been around that long.
 Canada also points to the evidence of Mr. Côté, which is that there were a number of cases of voter fraud in Québec provincial elections, leading in 1999 to the imposition of the voter identification requirements, and to the evidence or Dr. Robert Pastor that there have been numerous instances of voter fraud in the United States.
 Regarding the effect of irregularities on the electoral process, the Government expressed concern in its Response to the Committee’s Thirteenth Report that even a small number of wrongly-cast ballots could harm the integrity of the process. Canada says that in close races, a small number of fraudulent votes could change the outcome of an election. Further, Canada says, the existence of fraudulent votes dilutes the weight of legitimate votes cast, undermining the voice and meaningful participation of legitimate voters and lending the appearance that their votes do not count. This, in turn, can impair confidence in, and result in disillusionment with, the legitimacy of the electoral process.
 Although, as the plaintiffs assert, the record shows that Canada has a remarkably “clean” voting system given the low number of convictions and compliance agreements arising from fraudulent or mistaken voting, I accept Canada’s submission that the harm of electoral fraud lies not only in its potential to impact the actual outcome of an election, but also in its capacity to undermine public confidence in the electoral process. As Dr. Pastor indicates, an electoral system cannot inspire public confidence without safeguards to deter or detect fraud or to confirm the identity of voters. Dr. Archer similarly discusses how the existence of voter fraud and perceptions of voter fraud negatively affect public confidence in the electoral process.
 The plaintiffs and the BCCLA argue, however, that there is no evidence of any public perception of abuse under the former identification regime.
 Dr. Archer deposes there seems to be very little direct empirical evidence of any kind about public attitudes toward electoral fraud in Canada. Thus, while there is no evidence that there is a public perception of abuse, it is equally correct to say that there is no evidence that the public is not concerned about potential abuse. Dr. Archer points out that the political science literature indicates that there need not be a large number of instances of electoral fraud in order for electoral fraud to be a matter of public concern and undermine public confidence. As well, he suggests that the surveys conducted following recent by-elections in 2007 and 2008 and the general election of October 2008 indicate overwhelming levels of support for the new requirements, and that it is a fair inference that some or all of the people who hold this view do so because the new requirements make electoral fraud less likely.
 Mr. McDougall deposes that based on the first-hand electoral experience of Members of Parliament and other party officials, there is a “sentiment that voter fraud clearly does occur and is much more widespread than reported”, which was the basis for the Thirteenth Report recommendation on voter identification. He quotes comments from representatives of multiple parties (the Conservative, Liberal and New Democratic Party) during committee meetings to that effect.
 In sum, what can be concluded about the salutary effects of the legislation?
 Counsel for the plaintiffs argued, correctly in my view, that there must be solid evidence of salutary effects – if there is no problem to solve, then the legislative solution crafted for the problem cannot have any salutary effects. Bear traps in Stanley Park cannot have a salutary effect in the absence of any possible presence of bears.
 The evidence shows no history of systemic electoral fraud in Canadian federal elections and, indeed, no history of a problem of any magnitude with respect to electoral fraud or mistaken double voting. However, electoral fraud and mistaken voting are not wholly unknown. Votes have been, on occasion, cast by persons not entitled to vote as they did. Because it has been a complaint-driven process, it is not possible to know the full extent of fraudulent or mistaken voting.
 I find that there is sufficient evidence in this case to conclude that the legislation has some salutary effects.
 First, I conclude that the legislation makes it significantly more difficult to perpetrate electoral fraud through personation. Rather than simply announcing that one is “Jane Doe” (an elector), one must have the documentary evidence required, or a person willing to be a voucher, in order to establish one’s identity as “Jane Doe”. The legislation also makes it significantly less likely that someone will be able to vote erroneously in the wrong polling district or to vote twice. The potential for fraudulent or mistaken voting clearly existed in the previous system and, from time to time, was realized.
 Second, the voter identification measures provide some reassurance to those who are concerned about electoral fraud, and thereby would tend to enhance confidence in the integrity of the electoral process.
 The move from an “honour system”, where electors were permitted to vote simply by stating their names at the polling station where they are registered, to the new system where an elector’s identity is established before he or she obtains a ballot, will make the rare events of voter fraud through personation even rarer. Similarly, the move will reduce the number of instances (probably already very small) where a voter casts a second ballot by error or votes in the wrong polling district. Such reductions are salutary effects.
 To the (unknown) extent that members of the public at large have been concerned about the integrity of the electoral system, the legislation should have the effect of reducing such concern. The high levels of public support for the voter identification legislation (to be seen in the evidence discussed in the next section of these Reasons) probably indicates the existence of public concern about the integrity of the electoral system. In addition, the fact that Members of Parliament and party officials were concerned about electoral fraud is some evidence of public concern about the matter. Members of Parliament are, after all, representatives of the public and have first-hand experience in working within the electoral system. Reducing the concerns of Members of Parliament and increasing their confidence in the electoral system is likely to have some positive impact on public confidence in the electoral system. Those are also salutary effects.
 The likely effects of the legislation will be to prevent future instances of voter fraud or mistaken voting, to reduce the previously very small number of cases to an even smaller number, and to maintain or enhance public confidence in the electoral system. The salutary effects relate to a core element of the electoral system: the integrity of the vote, including the principle that each qualified citizen has the right to cast one vote in the district of his or her ordinary residence.
 The question will be whether the salutary effects (the benefits flowing from the legislation) are proportional to the deleterious effects (the costs). I turn next to an assessment of the deleterious effects.
 The defendant Canada argues that the only possible deleterious effects of the voter identification requirements are that they may impose a trivial or insubstantial burden on electors: to make reasonable efforts to obtain appropriate identification or someone to vouch for them.
 Canada submits that there is no evidence that the voter identification requirements will lead to a lower voter turn-out and relies on the evidence of Dr. Archer. He describes the long-term significant decline in voter turnout in Canada (from an average of 74.8% over 1949-1988, to 58.8% in 2008). He identifies age as the most important factor overall in the decision to vote, with young voters much less likely to vote than their older counterparts. The decline between 2006 and 2008 (when the voter identification requirements came into effect) was from 64.7% to 58.8%, respectively, but 2006 had marked a jump from the previous four federal elections (from 1993–2004), where the voter turnout was 70.9%, 67.0%, 64.1% and 60.9% in successive elections. He states the opinion that the available evidence shows no causal relationship between the voter identification requirements and voter turnout. He further states that “the change in voter identification requirements did not have a consistent negative impact across federal constituencies in Canada on the basis of the size of the young, elderly, residentially mobile, Aboriginal, visible minority or low income groups within constituencies.”
 Canada also submits that alleged deleterious effects on “third parties” should not be considered under the s. 1 analysis because the government is only required to justify infringement of the plaintiffs’ rights, and is not required to justify hypothetical infringements. Canada emphasizes that the plaintiffs were in fact able to vote in the last election (including Ms. Eddelstone, it says, because she possessed the necessary documents but simply failed to bring them with her to the polling station).
 The plaintiffs say, on the other hand, that in the s. 1 analysis the court can look at the systemic impact of the impugned measures, whether or not that impact constitutes a s. 3 infringement. Mr. Quail urges that a close attention to context is necessary, and that proportionality can only be evaluated through close attention to detail and factual setting. He refers to Thomson Newspapers at para. 87, discussed above.
 I think that the comments made by the Chief Justice for the majority in Hutterian Brethren at para. 69 provide the short answer to Canada’s submission: the court’s perspective is to be societal, not limited to the impact on the individual claimants. Chief Justice McLachlin wrote:
... Laws of general application affect the general public, not just the claimants before the court. The broader societal context in which the law operates must inform the s. 1 justification analysis. A law’s constitutionality under s. 1 of the Charter is determined, not by whether it is responsive to the unique needs of every individual claimant, but rather by whether its infringement of Charter rights is directed at an important objective and is proportionate in its overall impact. While the law’s impact on the individual claimants is undoubtedly a significant factor for the court to consider in determining whether the infringement is justified, the court’s ultimate perspective is societal. The question the court must answer is whether the Charter infringement is justifiable in a free and democratic society, not whether a more advantageous arrangement for a particular claimant could be envisioned. [Emphasis added.]
 It is in looking at the deleterious effects of the legislation from a societal perspective that the survey evidence filed in these proceedings by the CEO becomes relevant.
 Elections Canada commissioned evaluations of the implementation of the voter identification requirements in the seven federal by-elections and one general election held since June 2007 when the requirements came into effect. The plaintiffs rely on the results of these evaluations, not to prove that particular individuals were precluded from voting by the requirements, but to provide a “tally of the detriments” for the purposes of this stage of the s. 1 analysis. They refer to para. 77 of Hutterian Brethren, where the Court states that “[t]he final stage of Oakes allows for a broader assessment of whether the benefits of the impugned law are worth the cost of the rights limitation”. The plaintiffs argue that this entails a broad analysis that weighs the benefits of the harm of the measures on a societal level, “taking ten paces back and looking beyond the confines of the legislative objectives of the rules”. Impacts which do not amount to the absolute prevention of voting but that still discourage participation in the electoral process are examples of relevant detriments in this context. They submit that even though the surveys were not undertaken with the present litigation in mind, the Court is still entitled to glean what assistance it can from the available record of the experience in recent elections.
 I accept the plaintiffs’ position that the proportionate effects analysis, as discussed in Hutterian Brethren, permits this broader consideration of effects and that the survey evidence is relevant in this respect. Before reviewing some of that evidence, however, I will briefly address the expert evidence of Dr. Corbin and Dr. Archer that Canada tendered with respect to the interpretation of the surveys.
 With respect to the quantitative surveys (by Créatec, Environics, Strategic Counsel and Leger Marketing), Dr. Corbin does not take issue with the credentials of the researchers and states the opinion that the survey reports in evidence were conducted with accepted standards of quality control. However, she says, they cannot be used to draw inferences about individuals. She says that the surveys were designed to draw conclusions about large populations, and even then, only within a given margin of error, and with imperfect confidence.
 For the purposes of her opinion, Dr. Corbin defines “impedance to vote” as “the inability to reasonably access acceptable identification documents, or to find a person to vouch for the elector”. She states that if any individuals exist who were impeded from voting, they are not detected in the survey results. Rather, she says, the surveys do not produce a single definitive case of a person whose survey responses allow one to infer a structural barrier to voting.
 Dr. Corbin states that interpretation of the surveys reveals that the voting process went smoothly and successfully. She says that individual responses to isolated questions, to the effect that the identification requirements posed any sort of problem, were inconclusive as to what sort of problem was encountered and whether the problem was solved. Dr. Corbin states that some of the questions induced participants to name the new voter identification requirements as a problem, even if they did not unreasonably impede the participants from voting. In short, Dr. Corbin concludes that the surveys themselves are inconclusive as to whether the voter identification requirements created an impediment to voting.
 Dr. Corbin adds that the surveys draw no statistical conclusions about potentially marginalized groups, such as the homeless, Aboriginal Canadians, the visually challenged or rural residents. While one or more of the statistical surveys obtains information on the Aboriginal status of certain participants, there is no statistical analysis that permits one to make projections about the level of disadvantage that might be encountered among Aboriginal people. She says that the method of sampling Aboriginals, by way of non-representative “over-samples”, confirms that no reliable statistical projections can be made.
 The thrust of Dr. Archer’s evidence is to similar effect. He says that the survey data do not provide the level of detail necessary to permit the conclusions the plaintiffs seek to have the Court draw regarding the number of people who were unable to vote as a result of the identification requirements.
 Dr. Archer identifies a number of difficulties in making a definitive statement about whether the identification requirements precluded qualified electors from voting. He explains that in each of the surveys under consideration, a very small number of respondents reported that they had not voted due to identification issues. The questions used in the surveys did not probe this topic to the degree of detail that would be necessary to make definitive statements about whether any voters were precluded from voting due to the new requirements. For example, there is insufficient information regarding the level of knowledge of the voter identification requirements among those who cite this as a factor. Further information on the degree of awareness, understanding and experience with the vouching option is also necessary to fully appreciate the impact of the new voter identification requirements. Added to these factors, the survey questionnaires did not include the level of additional probing of respondents that would permit the kind of nuanced understanding that such a complex topic requires. Consequently, Dr. Archer says that one is left to “read into” the responses inferences about the motivations that lie behind responses such as not voting due to “voter identification issues”. Finally, in a number of instances, the wording of the question or of the response categories is such that the independent effect of the voter identification requirements cannot be isolated. All of these factors combine to lead Dr. Archer to conclude that, on the basis of the surveys he reviewed, he is “not able to determine that any qualified elector was actually precluded from voting due to the new voter identification requirements”.
 The plaintiffs challenge the evidence of both Dr. Corbin and Dr. Archer, arguing that “preclusion” of qualified electors from voting is a higher degree of impairment than is necessary when considering the detrimental effects of the impugned provisions. They submit that effects that do not amount to such a preclusion of voting but nevertheless discourage participation in the electoral process are examples of detriments that ought to be put into the balance.
 I accept that the creation of conditions that discourage voters (but do not absolutely prevent them) from exercising the franchise does constitute a detrimental effect.
 I accept the plaintiffs’ submission that the Court is entitled to consider a broader range of effects at this stage of the analysis than simply whether there has been an “impedance to vote”, as defined by Dr. Corbin, or whether any qualified voter was precluded from voting, as discussed by Dr. Archer. Thus, while I accept the experts’ evidence with respect to the limitations of the survey data, I nevertheless find that the survey evidence provides an overall picture of the general effects of the new identification requirements that I should consider.
 I turn now to some of the specific findings of the surveys regarding the implementation of the voter identification requirements.
 Two sets of by-elections were held between the effective date of the new identification requirements in June 2007 and the general election in October 2008. The first set of by-elections occurred on September 17, 2007 in three electoral districts in Québec. The second set took place on March 17, 2008 in the four electoral districts of Toronto Centre, Willowdale, Desnethé-Missinippi-Churchill River and Vancouver Quadra. Elections Canada commissioned evaluations of the implementation of voter identification in these seven by-elections.
 The evaluation of the September 2007 by-elections was conducted by Environics Research, “Application of the New Voter Identification Procedures Under Bill C-31 for the September 17 Federal By-Elections in Québec: Evaluation Synthesis”, January 15, 2008. This report was based on data gathered in two studies commissioned by Elections Canada: Les Études de Marché Créatec, “Evaluation of New Voter Identification Requirements at the September 17, 2007 Federal By-Elections (Electors and Election Officers)”, October 2007, and Impact Research, “Outremont / Roberval Lac-St. Jean / St. Hyacinthe-Bagot September 17th by elections – Final Report.”
 The key findings from the research were as follows. More than 90% of electors had a positive attitude towards the idea of voters having to prove their identity, with two-thirds being very positive. Only 2% were “very negative”. The tally sheets kept by elections officials showed that 80% of voters used an “Option 1 ID” (usually a driver’s licence) while most of the remainder used an “Option 2 ID”. Only 1% of voters were “vouched”. Of the 73,000 counted, 159 electors (or about 0.2%) left without voting. Virtually all electors who went to the polling station generally felt that meeting the new requirements was easy, with 95% considering it somewhat easy and 71% considering it very easy. About 1% reported that they did not have the correct identification, but went home, got it, returned and voted. Only six percent of electors who went to the polling station reported any specific problem with the identification requirements. The report indicates that no one specific problem stood out.
 Of the election officers surveyed, 94% felt that the overall process of voter identification went well with 64% saying it went very well. Lack of identification and lack of proof of residential address were the most common problems reported, yet 68% of election officers reported that they did not personally experience any specific problems related to voter identification and 67% of deputy returning officers reported that they never had to refuse a ballot for reasons related to the new identification requirements. Overall, deputy returning officers refused an average of less than one ballot each for reasons related to the new identification requirements.
 The Report of the CEO regarding the September 17, 2007 By-elections summarizes the findings regarding this first experience with the voter identification requirements as follows:
In broad terms, the findings indicate that the implementation of the Bill C-31 identification requirements went well. Some of the key findings of this evaluation were:
· More than 9 in 10 voters reported their attitude to the new identification requirements as positive; 2 percent characterized their attitude as “very negative.”
· Three percent of voters felt that voting took “a lot longer” as a result of the new identification requirements.
· Our communication efforts worked well in Saint-Hyacinthe-Bagot and Roberval-Lac-Saint-Jean, but slightly less in Outremont, where lower recall rates were observed.
· More than 9 in 10 voters expressed that they felt very well informed about the new identification requirements before going to the polls; in contrast, poll staff expressed a somewhat less positive view of voters’ level of understanding of the requirements.
· Virtually all those who went to the polls reported having proper identification on arrival; 6 percent reported a specific problem with identity verification.
· About one third of voters considered using their voter information card as identification, although the card clearly states that it is not an identification document.
· About 8 in 10 voters used a single photo identification showing name and address, most often a driver’s licence. Most of the remainder used two documents with no photo, both showing name and one showing address. One percent of electors swore an oath and were vouched for by another elector.
· Almost all election officers felt that the overall process of voter identification went well, and two thirds felt it went very well. Two thirds of election officers did not personally encounter any specific problems related to voter identification.
· About two thirds of deputy returning officers never had to refuse a ballot because of an identification-related issue, and 15 percent had to do so more than once. Based on the tally sheets filled out by deputy returning officers to track identification documents used by voters, about two in one thousand electors (0.2 percent) could not satisfy the requirements. Our survey of electors indicated that 0.5 percent reported having left the polling station without voting because they could not satisfy the new requirements.
Elections Canada expects to use these results to guide its communications approach in future electoral events; however, because electors in Québec have been required to show some form of identification in the past few provincial elections, their expectations – and consequently their reactions – may differ from those in other provinces.
 With regard to the discrepancy between what the tally sheets indicated (0.2 percent of electors in the 2007 by-elections in Québec were recorded as unable to satisfy the identification requirements) and what the post-election survey of electors indicated (0.5 percent of those surveyed reported leaving the polling station without voting because they were unable to satisfy the identification requirements), Dr. Archer suggests, and I accept, that the data from the tally sheets kept by elections officials are more likely to be accurate than the survey data because the survey data are based on a sampling population and are more likely subject to random and non-random error. He also points out that, of the 0.2% of electors refused a ballot, it is unclear what proportion returned with documentary identification and voted.
 Environics Research conducted an evaluation of the March 2008 by-elections for Elections Canada: Environics Research Group, “Evaluation of New Voter Identification Requirements, By-elections, March 17, 2008”, June 2008. The report summarized two surveys in four ridings (Toronto Centre, Willowdale, Desnethé-Missinippi-Churchill River, and Vancouver Quadra) designed to evaluate the new identification requirements: one survey was of electors and another of elections officers. Both used a quantitative methodology based on telephone interviews. For the survey of electors, 3,203 interviews were conducted among a random sample of eligible voters and an oversample of 100 eligible voters aged 18 to 24 in each of Toronto Centre and Vancouver Quadra. The margin of sampling error for the total sample was +/‑1.7 percentage points, 19 times in 20. For the survey of elections officers, a total of 1002 officers were interviewed. The margin of sampling error for this sample was +/‑3.1 percentage points, 19 times in 20.
 Among the findings regarding the impact of the new identification requirements on the voting process were these. Across the four electoral districts, 93% of respondents described their attitude toward the idea of electors having to prove their identity when voting in a federal election as at least somewhat positive, with 75% describing it as very positive. Attitudes in Desnethé (a riding in northern Saskatchewan with a large Aboriginal population) were noticeably less positive, with only 53% very positive towards the idea. With respect to those with a negative attitude, the report explained as follows:
The small group (7%) of voters with a negative attitude toward the idea that electors must prove their identity when voting in a federal election were asked why they feel this way. The most commonly cited reasons focus on the idea that the new requirements are unnecessary, either because the elections staff know the voters (mentioned by 20%, or just over 1% of all voters) or because the [voter information card] or registration list should be sufficient (15%, or just under 1% of voters). Just under two in ten (16%) cite the inconvenience of the new requirements, while 13 percent indicate the requirements discourage people from voting. In addition, just over one in ten (11%) indicate that they cannot meet the new requirements.
 Across the four districts, 93% of voters surveyed reported no difficulties related to the new requirements. Four percent reported not having the required documents. Voters in Desnethé were more likely (10%) than those in other districts (average of 2%) to report not having the required identification documents. Of the four percent overall who did not have the required documents with them, the most common response was that they did not have a document with their address (45%, or 2% of all voters). Some 32% (or 1% of voters) reported having no identification at all, while 23% (or 1% of voters) reported not having a document with a photograph.
 Six in ten among the small group who did not have the required documents either went home to retrieve appropriate documents or swore an oath and were vouched for. Only one in 20 (less than one-half of one percent of total voters) reported not voting as a result of not having the proper identification when first arriving at the polls.
 Across the four districts, 94% considered it easy to meet the new requirements, with 80% indicating it was very easy. Only 70% of voters in Desnethé considered it very easy. Aboriginal voters were also less likely to indicate that it was very easy to meet the new requirements: 56% compared to 82% of non-Aboriginals.
 With respect to the survey of election officers, 31% and 39% reported that electors reacted very or somewhat favourably to the new requirements, respectively, and 20% and 8% said electors reacted somewhat or very unfavourably, respectively. In Desnethé, 41% of the election officers perceived favourable reactions, while 59% perceived unfavourable reactions.
 In all districts except Desnethé, three-quarters of election officers felt that electors were generally well-informed about the new requirements.
 In order to gauge whether the new identification rules were implemented with relative ease on voting day, the survey asked election officers whether they experienced any specific problems verifying the identity and address of electors. In response, 70% claimed not to have experienced any problems, while 29% had experienced problems. Those in this latter group were asked to describe the problems they had experienced, through an open-ended question. At page 37, the report discussed the responses:
Two types of responses top the list of specific problems related to identity and address verification according to elections officers. The first of these, cited by nearly one-half of those who experienced a problem, was a lack of proper identification (46%). The second, cited by four in ten (40%), was a general verification issue, including a piece of identification that was not up-to-date. Other types of problems, each mentioned by roughly one in ten election officers who experienced a problem include confusion about the new requirements/not informed (14%), name/registry discrepancy on the voters list (12%), issues with new procedure (12%), electors not returning to vote/leaving without voting (10%), people upset about having to leave to get identification (9%) and voters having to be vouched for (7%). Less common problems included not being able to accept Voter Information Cards (5%), a language barrier (3%) and staff not following procedures/training (3%).
 The survey also addressed the topic of electors being refused a ballot. At page 40, the report explained:
Deputy returning officers were asked to estimate the number of electors to whom they had to refuse a ballot because they did not meet the new identification requirements. A majority of six in ten deputy returning officers (58%) claim not to have refused a ballot to any voters, while the remaining four in ten say they refused ballots to either one or two voters (22%) or else three or more (18%). The average estimate given by deputy returning officers is two electors (the median estimate is zero electors).
[Emphasis in original.]
 Mr. Molnar states that Elections Canada also undertook evaluations and consultations with different groups (northern Canadians, seniors, residents in long-term care facilities, students and the homeless), in order to identify specific barriers to voting that they might face because of the new identification requirements, determine the effectiveness of the list of identification documents, and determine how Elections Canada could best inform them about the new requirements. It received two draft reports outlining views and recommendations regarding those communities, to be posted on the Elections Canada website once finalized. It also asked the returning officer for Toronto-Danforth to prepare a report regarding specific problems for homeless voters. The report (“Homeless Elector Research Project”) was submitted a few months before the 40th general election.
 Mr. Molnar deposes that surveys following the September 2007 and March 2008 by-elections revealed similar results:
... among those who went to the polls, 0.5 percent reported that they could not vote because of the voter identification requirements. Furthermore, in the March 17, 2008 by-elections, 38 percent of survey respondents said that they had brought their voter information cards to the polls. While 2 percent reported using it as a single document to identify themselves, 16 percent said that they had produced it along with another acceptable document.
 The CEO summarized the findings from the research regarding the experience in the seven by-elections as follows at pages 42-43 of his Report on the 40th general election:
Evaluations prior to the 40th general election
Elections Canada evaluated the implementation of voter identification for the seven by-elections held after the adoption of the new identification requirements and before the 40th general election. The evaluation involved public opinion surveys and surveys with election officers. Survey results indicated that the implementation of the new voter identification requirements went smoothly overall. Over 90 percent of electors were aware of the new requirements and had a positive attitude toward the idea of proving their identity. Over 94 percent of voters indicated that they found the new requirements easy to meet. Some respondents said they did not vote because they lacked proper documentation (4 percent). A similar proportion of respondents (4 percent or lower) indicated that they did not have the required documents when they arrived at the polls. While most of them returned home to retrieve their identification or swore an oath and were vouched for, 0.5 percent finally did not vote.
The results from respondents in the northern Saskatchewan electoral district of Desnethé-Missinippi-Churchill River were an exception to these trends. Only 75 percent indicated that they were aware of the new provisions and also felt positive about the requirements to prove their identity and residence. Also, the proportion of electors without the required documents increased to 10 percent in this riding.
A detailed account of the results of the surveys carried out after the by-elections can be found on the Elections Canada Web site at www.elections.ca.
Anecdotal reports from election officers indicate that meeting the identification requirements was particularly challenging for some population groups, including Aboriginal people living on reserves and seniors in long-term care facilities.
 The Report of the CEO on the 40th general election describes at pp. 21-24 the outreach campaign prior to the 40th general election, whose theme was “Vote. Shape your world” and whose main message was “When you vote you must prove your identity and address”. Advertisements, including those for Aboriginal and ethnocultural communities, were carried on 144 television and 629 radio stations, and in 145 daily newspapers, 1,114 community newspapers, 22 cultural publications and 97 student papers. Advertisements also ran on 1,977 movie screens and banner ads appeared on 280 internet sites. He states that, based on industry standards, the campaign potentially reached 99.9% of people living in Vancouver. He further states that Elections Canada undertook specific outreach initiatives with respect to ethnocultural communities, electors with special needs, Aboriginal electors and students. He describes a number of other initiatives to inform voters of the new identification requirements including: flyers to all Canadian households; posting in several languages on Elections Canada’s Web site; providing information in large print and Braille; and employment of Community Relations Officers for specific communities, including youth, the homeless and seniors.
 Mr. Molnar deposes that Elections Canada will be undertaking a comprehensive review of all consultations, research and recommendations arising from the by-elections, from the 40th general election and from its own initiatives, and will decide in due time on required changes to its procedures and practices, including further public education and outreach initiatives.
 Elections Canada commissioned the Strategic Counsel to undertake a survey of electors (both voters and non-voters) following the October 2008 general election: The Strategic Counsel, “Survey of Electors following the 40th General Election” March 2009. The survey consisted of telephone interviews conducted with a representative sample of 3,348 Canadian electors (including 627 non-voters) between October 22 and November 17, 2008. The margin of error was within +/‑1.96 percentage points, 95 times out of 100. A particular focus of the survey was an assessment of the impact of the new voter identification requirements on the electoral process.
 The lengthy report explained the results as follows at pages 38 and 40:
While some did not bring their Voter Information Card, virtually all (98%) voters report bringing the required identification with them. The exception here is Aboriginal Canadians, among whom slightly fewer (89%) report having brought the appropriate identification.
Of the national sample of 2,500 Canadians interviewed and the 1,844 who report having gone to a polling station, just 42 respondents report being turned away because of incorrect, or insufficient, identification. For some it was a complete lack of identifying documents (n=13), for others no document with their address (n=10), while for others it was no document with a photo (n=6), and for still others the lack of a driver’s licence (n=5). In fact, the last two cases cited above of incorrect, or insufficient identification, can probably be considered as a misunderstanding of the requirements either by the electors or by elections staff who turned them away. Neither a photo ID nor a driver’s license were a mandatory requirement.
Among those 42 intending voters, the most common response (n=14) was to swear an oath regarding their identity before being allowed to vote. Others (n=8) reported going home to get the appropriate documentation and coming back to vote. Some appear to have voted without producing the required identification (n=4), while a very few (n=8) went home and did not vote. Overall, this suggests that the new requirements were an inconvenience for a few, but in virtually all cases were not a barrier to voting for those who went to cast a ballot.
The final evidence that the new identification requirements did not constitute a widespread impediment for voting comes from the fact that almost all (97%) electors report finding it easy to meet the new identification requirement; with over eight-in-ten (83%) saying it was “very” easy. However, as might be expected, those groups that report some difficulty in producing appropriate identification are somewhat less likely to be as positive. In particular, Aboriginal Canadians (66%), youth (72%) and those with annual household incomes of less than $20,000 (71%) are well below the average in reporting that it was “very” easy to meet the new identification requirements.
 The Strategic Counsel’s report also described the survey results with respect to the reasons given for not voting, in response to an open-ended question. Of those voters (8%) who identified issues related to the electoral process, 1% mentioned not having received a Voter Identification Card, 1% that they were unsure if they were registered, and 1% (eight respondents) mentioned a lack of proper identification. Of Aboriginal respondents, 4% cited a lack of identification documents, 3% mentioned being unsure if they were registered, and 3% that they had not received a Voter Information Card.
 The report comments on the eight persons who said they were unable to vote because of a lack of appropriate identification, as follows:
[W]hile the 8 electors only represent 0.32 percent of the population, when projected to the proportion of electors who turned out for the 40th General Election (59%), then approximately 43,000 potential electors could have been turned away because of the new identity requirements. When the margin of error is considered; then this would yield a range of 2.28 percent down to zero, which translates into a range for those who could have been turned away of from 310,000 down to zero.
 Dr. Corbin takes issue with this and says that the Strategic Counsel researchers made a statistical error because a sub-sample of just eight persons is too small to be reliably subjected to such arithmetic. Further, she says, there were in fact not eight persons, but only six (because a “weighting” program was used since the demographics of the obtained sample did not quite match the demographics of the Canadian population). Finally, she notes that there was a factual problem reconciling the raw data provided by Elections Canada in a computer file with the Strategic Counsel report and that she was advised by the representative of Strategic Counsel that the six people were not necessarily ones who had actually gone to the polls and been turned away.
 Elections Canada also commissioned Leger Marketing to compile the “Survey of Election Officers – 40th General Election” to analyze the perspective of election officers regarding the proceeding of that election. The information was gathered through a telephone survey administered to a total of 3,115 officers who had worked in polling locations across Canada. The margin of error was +/‑1.8%, 19 times out of 20. The survey was published in March 2009.
 An average of 95% of election officers reported that the identification process went well, though more than one in two (56%) reported that the new requirements slowed the process down. Voter reaction to the new requirements differed significantly between regions. Less than one deputy returning officer in five reported experiencing problems when verifying the identity (19%) or address (16%) of voters. One in four deputy returning officers in Manitoba, Saskatchewan, British Columbia and the North reported experiencing specific problems validating voters’ addresses. When asked to describe them, they reported that the largest problems were with electors showing up with either improper identification (45%) or outdated addresses on their identification documents (15%). Of the sample, 8% reported that it was a common problem for electors to be refused their ballot due to inability to satisfy the voter identification requirements.
 In regional post-mortems held with returning officers after the 40th general election, some returning officers reported anecdotal evidence of problems at the polls, especially for particular groups of electors, including students and seniors in long-term care facilities. Representatives of the registered parties expressed mixed feelings about the new voter identification rules to the CEO. In de-briefings held after the last general election, some reported that the new rules caused problems for some electors, especially Aboriginal electors, students, residents of long-term care facilities and transient populations. Others, however, were unaware of any issues or problems.
 In February 2009, the CEO submitted his Report to Parliament regarding the 40th general election. With respect to the experience during the election, the CEO wrote at pages 43 and 45:
Issues faced by electors at the polls in the 40th general election
Our communications and outreach activities undoubtedly reached many electors, but there were still difficulties at the polls in the 40th general election.
Electors without required pieces of identification
We received anecdotal reports that some electors were unable to present pieces of identification and official documents that could prove their residence.
It appears that some electors relied on an attestation of residence (that is, a letter from the administrator of a homeless shelter) but were unable to supply an additional piece of identification as required. In some cases – especially in seniors’ residences – guardians or family members assume responsibility for an elector’s identification documents. These pieces of identification were not always readily available on voting day.
Similarly, residents in long-term care facilities often no longer carry acceptable identification. For this reason, the Chief Electoral Officer expanded the list of acceptable identification documents to include bracelets issued by these facilities, provided the elector was wearing the bracelet when voting.
Confusion about procedure
Some electors did not understand why certain types of identification were not sufficient. In the case of passports, for example, many electors did not appreciate that the issuing authority does not include address information, which is frequently filled in by hand by the passport holder.
Some electors complained that it was redundant to require that they prove their identity and residence after they had made a verbal declaration. They also noted that they could have been overheard by people who were not under oath to protect the privacy of their information.
Seniors were upset at having to show identification documents proving their residence when they were voting at a mobile poll set up in the long-term care facility where they lived.
Attestations of Residence
Some electors could not provide any proof of residence. Examples included persons residing in homeless shelters, student residences and long-term care facilities, or on Aboriginal reserves. In these cases, the elector could request an attestation of residence from the administrator of the facility where that person lived, confirming the elector’s place of ordinary residence. This provision offered many electors the opportunity to vote but also gave rise to several issues:
· The task of completing and signing attestations proved burdensome for many administrators of student residences and long-term care facilities.
· There was some confusion about how to provide the attestation of residence. In one case, a returning officer was given the entire list of persons living in a university’s student residences. This raised privacy concerns.
Elections Canada is evaluating the implementation of the new voter identification requirements. With the findings from the evaluation as well as the consultations preceding the 40th general election, we expect to be able to identify administrative improvements that Elections Canada can make or propose to Parliament.
 The plaintiffs also refer to other evidence and reports in demonstrating what they say are some of the detrimental effects of the impugned provisions. I will review some of that evidence.
 Following the 40th general election, Elections Canada prepared an internal analysis of complaints received from electors regarding the new identification requirements (“Analysis of Elector Complaints Instigated by the New Identification Rules”). From the 197 letters it received relating to the new identification policy, Elections Canada identified 13 major issues:
(a) Difficulties in producing acceptable pieces of identification;
(b) Polling day workers did not fully understand the new identification rules;
(c) New rules disenfranchised electors;
(d) Refused pieces of identification;
(e) Electors concerned about their privacy;
(f) Don’t understand the need for identification;
(g) Identification rules should be adapted to specific types of electors (rural areas, students, seniors);
(h) Asking for identification caused hard feelings in small communities;
(i) Insufficient provision in the identification policy to allow an elector to vote;
(j) More public education is needed;
(k) Acceptable forms of identification are too limited;
(l) Difficulty understanding what pieces of identification to provide; and
(m) National voting card should be issued by the government.
 The most prevalent concern expressed was with respect to producing acceptable pieces of identification. The report (dated I 2009) states that electors residing in rural areas with only mailing addresses were greatly affected because they did not have documents showing a current residential address. Senior voters comprised the second largest group of electors who wrote to Elections Canada. The main issue was that people who reside in seniors facilities generally do not have driver’s licenses, do not receive bills and frequently leave identification documents with family members. Seniors may also have limited mobility, so returning to the polling station with appropriate identification documents can be onerous or impossible.
 Others who reported difficulties producing acceptable identification were students, women voters who have changed their names, Aboriginal voters and people with disabilities.
 One group that is particularly vulnerable to disenfranchisement is the homeless. Ms. Henry and Mr. Wright, two of the plaintiffs, gave evidence about the challenges faced by homeless people in obtaining and keeping identification.
 The evidence of Ms. Graves mirrors that of Ms. Rose and Mr. Wright. As mentioned earlier, Ms. Graves is the Tenant Assistance Co-ordinator for the City of Vancouver, and thus is the City’s “responder” to low income tenants and the homeless. Her work includes assisting homeless people to obtain registration on voters’ lists and to vote in elections at the federal, provincial and municipal levels. She identifies the requirement that they indicate a place of ordinary residence as one of the most significant issues for homeless citizens wishing to vote.
 With reference to the list of CEO-approved documents, Ms. Graves deposes that few homeless people possess even the most commonly-held documents listed. For instance, few, if any, receive utility bills. A great many do not have accounts with financial institutions and rely on payday lending outlets if they have cheques to negotiate. They have no leases or mortgages, and receive no property tax assessments. In her experience, a significant number of homeless people have no reported income and do not file tax returns. It is very common that they have no identity documents of any sort, frequently because such documents have been stolen.
 Regarding the attestation of residence option, Ms. Graves’s evidence is that there are a few facilities that provide free meals in the Downtown Eastside that might fit the description of “soup kitchen”. In such facilities, people line up to be provided with a meal. In her experience, such facilities do not record the names of their patrons, many of whom remain anonymous to staff and volunteers. A homeless person does not “reside” at a shelter, but stays there occasionally according to the rules of the facility.
 It appears, however, from the evidence of Clyde Wright, that Elections Canada has implemented the letters of attestation provision in a way that goes well beyond soup kitchens and shelters, and extends to social service facilities, such as Pivot Legal Services, as agencies that can issue letters of attestation.
 The plaintiffs also rely on the report of Peg Lahn, a returning officer for the Toronto-Danforth electoral district, entitled “Homeless Elector Research Project” and commissioned by elections Canada prior to the 40th general election. The objective of Ms. Lahn’s project was to identify strategies that could allow Elections Canada to more effectively address identification-specific problems for homeless electors.
 Dr. Corbin correctly observes that Ms. Lahn did not purport to conduct a scientific survey, and did not interview homeless individuals. Dr. Corbin reports that Ms. Lahn did speak with Elections Canada personnel, a Registrar General of Ontario employee and some representatives of social services agencies, but did not retain documentation of these interviews.
 Ms. Lahn’s general findings were that identification-related issues persist as a challenge for homeless electors. In some cases, she said, the barriers are real, while in others they are perceived but with an equally limiting impact on voter participation. She stated that her research confirmed that the breadth of the list of documents accepted at the polls is sufficiently comprehensive that many electors without traditionally recognized identification will be able to exercise their right to vote, but that perceptions of identification-specific barriers to voting are broadly held. She wrote that proof of residence persists as a barrier to voting for people with no fixed address but that the attestation of residence procedure, if effectively communicated and adopted, could reduce this obstacle for many.
 Ms. Lahn’s report outlines strategies for maximizing the effectiveness and efficiency of initiatives to reduce the barriers to voting for homeless electors. For example, one barrier she identified regarding attestations of residence was confusion over where to vote. Because people without a fixed address will not likely receive a voter information card and may not have voted consistently in past elections, they may not be familiar with the requirement to vote only at the poll specific to their declared address, that being the agency providing the attestation of residence. She suggested that returning officers consider locating polling sites in homeless-serving agencies where significant numbers of attestations of residence are anticipated to be used.
 Ms. Lahn is an Elections Canada Returning Officer with some experience in managing elections under the new legislation, and some practical experience in dealing with homeless individuals, specifically in connection with their obtaining of identification documents. If her report were in the form of an affidavit, some portions of it would be admissible, direct evidence about the implementation of the voter identification requirements. There is no evidence from Ms. Lahn under oath. However, Canada had the opportunity to cross-examine her, and had the opportunity to lead evidence to contradict what she wrote in her report. Given the limited purpose for which the plaintiffs seek to rely upon this evidence, I will consider it as bearing on the section 1 proportionality issues and will take Dr. Corbin’s comments into account in assessing its weight.
 Another qualitative report upon which the plaintiffs rely is “Implementation of the Identification Requirements in the Canadian North”, October 7, 2008. This report was written by Ilona Dougherty and Adrienne Smith of Apathy is Boring, a national non-partisan project engaged in encouraging active citizenry, particularly among young Canadians. Upon the request of Elections Canada, Apathy is Boring conducted a series of interviews to seek an understanding of the potential challenges with the new identification requirements unique to northern Canadians. The authors conducted over 70 interviews with a variety of identified stakeholders in the Yukon, Northwest Territories and Nunavut and the electoral district of Desnethé-Missinippi-Churchill, as well as Members of Parliament whose ridings are directly above or below the 60th parallel. The report explains that the interviews were guided by a questionnaire developed in collaboration with Elections Canada and were conducted between I and August 2008.
 Under the heading “Attitudes Towards Voter Identification at the Polls”, the report summarizes its general findings:
Throughout the interviews, participants expressed the consensus that the requirement to produce ID at the polls is not seen as necessary in the Canadian North, especially in small isolated communities. “Everyone knows everyone else” was a phrase repeated in almost every interview conducted. Producing ID is seen as a strange and redundant process, especially when it is likely a family member or friend who will be staffing at the polls. As expressed by participants, this issue is compounded because it is not general practice for northerners, especially First Nations, to carry ID because they don’t often use it in day-to-day life: “People won’t have their ID in their wallet.”
Simply put, participants felt that this legislation “will create difficulty” in the North, and many people will “fall through the cracks”.
There was an overall concern expressed in Nunavut and the NWT that even when locals do have ID there will be many issues with inaccuracy including misspellings, incorrect information, or information that does not match or is inconsistent across multiple pieces of ID. For example, participants referred to the recent admission by the Government of the Northwest Territories that there are more health cards in circulation than there are people living in the NWT. It was pointed out that birth dates of many Elders are unclear, or not known, and many First Nations people have names that have been changed or adapted throughout their lifetime. Debit cards are often stolen in Inuvik and employee cards and firearm acquisition cards (FACs) are often inaccurate. Given this, it was made clear by participants that identification in the North is not a fail-safe way to ensure that the most accurate and necessary information about voters will be obtained.
Participants identified youth, Elders, First Nations and middle-aged women as those who will have particular trouble proving their residential address because their names generally do not appear on many of the documents accepted by Elections Canada for use at the polls. There is also confusion among families around who exactly needs to prove their identity.
 I accept Dr. Corbin’s opinion that this was not a statistical study from which it is possible to draw inferences as to whether the voter identification requirements have actually impeded homeless electors or electors in the Canadian North from voting. The plaintiffs submit that the report nevertheless has evidentiary value as identifying the issues arising in connection with the voter identification requirements as well as some potential solutions. I will consider it, but can give it only limited weight since the survey questions used were not available, and the stated credentials of the authors do not show expertise in conducting this kind of research.
 Having reviewed the evidence bearing on the actual or potential deleterious effects of the legislation, and giving it the weight that is appropriate, I conclude that the deleterious effects of the voter identification requirements to date have been very modest. I further conclude that in all likelihood the deleterious effects will decrease over time as the voter identification requirements become widely understood.
 First, there is no clear evidence that any individual Canadian has actually been impeded from casting a ballot as a sole result of the voter identification requirements. I accept the evidence tendered by Canada that the survey results do not permit such a conclusion, and the plaintiffs did not provide evidence of individuals for whom the identification requirements made it impossible to vote in any of the elections held since the legislation came into effect.
 Second, past deterrence from voting, and deterrence or prevention from voting in the future, would also be deleterious effects of the legislation. There is direct evidence (Ms. Eddelstone’s) and circumstantial evidence (the reports of the CEO, the underlying surveys and qualitative reports, and the evidence of Ms. Graves) supporting the inference that the voter identification requirements have been a factor in making it more difficult for people to cast a ballot in the by-elections and general election held since the legislation became effective.
 It is a fair inference that, among the millions of Canadians who were eligible to vote, for a small number, the voter identification requirements played some role in deterring them from voting – as they did for Ms. Eddelstone. These deterrent effects would likely be greater on the Canadians who are economically disadvantaged, are seniors, live in rural or remote areas, or have disabilities. It is also a fair inference that the requirements may continue to play some role in making it more difficult for some Canadians to vote to the point that, in very rare cases, they might altogether prevent some electors from voting.
 The evidence reveals, however, that Elections Canada engages in significant public education and outreach initiatives to ensure that voters are aware of the identification requirements, thus enabling voters to make appropriate arrangements to have identification available on voting day. Elections Canada has devised, and has the authority to continue to update, an extensive list of documents that can serve as identification. It has in place the letter of attestation system, specifically to meet the needs of voters who are homeless. It is a reasonable inference that as familiarity with the voter identification requirements increases, the small number of voters who are adversely affected by those requirements will decrease even further. Further, the perception that it will be difficult or impossible for persons to vote who do not possess conventional forms of documentary identification should be dispelled as Elections Canada continues to take the kinds of steps it has and as Canadians gain experience with the new system.
 The plaintiffs submit, rightly, that the voter identification requirements may affect both individual and societal interests. This is because a lower voter turn-out harms democracy and faith in democracy. However, the available evidence does not support the conclusion that the voter identification requirements have had, or will have, that consequence. There is no evidence supporting the conclusion that lower voter turn-out in the 40th general election was related to the innovation of the voter identification requirements and there is no evidence connecting the voter identification requirements with declining faith in the electoral system.
 I also note, and agree with, the observation in the Environics report on the 2007 by-elections in Québec that any change that significantly impacts satisfaction with the act of voting runs the risk of negatively impacting elector participation in future elections. However, the evidence does not disclose that the voter identification requirements have had a significant impact on voters’ satisfaction with the voting experience and accordingly there is no indication of a deleterious effect in this regard.
Balancing the Effects
 The final question is whether the deleterious effects of the voter identification requirements are out of proportion to the public good they accomplish. In other words, are their benefits worth the costs they impose?
 In making that assessment one must bear in mind what Chief Justice Dickson stated in Oakes at para. 64. The ultimate standard against which limits on Charter rights are measured is correspondence with the “values and principles essential to a free and democratic society”, including “respect for the inherent dignity of the human person, commitment to social justice and equality, accommodation of a wide variety of beliefs, respect for cultural and group identity, and faith in social and political institutions which enhance the participation of individuals and groups in society.”
 I have found that the salutary effects of the legislation are modest, in that there was little evidence of voter fraud before the requirements were imposed. However, the legislation will increase the assurance that votes in federal elections are cast by persons who are entitled to vote, and are cast in the polling districts where the persons voting are resident. It will reduce voter fraud to an even lower level, prevent it from increasing, and will enhance public confidence in the integrity of the electoral system and in the Canadian democratic process.
 Although I have concluded that s. 148.1(1) of the Act is inconsistent with s. 3 of the Charter, and that the voter identification requirements overall have the effect of interfering with the capacity of some citizens to play a meaningful role in the electoral process, I have found that the deleterious effects of the legislation are very modest, amounting for the most part to minor inconvenience for a minority of electors. For a small number of electors, however, there may be a higher level of inconvenience and, in extremely rare cases, preclusion from voting.
 The plaintiffs submit that precluding even one voter from casting a ballot would be a significant deleterious effect. Indeed, that would be a deleterious effect, as would be the creation of inconvenience that discourages voters from coming to the polls. In an ideal world, no elector would ever be inconvenienced or precluded from voting by any aspect of the electoral system. However, that is not the constitutional requirement. Just as it cannot be constitutionally required for an individual polling station to be set up for each individual voter, a system of voter identification need not be such that not one person is ever inconvenienced or precluded from voting by its requirements. The state has a positive obligation to create an electoral system that is sensitive to the needs of all electors and that maximizes access in every way possible, but a standard of absolutely perfect access cannot be imposed.
 On the premise (which I find is well-founded in the evidence) that the CEO will continue with public outreach initiatives and implementation of the voter identification requirements so as to further minimize difficulties for voters and maximize access, in particular for those voters who are marginalized or impoverished, or who have been identified in the research to date as experiencing the most difficulty with the identification requirements, it is likely that the deleterious effects will dwindle further, over time, toward a vanishing point.
 In my view, the defendant Canada has met the burden at this final stage of the analysis and has established that the salutary effects of the legislation outweigh its deleterious effects, and that the limit on Charter rights imposed by the legislation is consistent with the values and principles of a free and democratic society.
 I have found that the legislation is inconsistent with the electoral rights guaranteed in s. 3 of the Charter. However, I find that it constitutes a reasonable limit, prescribed by law and demonstrably justifiable in a free and democratic society, under s. 1 of the Charter.
“Lynn Smith J.”