COURT FILE NO.: CV-18-590584
DATE: 20210608
ONTARIO
SUPERIOR COURT OF JUSTICE
BETWEEN:
WORKING FAMILIES ONTARIO, PATRICK DILLON, PETER MACDONALD, THE ELEMENTARY TEACHERS’ FEDERATION OF ONTARIO, FELIPE PAREJA, THE ONTARIO ENGLISH CATHOLIC TEACHERS’ ASSOCIATION AND ON BEHALF OF THE MEMBERS OF THE ONTARIO ENGLISH CATHOLIC TEACHERS’ ASSOCIATION, THE ONTARIO SECONDARY SCHOOL TEACHERS’ FEDERATION AND LESLIE WOLFE
Applicants
– and –
THE ATTORNEY GENERAL OF ONTARIO and THE CHIEF ELECTORAL OFFICER OF ONTARIO
Respondents
– and –
THE CANADIAN CIVIL LIBERTIES ASSOCIATION
Intervenor
Paul Cavalluzzo, Adrien Telford, and Tyler Boggs, for the Applicants, Working Families Ontario, Patrick Dylan, and Peter MacDonald
Paul Cavalluzzo, Adrien Telford, and Christopher Perri, for the Applicant, Ontario English Catholic Teachers Association
Susan Ursel, Kristen Allen, and Natasha Abraham, for the Applicants, Ontario Secondary School Teachers Federation and Leslie Wolfe
Howard Goldblatt, Christine Davies, Daniel Sheppard, and Melanie Anderson, for the Applicants, Elementary Teachers Federation of Ontario and Felipe Pareja
Yashoda Ranganathan and David Tortell, for the Respondent, Attorney General of Ontario
Stephen Aylward, for the Respondent, Chief Electoral Officer of Ontario
Colin Feasby, Lindsay Rauccio, Graham Buitenhuis, Stephen Armstrong, for the Intervenor, The Canadian Civil Liberties Association
HEARD: June 2-3, 2021
e.m. Morgan, J.
[1] Since at least the late 1980s, there has been a concern in Canada that third party spending on political advertisements be restricted in the run-up to an election so that private interests do not drown out the political parties waging the electoral contest. But at what point do rules that protect democracy drown out civil society voices protected by the Canadian Charter of Rights and Freedoms (the “Charter”)?
I. The legislative background
[2] In 2017, the Ontario legislature amended the Election Finances Act, RSO 1990, c E7 (“EFA”) to impose, for the first time, spending limits on political advertising by third parties in the six months leading up to the issuance of an election writ. Section 37.10.1(2) of the then-current EFA placed a $600,000 overall spending limit on political advertising by any one individual or organization in the six-month pre-election period, of which a maximum of $24,000 could be spent in any given provincial riding.
[3] Prior to that time, there were no spending limits on political advertising by third parties, although since 2007 third party advertisers have been required to register with the province’s Chief Electoral Officer (“CEO”) and report on how much they spend on political advertising during the election period – i.e. during the period between the issuance of the election writ and polling day: EFA, s. 37.1. In the first generation of this legislation, “political advertising” was defined as advertising in any medium with the purpose of promoting or opposing a registered party, its leader, or the election of a registered candidate.
[4] The 2017 iteration of the EFA not only extended the restricted period to 6 months prior to the dropping of the writ, but added in a significant way to the definition of “political advertising”. In addition to the previous regulation of advertising which directly promotes or opposes a political party, its leader or a candidate, the amended version included so-called issue-based advertising – i.e. advertising that that can be reasonably regarded as closely associated with the election of a party, leader, or candidate. This 6-month pre-election spending regulation, with its inclusion of issue-based advertising, in turn prompted a constitutional challenge by a number of the Applicants alleging violation of Charter sections 2(b) (freedom of expression) and 2(d) (freedom of association).
[5] The challenges to the 2017 amendments were launched in January 2018. They did not impugn the enactment of election spending limits during the election period subsequent to the issuance of a writ. The Applicants accepted that restrictions during that period were justifiable to ensure that properly functioning elections would take place. However, they challenged the regulation of election spending during the preceding six-month period as well as the definition of “political advertising” that accompanied that extension.
[6] The 2018 constitutional challenges were responded to by the government of Ontario with, among other things, a series of expert reports and affidavits explaining that the 6-month pre-election spending limit on third-party political advertising was necessary and reasonable to ensure a fair and proper election process. Those Applications, however, have never been adjudicated. Instead, unannounced and to the Applicants’ surprise, the government amended the EFA again before the cases got to court. The 2021 amendments were contained in the Protecting Ontario Elections Act, 2021 (“Bill 254”), which was enacted in February 2021, received royal assent on April 19, 2021, and came entirely into force on May 4, 2021. The relevant provisions of Bill 254 are now incorporated into the current version of the EFA.
[7] Bill 254 introduced a number of significant amendments to the EFA. Most significantly, the amendments doubled the pre-election restricted spending period from six months to twelve months (while the $600,000 per third party limit remains the same). Accordingly, as matters now stand, third parties are limited to $600,000 in overall spending on “political advertising”, including issue-based advertising, for 12 months preceding the issuance of an election writ. This is by a significant margin longer than any restricted pre-election period in any other jurisdiction in Canada.
[8] The Applicants in the 2018 proceeding issued Amended Notices of Application on April 20, 2021. In an initial case conference, I ordered the several Applications to be consolidated into one proceeding challenging the 2021 version of the EFA. All parties have updated their materials, so that the record now reflects the Applicants’ challenge to, and Ontario’s defense of, the relevant sections of the current EFA which include the Bill 254 amendments.
II. The impugned sections
[9] The sections of the EFA at issue here include certain of the definitions in section 1(1), together with sections 37.0.1, 37.10.1(2), 37.10.1(3)-(3.1) and 37.10.2 (the “Impugned Sections”). These statutory provisions encompass the 12-month pre-election restricted spending period, the inclusion of issue-based advertising, anti third-party coordination provisions that prohibit collusion among political advertising groups, mandatory registration and reporting requirements that accompany expenditures on political advertisements, and the punishments and administrative penalties that have been enacted to enforce all of these policies.
[10] Section 37.10.1(2) of the EFA sets out the pre-election expenditure limit. It is this section that is the primary focus of the Charter analysis pursued here. The other Impugned Sections, including the definition of “political advertising” and the several reporting and enforcement-oriented sections, are in effect ancillary to the enactment of the pre-election spending limits. It stands to reason that if there were no spending limits there would be nothing to define, report, or enforce.
[11] The current $600,000 limit in section 37.10.1(2) is for a full one-year period before the writ drops in a fixed-date provincial election, of which a maximum of $24,000 can be spent in any one electoral district. Section 37.1 defines a “third party political advertising expense” as “an expense incurred in relation to: (a) the production of a third-party political advertisement, or (b) the acquisition of the means of transmission of a third-party political advertisement to the public…”. These include all costs associated with researching (including surveys, polling and focus groups), designing, producing, and transmitting an advertisement to the public. The Applicants contend that this amount is prohibitively low given the realistic costs of modern television and other media advertisements.
[12] Section 1.1(1) of the EFA defines “political advertising” broadly to include advertising that directly or indirectly promotes or opposes a party, leader or candidate. It also covers “… advertising in any broadcast, print, electronic or other medium with the purpose of promoting or opposing any… position on an issue that can reasonably be regarded as closely associated with a registered party or its leader or a registered candidate”.
[13] Supplementing the statutory definition is a non-exhaustive list of factors for the CEO to consider when determining whether an ad constitutes “political advertising”. Counsel for the Attorney General submits that this list, set out in full in section 37.0.1 of the EFA, provides assurance that expression not related to elections is not captured by the statutory definition.
[14] Counsel for the Applicants, on the other hand, submit that the list is vague and unhelpful. The listed factors include consideration of whether the third party has typically advertised on a particular subject matter, at a particular time of year, or in a particular volume, and, in section 37.0.1(h), “whether the advertising is within the normal parameters of promoting a specific program or activity.” These factors are aimed at distinguishing ongoing issues advocacy from election-oriented issues advocacy, which may, of course, frequently contain considerable overlap.
[15] There are also a number of specific limitations and exceptions to the “political advertising” definition, for the most part aimed at exempting unpaid-for communicative activities. For example, the section 1.1(1) definition of excludes editorial and column writing, commentary, newswriting, production and distribution of a book, presentation or communication with shareholders, employees, trade union members, and non-commercial internet transmission of personal political views. It also excludes ‘get out the vote’ messages and telephone campaigns designed to encourage voters to participate in the election.
[16] The Applicants contend that, even with the noteworthy exceptions and the listed factors guiding the interpretation, the definition of “political advertising” is unfairly broad and unworkably vague. They submit that elected politicians, and in particular the Premier, cabinet ministers, and the governing party, are frequently associated with all of the important public policy issues of the day. It is their view that the Impugned Sections therefore prevent individuals and civil society organizations from effectively communicating their views on virtually all major issues – from health care to education to policing to the environment, etc. – for a full 13-months prior to election day.
[17] Section 37.10.1(3) of the EFA prohibits third parties from circumventing a spending limit by coordinating with another third party. Section 37.10.1(3.1) is the most straightforward part of this provision, stating that any contribution from one third party to another third party for the purposes of political advertising shall be deemed as part of the expenses of the contributing third party. The section also goes on to prohibit coordination between advertising third parties by sharing information, a common vendor, or a common set of political contributors or donors with another third party. The Applicants submit that this, like the definition of “political advertising”, is unworkably broad and vague. They point out that “information”, “vendor”, and “political contributors or donors”, together with the concepts of “sharing information” and sharing “a common advocacy, cause or goal” are all undefined in the EFA.
[18] As for the reporting requirements under the EFA, section 37.5 compels third parties to register in the pre-election period upon incurring $500 or more in political advertising expenses. Each advertising third party is required under section 37.5(3) to appoint a chief financial officer and, under section 37.7, an auditor once their advertising expenses reach $5,000. Six months after election day, third parties must file a report with the CEO listing “all third-party political advertising expenses and the time and place of broadcast or publication”, as well as details about the identity of individuals who contributed funds to the third party for political advertising purposes in the period beginning six months before the pre-election period and ending three months after polling day.
[19] In addition, under s. 37.10.2(1) of the EFA, every third party is required to promptly file with the CEO an interim report each time it “has paid or committed to any person or entity to spend any funds on political advertising” in the amount of at least $1,000. The third party must report on “the amount spent or committed, with a separate report being required each time aggregate spending increases by an amount of at least $1,000”. And, finally, the third party must also report when it has reached the overall spending limit for the pre-election period. Section 37.10.2(2) mandates the CEO to publish all of the advertising expenditure reports on its website within two days of receiving them.
[20] As might be expected, the Attorney General’s position is that the reporting requirements are an important element of transparency. It is the government’s view that these sections contribute to the aim of the legislation in creating an equal playing field for all electoral participants and contributors. The Applicants, by contrast, perceive the reporting requirements as unnecessarily onerous. They contend that these features of the EFA are so burdensome and intrusive as to effectively prevent all but the largest and best resourced organizations from engaging in any meaningful political advertising.
[21] Finally, the Applicants submit that the penalty provisions under the current version of the EFA are severe and that the administrative remedies are subject to unchecked discretion. There are substantial fines imposed under sections 46 through 49 if third parties are found to have “knowingly contravened” the spending limits and reporting requirements; specifically, under section 46.0.2, any third party that contravenes the spending limits is, in addition to other penalties, liable to a further fine of up to five times the amount by which they overspent. Moreover, separate and apart from the offence provisions, section 45.1 creates administrative penalties of up to $100,000 for each contravention by a third party. These penalties may be ordered by the CEO if that office believes on reasonable grounds that a person or entity has contravened specified registration, contribution, and reporting provisions.
[22] Counsel for the Attorney General see these remedial provisions as part of ordinary administrative procedure. They point out that the regulatory system in place under the EFA is administered by the CEO, which is an office independent of government exercising non-political duties and judgment. In the Attorney General’s view, the administrative enforcement mechanism is part and parcel of the very rule of law that the Applicants seek to uphold.
[23] Counsel for the Applicants, by contrast, view this discretion in administrative enforcement as a serious infringement of the rule of law. They submit that the severity of the penalties, and especially the way that the administrative penalties can be levied in the context of already vaguely defined regulations, impose a chilling effect on third parties’ political expression. In the Applicants’ view, the remedies sections on their own amount to a violation of Charter rights.
III. Freedom of expression and third-party political advertising
[24] Since the first Charter decade and its early jurisprudence on freedom of expression, the Supreme Court of Canada has adopted a contextual approach to freedom of expression. Chief Justice Dickson, writing for the majority in R v. Keegstra, 1990 CanLII 24 (SCC), [1990] 3 SCR 697, 762, indicated that not all expression was equally deserving of the same level of protection under section 2(b) of the Charter, and that some forms of expression were more “closely linked” with the rationales underlying the Charter right than are others. Justice McLaughlin (as she then was), writing in dissent in Keegstra but not dissenting on this point, followed up on this idea with a specific comment on political expression: “Salient among the justifications for free expression in the first category is the postulate that the freedom is instrumental in promoting the free flow of ideas essential to political democracy and the functioning of democratic institutions. This is sometimes referred to as the political process rationale”: Ibid., at 802 [citations omitted].
[25] Accordingly, the section 2(b) guarantee has often been said to give expression in a political context the highest level of protection. As Justice McLachlin put it in R v Zundel, 1992 CanLII 75 (SCC), [1992] 2 SCR 731, “The purpose of the guarantee is to permit free expression to the end of promoting…political or social participation”.
[26] On the other hand, the early Supreme Court case law also put a qualifier on the Court’s enthusiasm for protecting political expression, indicating that the contextual analysis can cut both ways depending on the nature of the controversy at hand. Accordingly, in Ford v Quebec, 1988 CanLII 19 (SCC), at para 54, the Court went out of its way to state that “political expression is only one form of the great range of expression that is deserving of constitutional protection”. Political speech engaged in by individuals or organizations in civil society can therefore be at either the high end or the moderate end of constitutional protection. The context of the particular mode of expression is an all-important factor.
[27] When it comes to spending and limitations on spending for political advertisements, the Supreme Court has said in each case in which the issue has arisen that while the subject matter of the expression may be quintessentially political, free speech is not the only constitutional value to be protected in an election. Equality of speaking opportunities is also of high concern in the democratic process.
[28] In furtherance of this view, the Court indicated in Libman v Quebec, 1997 CanLII 326 (SCC), [1997] 3 SCR 569, at para 49, that, “[t]he actions of independent individuals and groups can directly or indirectly support one of the parties or candidates, thereby resulting in an imbalance in the financial resources each candidate or political party is permitted.” In other words, the financing of political expression – specifically, election-oriented advertisements – is certainly an aspect of expression deserving protection under section 2(b) of the Charter; but its level of protection is a matter of context, to be weighed with and against other values underlying democracy itself.
[29] Canadian regulation of third-party spending in elections commenced in the wake of the 1991 Royal Commission on Electoral Reform and Party Financing (the “Lortie Report”). The Lortie Report was of the view that unlimited third-party spending would undermine party and candidate spending limits. Specific concerns about large-scale third-party engagement in the 1988 federal election fought over the issue of U.S.-Canada free trade was the engine that for the most part drove the Commission’s work. Following the Lortie Report, federal legislation for the first time established limits on third party advertising spending during the 1992 federal referendum campaign. The legislation required all third parties to report to the Chief Electoral Officer of Canada all advertising spending and the contributions they received toward that spending.
[30] In December 2018, the Canada Elections Act was amended to create a formal “pre-election period”. During the pre-election period, third-party advertising was limited to $700,000, of which no more than $7,000 could be spent to promote or oppose a particular candidate. Ontario followed the same pattern, and its EFA was amended in late 2017 to provide for a $100,000 limit on third-party political advertising spending during the election period and a $600,000 limit for the six months preceding the election period. As already indicated, the Bill 254 amendments introduced in February 2021 extended the six-month pre-election period to twelve months, and left in place the $600,000 overall spending limit for the extended pre-election period.
[31] Under the egalitarian principle espoused by the Lortie Report, freedom of expression is counterbalanced by a need to ensure that all citizens have an equal opportunity to participate in the electoral process. This principle has, in turn, been embraced by the Supreme Court, which in a series of cases has reiterated that “spending limits are necessary to prevent the most affluent from monopolizing election discourse”: Libman, at para 47.
[32] This egalitarian model of democracy views it as a constitutional imperative to ensure that voices backed with large financial resources do not come to “deprive their opponents of a reasonable opportunity to speak and to be heard”: Figueroa v Canada (Attorney General), [2003] 1 SCR 91, at para 52. The Charter has therefore become not just a vehicle for guaranteeing political speech, but a vehicle for “promot[ing] the equal dissemination of points of view by limiting the election advertising of third parties: Harper v Canada (Attorney General), 2004 SCC 33, [2004] 1 SCR 827, at para 63.
[33] With all of that, the Supreme Court has also stayed true to the idea that the freedom to speak one’s mind is not only essential to human fulfillment, but essential to Canada’s way of government. As Justice L’Heureux-Dubé said in Committee for the Commonwealth of Canada v Canada (Attorney General), [1991] 1 SCR 139, “Freedom of expression cannot be jettisoned in any system which values self-government – political participation is valuable in part because it enhances personal growth and self-realization.” To put it another way, “our democratic system is deeply rooted in, and thrives on, free and robust public discussion of public issues. As a general rule, all members of society should be permitted, indeed encouraged, to participate in that discussion”: Fraser v. Public Service Staff Relations Board, 1985 CanLII 14 (SCC), [1985] 2 SCR 455, 467.
[34] As a starting point for the analysis, there is little doubt that third party political advertising spending limits such as those in section 37.10.1(2) do restrict freedom of expression. Counsel for the Attorney General have conceded as much in their factum as well as in oral argument before this Court. In doing so, they acknowledge that the right guaranteed in section 2(b) of the Charter is a very broad one. It is engaged whenever government limits an activity that conveys or attempts to convey meaning: Montréal (City) v 2952-1366 Québec Inc., [2005] 3 SCR 14, at paras 58-59. There are, of course, many such activities; commercial advertising is one and, of course, political advertising is another: Irwin Toy Ltd v Québec (Attorney General), 1989 CanLII 87 (SCC), [1989] 1 SCR 927, 969.
[35] The question remains, however, whether the infringement can be justified under section 1 of the Charter. That is, is there a constitutionally compliant reading of the Impugned Sections of the EFA? The Supreme Court observed in R v Sharpe, 2001 SCC 2, [2001] 1 SCR 45, at para 38, that “[i]f a legislative provision can be read both in a way that is constitutional and in a way that is not, the former reading should be adopted.”
[36] The Attorney General’s position is that the Impugned Sections are necessary for the equality model of elections embraced by the Lortie Report, and, further, that they do not bear the broad meaning attributed to them by the Applicants. Counsel for the Attorney General submit that a common sense reading of the EFA, supported by the statute’s very title, is that it regulates advertising by third parties in relation, and only in relation, to an election.
[37] That interpretation may beg the question of whether any given instance of issues-based advertising is or is not in reference to an upcoming election. However, it does identify that it is the election context, and not a broader reading of “political advertisements”, that is the focus of the section 1 analysis. That is, if justifiable in a free and democratic society, the Impugned Sections must be shown to be a pressing, rational, minimal, and proportionate form of election-oriented regulation.
IV. Reasonable limits on expression
[38] It is well known that the section 1 analysis generally follows the pattern first set out in R v Oakes, 1986 CanLII 46 (SCC), [1986] 1 SCR 103. Before launching into the multi-part Oakes test, however, the Applicants raise the question of whether the restrictions on political advertising are void for vagueness. At its most general level, “[t]he doctrine of vagueness finds its sources in the rule of law”: Committee for the Commonwealth of Canada (L’Heureux-Dubé J., concurring), citing Gary T. Trotter, “LeBeau: Toward A Canadian Vagueness Doctrine” (1988), 62 CR (3d) 183, at 18. An excessively vague law does not provide fair notice to persons of what is prohibited, and does not provide clear standards for those entrusted with enforcement: Peter Hogg, Constitutional Law of Canada (Toronto: Thomson Canada Ltd., 2004), at 1039; R v Nova Scotia Pharmaceutical Society, 1992 CanLII 72 (SCC), [1992] 2 SCR 606, 639.
[39] Both of these criticisms – uncertain compliance and arbitrary enforcement – are levelled by the Applicants at the Impugned Sections of the EFA. Both arguments, however, overstate the concern.
[40] I do agree that it is sometimes hard to distinguish politics from anything else. In fact, it is a cornerstone of contemporary scholarship that even the personal is political: see Carole McCann and Kim Seung-Kyung, Feminist Theory Reader: Local and Global Perspectives (London: Routledge, 2013), at 191; Carol Hanisch, “The Personal is Political”, in: Shulamith Firestone and Anne Koedt, eds, Notes from the Second Year: Women’s Liberation (New York: Radical Feminism, 1970). This all-encompassing understanding of the term makes the question ‘what is political?’ a notoriously difficult one: Eugene F. Miller, “What Does ‘Political’ Mean?” (1980), 42 Rev. of Politics 56-72.
[41] That said, distinguishing the political from the partisan is a less difficult exercise: Canada Without Poverty v. v. AG Canada, 2018 ONSC 4147, at paras 1-8. For the purposes of the Income Tax Act, for example, the courts have had difficulty distinguishing political activities from charitable activities directed toward a wide array of public goods; but they have had little difficulty identifying politically partisan advocacy from other types of politics writ large: see Vancouver Society of Immigrant and Visible Minority Women v. MNR, 1999 CanLII 704 (SCC), [1999] 1 SCR 10, at paras 135, 175-189.
[42] It is generally easy to know who is running for election and whether an advertisement or media campaign is targeting a candidate or party; it is also possible to discern whether an advertisement targets a candidate’s or a party’s election talking points or policies. The regulatory decision under the EFA is made in first instance by the CEO, a non-partisan, independent office, subject to the safeguards of administrative law. The CEO’s particular attunement to the electoral issues of the day makes it a reliable decision-maker.
[43] As counsel for the Attorney General argue here, the EFA itself gives ample guidance as to what types of advertisements fall into the restricted category. One would expect common sense to prevail in interpreting and applying that guidance. This is not legislation where, as with parallel provisions in the British Columbia Elections Act, there is “a seemingly limitless range of activities” to which it applies, and virtually “any public communication on government action” would be subject to pre-election regulation: Reference re Election Act, 2012 BCCA 394, at paras 3, 11, 20, 25. Rather, the guidance and context provided by the EFA effectively limit the application of the term “political advertisement” to election-oriented communications.
[44] The Supreme Court has stated that a law need not define its terms with unmistakable precision; rather, “a law will be found unconstitutionally vague if it so lacks in precision as not to give sufficient guidance for legal debate”: R v Nova Scotia Pharmaceutical Society, 1992 CanLII 72 (SCC), [1992] 2 SCR 606. While the term “political advertisement” as used in the EFA may open for a certain amount of debate, the debate has a defined context with discernable parameters that make it no different than most debates over statutory interpretation.
[45] The very fact that what is and what is not a “political advertisement” can be coherently debated in terms of the EFA’s own guidelines saves it from the kind of stark indeterminacy that the rule of law eschews. It is not mathematically precise, but it is also not so vague as to make either compliance or regulatory decision-making incoherent or unknowable.
a) Pressing and substantial objective
[46] Turning to the Oakes test, the first question is whether the EFA’s restrictions on expression embody a pressing and substantial objective. McLachlin CJ opined in BC Freedom of Information and Privacy Association v Attorney General of British Columbia, 2017 SCC 6, [2017] 1 SCR 93, at para 16, that for limitations on political speech to be justified under section 1, they “must be supported by a clear and convincing demonstration that they are necessary, do not go too far, and enhance more than harm the democratic process.”
[47] In introducing the second reading of Bill 254 in the Legislature, Attorney General Doug Downey set out the objectives behind the amended EFA:
This proposed time-limit increase would help to responsibly regulate third party advertising between elections, and would protect the essential voice of individuals and ensure they, and not pop-up political groups, remain the driving force of our elections. We want individuals to make decisions based on what each party stands for and based on their record. The proposed change strikes an important balance. It maintains the ability of third parties to participate in elections, and it builds on previous changes made by the Ontario Legislature to ensure that voters, not third-party political action pop-up groups, have the loudest voice in our elections.
Legislative Assembly of Ontario, Hansard, March 29, 2021, M-402–M-403.
[48] In explaining these objectives, Attorney General Downey in effect reiterated those embraced by the Lortie Report and the Supreme Court of Canada. Stating the matter a tad more eloquently, Chief Justice McLachlin has identified the pressing and substantial objectives of political advertising financial restrictions as being threefold: “first, to favour equality, by preventing those with greater means from dominating electoral debate; second, to foster informed citizenship, by ensuring that some positions are not drowned out by others (this is related to the right to participate in the political process by casting an informed vote); third, to enhance public confidence by ensuring equality, a better informed citizenship and fostering the appearance and reality of fairness in the democratic process”: Harper, at para 23.
[49] One can hear in the government’s introduction of the legislation that buried in the policy objectives is a warning about private parties coming to dominate the electoral scene. That is not surprising. As counsel for the Attorney General describes it, for Canada the spectre of the American political and constitutional position serves as a cautionary tale on the impact that weak third-party political advertising restrictions can have on democratic elections.
[50] Pursuant to the ruling in Citizens United v. Federal Election Commission, 558 U.S. 310 (2010), government in the United States is prohibited from restricting expenditures for political campaigns by corporations, wealthy individuals, and committees established for the purpose of fundraising (PACs). Evidence in the record before me demonstrates that the 2016 U.S. presidential election displayed the full impact of this laissez-faire approach to third-party spending. Independent groups spent some $968 million primarily on advertising, while the top 50 donors in the United States contributed roughly $571.1 million to so-called Super PACs.
[51] Political science experts have opined that the rise in spending by a small number of wealthy third parties has had a corrosive impact on the U.S. electoral process. Two-thirds of Americans surveyed believe that they have no chance to influence electoral outcomes. And the perception that donors can “buy” politicians has fostered a lack of trust in government even beyond the electoral process. In addition, unlimited third-party expenditures have produced political advertising that is often more polarizing than the electoral advertisements from the actual candidates.
[52] There is evidence in the record to suggest that the situation in Ontario is headed in a similar direction. The province’s 2014-2015 Annual Report of Elections Ontario recorded that in 2014 a relatively small number of groups spent significant amounts of the aggregate spending among third parties. According to the government’s statement in the Legislature, third-party political advertising limits are needed to ensure the public’s faith in democracy, to provide candidates with an equal opportunity to communicate their political agendas, and to maintain the egalitarian model identified by the Lortie Report – i.e. a sense of parity between participants in the electoral process.
[53] Counsel for the Attorney General argues that the need for spending limits becomes all the more imperative with fixed date elections. They submit that when the election date is known in advance, politically interested groups and individuals can spend significant amounts of money on election advertising in the pre-election period, effectively blurring the distinction between election and non-election advocacy.
[54] The evidence before me, including from the Applicants’ own elections expert, Professor Andrea Lawlor, is that with Ontario adopting fixed-date elections, limits on third-party political advertising spending for some period of time prior to the election period are necessary to preserve the equality principle on which the electoral process is based. Those limits ensure that third parties cannot spend massively just prior to the election period, but then undermine the statutory limits by ceasing to spend during the election period following issuance of the writ.
[55] All of which points to a pressing and substantial objective for the spending limits and definitions in the Impugned Sections. It also speaks to the need for reporting mechanisms and remedial provisions that give the spending restrictions their teeth. While the Applicants portray the provincial government as acting only out of partisan self-interest, seeking to preserve their majority government above all else, I do not see the enactment of the relevant EFA policies entirely in that way.
[56] As discussed below, skepticism of government’s motives is not misplaced where new election procedures are concerned. However, at the highest level the objective of the spending restrictions in the EFA can be seen as genuine. I am willing to agree with counsel for the Attorney General that they are aimed at fortifying not so much the government of the day, but democratic governance itself.
b) Rational connection
[57] It is the Applicants’ view that the Bill 254 amendments are not rationally connected to the objective of fostering a fair and more egalitarian election process. In their submission, the Impugned Sections undermine the egalitarian model of election finance by failing to equalize the Applicant organizations with the government whose policies they wish to counter. In this view, the problem with the restrictions on political advertisements, as enacted, is that they give an advantage to the government. That is, they limit civil society organizations in their participation in electoral discourse, but do not in the precisely same way limit political advertisements and participation in electoral discourse by government.
[58] With all due respect, this seems to present a slightly skewed view of elections. The equality model embraced by the Supreme Court seeks to prevent wealth from dominating electoral policy debate; it does not seek to treat organizations who advertise their political views as if they are parties or candidates actually running for office. Electoral contests are not between the government and the organizations represented by the Applicants. Rather, they are between incumbent government parties and opposition parties, the latter of whom enjoy the same freedom to engage in political advertising as the governing party.
[59] The Applicants’ view also takes little or no account of the danger that the Lortie Report and every Supreme Court of Canada case on election financing has identified – i.e. that private wealth, whether by an individual or by many individuals pooling their resources – may capture an election in an unfair and, ultimately, undemocratic way. The Applicants also appear to perceive themselves – all of whom are labour unions and their representatives – as the sole targets of the legislation. As counsel for the Attorney General point out, however, Bill 254 and the EFA more generally will impact on all politically engaged organizations regardless of their political leanings.
[60] It is noteworthy that the Lortie Report identified corporate interests, not labour interests, as the subject of concern at that time. While that conclusion can be attributed to the late 1980s federal election issues that spawned the inquiry, it is valuable as a matter of perspective. What seems obviously true is that as political times change, so will the third-party political advertisers.
[61] The Supreme Court stated in Harper, at para 29, that “reason or logic” alone can establish the causal link between limits on citizen spending and electoral fairness. Thus, for example, in Libman, the Court concluded that, as the Attorney General argues here, election spending limits are rationally connected to the objective of fair elections. Indeed, the Court observed that “the conclusion that limits may in theory further electoral fairness is difficult to gainsay”: Ibid.
[62] I have little trouble coming to the same conclusion on rational connection that the Court did in Libman and Harper, whether it applies to the election period or to the pre-election period. The spending limit contained in section 37.10.1(2), along with the supporting and ancillary matters contained in the other Impugned Sections, is rationally connected to the objective of fostering electoral fairness.
c) Minimal impairment
[63] It is with the minimal impairment portion of the Oakes test that the rubber of Bill 254 hits the slippery road of justification, causing the EFA vehicle to skid off course.
[64] The government’s own evidence demonstrates that less impairing measures were available to it than those contained in the Impugned Sections. For example, the CEO, an obvious source of insight into electoral fairness, expressly recommended in June 2016 against regulating issue-based advertising prior to the writ period. The very office charged with ensuring fair elections in the province came to the conclusion that a prohibition on issue-based advertising in any pre-election period does not augment the fairness and equality that such regulations are meant to address.
[65] Even more graphically, the government’s own expert witnesses in the present case have all testified that a 6-month pre-election period was the appropriate and effective period in which spending restrictions for political advertisements should operate. In the predecessor litigation, the experts produced by counsel for the Attorney General – Professor Harold Jansen and former CEO Jean-Pierre Kingsley – both opined that a 6-month period of pre-writ regulation was reasonable. Those same experts have now opined that the new 12-month period introduced by Bill 254 is “also reasonable”.
[66] Without meaning to stress the obvious, it is hard to see how 12 months is minimal if 6 months will do the trick. In certain forms of minimalist design, less may be more: see Ruth Malan and Dana Bredemeyer, “Less is More with Minimalist Architecture”, Perspectives, Sept-Oct 2002, at 48. But this does not apply to the Oakes analysis. This analysis asks “whether the means adopted to achieve the end sought do so by impairing as little as possible the right or freedom in question”: R v Big M Drug Mart Ltd., 1985 CanLII 69 (SCC), [1985] 1 SCR 295, at para 139; Oakes, at para 70. The more the restriction on their political expression, the more the Applicants’ rights are limited. When compared with a twelve-month impairment, a six-month impairment will always be the more minimal one.
[67] Counsel for the Attorney General responds that when it comes to a policy choice like regulating political advertising, the Court should defer to the government’s chosen policy. This reflects the Supreme Court’s view that courts “should not lightly use the Charter to second-guess legislative judgment”: McKinney v University of Guelph, 1990 CanLII 60 (SCC), [1990] 3 SCR 229, at para 318. The Attorney General therefore submits that the relevant Charter jurisprudence “establishes that courts ought to take a natural attitude of deference toward Parliament when dealing with election laws”: R v Bryan, 2007 SCC 12, [2007] 1 SCR 527, at para 9.
[68] I am not convinced that this attitude of deference is applicable here. The policy choice approach comes from cases in which two different values are in competition. That is not exactly the way the policy choice has arisen here.
[69] By way of illustration, in the early Charter case of R v Jones, 1986 CanLII 32 (SCC), [1986] 2 SCR 284, the question was whether a prohibition on home schooling amounted to a reasonable limit on freedom of religion for parents whose beliefs were contrary to what is taught in schools. The Supreme Court indicated that the legislature had faced a choice between protecting religion and protecting universal public education, and that the judiciary had no greater insight into that choice than did the elected branches of government. A similar choice of policy options confronted the legislature in R. v. Edwards Books and Art Ltd., [1986] 2 S.C.R. 71, where the statute elevated a day of rest for labour over the protection of religion sought by non-Sunday worshipers.
[70] Justice LaForest explained these cases in RJR Macdonald v Canada (Attorney General), 1995 CanLII 64 (SCC), [1995] 3 SCR 199, at para 62, as arising where section 1 of the Charter demanded an analysis of legislation that “strikes a delicate balance between individual rights and community needs.” But, as Dickson CJ stated in Edwards Books, at 768, “the nature of the proportionality test would vary depending on the circumstances”. Given the way that Bill 254 amended the 2017 amendments to the EFA before they could be considered in the Applicants’ legal challenge, the choice here is that posed by the new spending restrictions as opposed to the previous ones.
[71] Given that legislative context, there are not really competing socio-economic interests in need of a delicate balance here. That is, the policy choice embodied by section 37.10.1(2) of the EFA is a choice between a 6-month restricted period and a 12-month restricted period. Both are said by the government’s own experts to reflect the same set of values; and both are likewise said by the government’s experts to be effective and to work.
[72] Where the two policies being considered are not competing social values but rather competing ways of accomplishing the same social value – i.e. fair and egalitarian elections – no nuanced balancing is engaged. To defer to the legislature in such a case would not be to assess reasonable limits on a Charter right, but to impermissibly elevate legislative process over Charter conformity: Lorraine Weinrib, “Canada’s Charter of Rights: Paradigm Lost?” (2002) 6 Rev Const Studies 120, 167. Without a contest between competing social or economic values, the constitution cannot support deferring to the legislature for its own sake.
[73] There is no justification or explanation anywhere in the Attorney General’s record as to why the doubling of the pre-election regulated period was implemented. This lack of explanation has to be taken seriously. As counsel for the Canadian Civil Liberties Association points out, the subject of electoral design is one in which the incumbent government has a structural conflict of interest in that its interest in self-preservation may dominate its policy formulation.
[74] This potential for partisan self-dealing poses a fundamental challenge to the democratic system, and represents a context in which a more rights-oriented logic is called for to safeguard democratic institutions: Yasmin Dawood, “Electoral Fairness and the Law of Democracy: A Structural Rights Approach to Judicial Review” (2012) 62 UTLJ 499, 503, 547. It is in this context that McLachlin CJ’s admonition in Sauvé v Canada (Chief Electoral Officer), 2002 SCC 68, [2002] 3 SCR 519, at para 15, is apt: it is “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.”
[75] In the case at bar, “the minimum impairment test requires only that the government choose the least drastic means of achieving its objective: Alberta v Hutterian Brethren of Wilson Colony, 2009 SCC 37, [2009] 2 SCR 567, at para 54. It is self-evident that if a six-month impairment on free expression accomplishes the desired objective, a twelve-month impairment cannot be the least drastic means. It does not matter that both a 12-month and a 6-month restricted period are “reasonable” in the view of the experts. A 12-month impairment is twice as long, and twice as restrictive, as a 6-month impairment, and so by definition is not minimal. There is nothing else to factor into the analysis.
[76] The Bill 254 amendments, all of which are built on implementing and enforcing the 12-month pre-election regulated period in now found in section 37.10.1(2), do not pass the minimal impairment test. They are therefore not reasonable limits on the Applicants’ rights of free expression. They violate section 2(b) of the Charter and are not saved under section 1.
d) Proportionality
[77] The focus of the final step of the analysis under section 1 of the Charter is whether the deleterious effects are disproportionate to the salutary effects of the infringing measures: Dagenais v Canadian Broadcasting Corporation, 1994 CanLII 39 (SCC), [1994] 3 SCR 835, at 889. In the present case, the contest is whether the salutary impact on those “who wish to engage in the electoral discourse” outweighs the deleterious impact on the third parties whose freedom of expression is curtailed: Harper, at para 62.
[78] I agree with counsel for the Attorney General that the deleterious effects of the Impugned Sections are minimized by using a narrow definition of political advertising. As already discussed above in these reasons, properly interpreted, the definition of “political advertisements”, including issues-oriented advertisements, includes only election advertising expressly or impliedly directed to political parties or their leaders, candidates or platforms.
[79] The Impugned Sections are further limited in their deleterious effect by the exemptions built into the definition of “political advertising” at section 1(1) of the EFA. In addition, the burden of the reporting and disclosure sections is offset by the benefit of transparency that those sections embody. Finally, the spectre that the administrative enforcement will be abused by the regulator is offset by the fact that the CEO is an independent, non-partisan and non-political official subject to the safeguards of administrative law.
[80] All of these features of the Impugned Sections, however, are policies enacted in aid of the primary regulatory initiative of Bill 254 – that is, the section 37.10.1(2) creation of a pre-election regulated period of 12 months. That section fails the minimal impairment test and it fails the proportionality tests for much the same reason. It is disproportionate in terms of the deleterious effects that it imposes, as the expert evidence shows that it is twice as long a period, and thus twice as deleterious, as it need be to accomplish its salutary effects.
[81] Moreover, since the definition and examples of “political advertisement”, along with all of the ancillary reporting and enforcement provisions contained in the Impugned Sections, are only enacted insofar as they support implementation of the 12-month regulated expenditure period, it makes no sense for them to survive in the EFA on their own. If there is no restriction on political advertisements there is no reason to define the term, provide examples in aid of its interpretation, require reporting of expenditures, or enforce anything in relation to such expenditures.
[82] In other words, once the primary policy in section 37.10.1(2) falls, the subordinate policies also must fall. None of them are proportionate to anything if the 12-month pre-election period cannot be sustained.
V. Freedom of association
[83] Separate from the challenge under section 2(b), the Applicants allege that the provisions of the EFA that prevent third parties from circumventing spending limits infringe section 2(d) of the Charter (freedom of association). It is not strictly necessary to discuss this aspect of the challenge since the Impugned Sections cannot survive the challenge under section 2(b). However, a few brief comments may aid an analysis of such legislation down the road.
[84] As the Attorney General’s counsel points out, a similar argument was made in Harper in relation to the anti-circumvention provisions under the Canada Elections Act. The Supreme Court held that section 2(d) was not infringed because the challenged provisions did not prevent individuals from joining to form an association in pursuit of a collective goal, which is the essence of the guarantee of freedom of association. Rather, the anti-circumvention provisions only precluded persons from undertaking an activity – namely, circumventing the third-party election advertising limits under the EFA – which is not protected under the Charter: Harper, at paras 125-127.
[85] By logical extension, the same is true of the Impugned Sections. Section 37.10.1 of the EFA states that, “No third party shall circumvent, or attempt to circumvent, a limit set out in this section in any manner”, and then goes on to list a number of ways in which a third party might attempt to do so. These circumvention methods include third party advertisers sharing a common vendor or pooling donations. The terms of the provision are worded in a way that ensures that it is the activity of circumventing the spending limits, and not the associational aspect of that activity, that is prohibited.
[86] The Supreme Court of Canada has indicated that section 2(d), viewed purposively, protects “(1) the right to join with others and form associations; (2) the right to join with others in the pursuit of other constitutional rights; and (3) the right to join with others to meet on more equal terms the power and strength of other groups or entities”: Mounted Police Association of Ontario v. Canada (Attorney General), 2015 SCC 1, [2015] 1 SCR 3, at para 66. It is not the joining that is addressed by the Impugned Sections; it is the spending on political advertisements that is the target.
[87] As in Harper, the anti-circumvention sections are enacted to ensure that third parties cannot do together what they cannot do on their own. The restriction on expenditures, and not the fact that those expenditures may be done either in concert or individually, is what the legislation is about. The Applicants are permitted to coordinate and share as much information as they wish, and are only prohibited from doing so for what other provisions in the EFA identify as impermissible ends.
[88] To be clear, nothing in the EFA prohibits individuals or organizations from joining together to collectively express their views on election issues. If the Impugned Sections could have survived the challenge under section 2(b) of the Charter, they would have survived the Applicants’ challenge under section 2(d) as well. However, since they are rendered inoperative by virtue of their violation of freedom of expression, there is nothing to circumvent and therefore nothing to preserve in section 37.10.1.
[89] The Impugned Sections must, in my view, all rise or all fall together.
VI. Disposition
[90] There shall be a declaration that sections 37.0.1, 37.10.1(2), 37.10.1(3)-(3.1), and 37.10.2 of the EFA infringe section 2(b) of the Charter and are not justified under section 1. Pursuant to section 52(1) of the Constitution Act, 1982, these Impugned Sections are of no force or effect.
[91] I have discussed with counsel the prospect of suspending the operation of my judgment for some period of time to allow the government an opportunity to introduce new amendments to the EFA that might comply with the Charter. Given the timing parameters involved, however, I am not inclined to order such a suspension.
[92] The next election in Ontario is scheduled to take place on June 2, 2022. That means that we are already within the 12-month restricted period for political advertisements under section 37.10.1(2) of the EFA. It would be unfair to the Applicants (and to other potential third-party political advertisers) for statutory provisions that have been declared unconstitutional to remain in operation during this time. That would effectively prohibit expression that I have already determined may not be prohibited.
[93] If the government wishes to amend the EFA yet again in time for the next election, it will have to navigate the legislative process in a way which allows it to do so. The burden of this timing cannot be made to fall on the Applicants whose rights have been infringed.
Morgan J.
Released: June 8, 2021
COURT FILE NO.: CV-18-590584
DATE: 20210608
ONTARIO
SUPERIOR COURT OF JUSTICE
BETWEEN:
WORKING FAMILIES ONTARIO, PATRIC DILLON, PETER MACDONALD, THE ELEMENTARY TEACHERS’ FEDERATION OF ONTARIO, FELIPE PAREJA, THE ONTARIO ENGLISH CATHOLIC TEACHERS’ ASSOCIATION AND ON BEHALF OF THE MEMBERS OF THE ONTARIO ENGLISH CATHOLIC TEACHERS’ ASSOCIATION, THE ONTARIO SECONDARY SCHOOL TEACHERS’ FEDERATION AND LESLIE WOLFE
Applicants
– and –
THE ATTORNEY GENERAL OF ONTARIO and THE CHIEF ELECTORAL OFFICER OF ONTARIO
Respondents
– and –
THE CANADIAN CIVIL LIBERTIES ASSOCIATION
Intervenor
REASONS FOR JUDGMENT
E.M. Morgan, J.
Released: June 8, 2021