Corps de l’article

Introduction

In the Carter decision, the Supreme Court of Canada upheld the constitutionality of variances from voter equality in Saskatchewan’s provincial redistricting plan.[1] Under the electoral map, rural, northern voters would be overrepresented and urban voters correspondingly under-represented. As the governing party benefitted disproportionately from support among the voters who were to be overrepresented,[2] there were legitimate reasons to consider whether the underlying issue in the case was an attempt to impose a partisan electoral map.[3] The Court, however, quickly dismissed the argument that partisanship distorted the electoral map.[4] The decision not to consider more thoroughly the possibility of self-dealing by elected representatives hints that the Court was grappling with how to understand the use of state power to distort the democratic process to further a private purpose.

Literature on the law of democracy in Canada is growing,[5] but it remains underdeveloped in comparison to other areas of constitutional law.[6] In this article, I seek to develop a theoretical framework through which to view the law of democracy in Canada, not confined to the redistricting issues raised in Carter, but rather encompassing the law of democracy more generally. Such a framework has been missing from the Court’s reasoning on the right to vote in section 3 of the Charter[7] and has been largely absent from the literature.[8] I will argue that the Canadian law of democracy should be interpreted in light of structural concerns, particularly the problem of self-dealing by elected representatives and, flowing from this analysis, that courts should scrutinize more closely infringements of democratic rights.

I advance this argument by focusing on the potential breakdowns in the democratic process that occur when legislatures and executives engage in self-serving behaviour in election law. This type of behaviour was arguably the case in Carter, but is not limited to redistricting, so I examine various areas of the law of democracy. I use the term “breakdown” to refer to the manipulation by self-interested insiders of the laws, regulations, and institutions of the democratic process.[9] At their core, concerns about the democratic process are based on concerns about democratic legitimacy. When the democratic process breaks down, then the democratic legitimacy of the institutions and laws it produces suffers in consequence.

Acknowledging the potential for self-interested manipulation of election laws by elected representatives is necessary to understand the law of democracy. Theories of judicial review that fail to take into account the incentives of legislators operating under the constraints of election laws miss something fundamental about how representative democracy actually functions.[10] They ignore the likelihood of distortion of election laws by self-interested parties seeking to remain in office. The potential for distortion unites the myriad subject matter (redistricting, campaign finance, political speech, regulation of political parties) that forms the law of democracy.[11] Existing Canadian scholarship has rightly identified the self-interested manipulation of election laws as a problem to be addressed,[12] but much work remains in understanding the consequences of this insight.

This article will proceed as follows. Section I details the structural approach to the law of democracy, which is the leading model in the US literature, as well as the major alternative, which is known as rights theory. Section II investigates the hints of structural theory that have appeared in the Canadian jurisprudence and argues that the absence of full-fledged structural reasoning by the courts has been problematic. In Section III, I engage with the major critiques of structural theory. I argue that the structural approach to the law of democracy in Canada is preferable to the alternatives because it provides a deeper descriptive understanding of the democratic process and a convincing normative framework through which to evaluate democratic institutions. Section IV outlines three types of breakdowns in the democratic process. I argue that a structural theory of the law of democracy should recognize the problematic nature of three types of breakdowns: partisan, incumbent, and interest entrenchment breakdowns. I cite various examples from the Canadian and comparative law of democracy to explain the preliminary typology that the article sets out.

I. Theories of the Law of Democracy

A. Introduction

At this early juncture in the study of the law of democracy in Canada, the academic literature and judicial accounts remain undertheorized. The Supreme Court first introduced the concept of “effective representation” in Carter. Since Carter, the Court has been applying the doctrine on a case-by-case basis, often with incongruous results. Like the US Supreme Court, the Supreme Court of Canada has no theory of democracy to inform the content that it must pour into the effective representation doctrine.[13] The Court has also interpreted section 3 of the Charter to include the right to “meaningfully participate” in an election, though there is potentially some conflict between effective representation and meaningful participation.[14] The meaning of that phrase shifts with each new area of election law attended to by the Court. This confusion stems to a great extent from the lack of any overarching, consistent theoretical framework to link the various areas of the law of democracy together. In the US literature, two main theories have developed seeking to guide thinking in this area: structural theory and rights theory. I examine each in turn, though focusing at this juncture more extensively on structural theory, which is the leading approach. I examine rights theory in greater detail in Section III. B.

B. The Structural Approach

Structural theory of the law of democracy derives from the “political competition” approach advanced by Samuel Issacharoff and Richard Pildes.[15] The political competition approach defines the primary role of courts in a democracy as ensuring that political insiders do not use their existing political authority to “chok[e] off the channels of political change to ensure that they will stay in and the outs will stay out.”[16] Courts must function as antitrust regulators do in the economic market to ensure that the political market remains competitive. Incumbents must be prevented from self-interestedly frustrating the proper formation of democratic majorities or restricting the political power of minorities. The normative thrust of this approach is to ensure that incumbents are not able to insulate themselves from political and legal accountability. The political competition approach deviates from the traditional posture of constitutional law by emphasizing the systemic effects on voters and democratic institutions of legal rules rather than the constitutional harm those rules may perpetrate against individual rights.

New versions of the structural approach have critiqued the use of the antitrust model[17] or looked to anti-domination as a concept to guide thinking on the law of democracy.[18] Another has critiqued the earlier versions for not recognizing the need for dual-track consideration of individual rights with structural themes.[19] These new versions, however, all recognize the need for a general structural framework and are justly classified as variants of structural theory.

As I understand it, there are four main features of the competition approach: (1) a view of the state as perpetually subject to capture by self-interested incumbents; (2) a claim that a singular focus by courts on the constitutional harm to the individual claimant is insufficient for election law because the real concerns to be addressed are systemic harms; (3) a normative valuation of electoral competition as the best guarantor against self-interested lock-ups of the political process by incumbents; and (4) an adoption of a minimalist version of democracy as the baseline level for interventions to break up lock-ups of the process. I deal with each aspect of the theory in turn.

1. Self-Interest and the State

First, Issacharoff and Pildes reject a view of the state and the incumbent representatives who form the government as altruistic trustees of the public will deliberating and acting upon the common good.[20] Rather, they identify state actors as individuals capable of acting in their own interests, rather than in the interests of those they are supposed to represent. Though Pildes in particular has criticized public choice approaches to democracy,[21] the link between public choice and the competition approach to the law of democracy, at least in how the state is conceived, seems clear. Issacharoff and Pildes depart from public choice by adopting a weaker version of the rational choice assumptions about the state.[22] Public choice theory asserts that all aspects of the state are likely to be captured or plagued by self-interested manipulation. Issacharoff and Pildes make the narrower claim that it is on the law of democracy that political actors are particularly self-interested.[23] The potential for self-interested political actors to use the legislative and regulatory tools of the state for their own ends is particularly endemic on their view on election law. Politicians are able to design the rules of the very game that they must compete in to attain and remain in office. Seemingly small manipulations of the democratic process, such as access to primary ballots, can lead to far-reaching electoral consequences.[24]

Whether one is willing to adopt a public choice approach to election law or not, the narrower claim that self-interested political actors can manipulate democratic procedures to their own benefit is plausible and convincing. Political science scholarship has established that the leading determinant of electoral system change, for example, is whether the governing party will gain seats in the next election.[25] These results suggest that the anticipated partisan consequences of election law reforms guide government decision making. Examples abound in both Canada and the United States of incumbents manipulating election laws for their own benefits. The so-called “White Primary Cases” are a notable example in the United States.[26] In Canada, changes to campaign finance regulation by Prime Minister Harper’s government have had a marked partisan impact and were likely the product of partisan motivation.[27]

2. Individual Rights and Systemic Harm

Second, competition theory claims that an individual rights perspective alone is insufficient to grapple with the full reality of what is at stake in the law of democracy. This is so in their argument not because election laws necessarily engage group rights, but because the underlying concerns are generally structural ones about the operation of democracy.[28] Take vote dilution, for instance. In the seminal American case of Baker v. Carr,[29] the US Supreme Court found the overrepresentation of rural voters due to a failure to reapportion Tennessee’s legislative districts to be justiciable. Of concern in the case, however, was not only the violation of Mr. Baker’s right to an equal vote, but also the systemic consequence of permitting the legislature to over-represent the interests of one aggregation of voters.[30] Similarly, in campaign finance cases, at stake is not simply the application of the spending or funding rule in the given case but also the consequences that the preservation or elimination of the rule will have downstream. Eliminating the ban on corporate spending during election campaigns, as the US Supreme Court recently did in Citizens United v. Federal Election Commission,[31] for example, has the potential to augment the leverage of lobbyists, who can threaten corporate spending against an incumbent. On a structural view, what was truly at stake in Citizens United went far beyond the specific issue of corporate spending.

3. Electoral Competition

Third, structural theory in the United States views electoral competition between parties as the most effective means by which to achieve electoral accountability. Incumbents attempt to insulate themselves from the disapproval of voters during election time by designing rules to favour themselves and to disadvantage their electoral competitors. They alter electoral maps under the practice of partisan gerrymandering, for example, to enhance the possibility that their party will win a majority of districts. The goal is to enhance their electoral prospects; the route that is taken is to manipulate the rules of the game. Structural theorists argue that these types of manipulations reduce electoral competition and thereby hinder electoral accountability. Accountability is preserved when political actors are prevented from reducing competition.

4. Minimalist Assumptions Regarding Democracy

Fourth, Issacharoff and Pildes’s approach adopts a minimalist view of democracy as its starting point. In their work, they explicitly cite Schumpeter as an influence.[32] Schumpeter argued that the basic content of democracy was the alternation of electoral power among elites. Free and fair elections were thus the minimum content of a democracy. The emphasis on competition can be seen as a successor to Schumpeter’s view.[33] Without adequate electoral competition, individuals are deprived of the right to choose among alternative representatives, rendering elections meaningless.

C. Rights Theory

A rights approach takes the more traditional view that courts must apply constitutional law to redress individual harm to a particular voter.[34] It is more skeptical of judicial intervention in the law of democracy and puts greater emphasis on constitutional text and existing doctrine. Rights theorists have criticized the political competition approach for having too much faith in the ability of judicial review to solve political problems, especially after Bush v. Gore.[35] They have also questioned what democratic theory lies behind the political competition approach[36] or argued that competition theory is insufficiently attentive to myriad democratic values because of a singular focus on competition. While rights theorists are not united around any single logic as an alternative to political competition, they are joined together in their critique of the competition approach.

The leading rights theorist is Richard Hasen. Hasen argues that election law, a term he prefers to the law of democracy, should be approached with substantive and not procedural theory. On his argument, equality rights form the basis of a normative evaluation of election laws. He argues that election laws should be understood as engaging either “core” equality rights or “contested” equality rights.[37] On his view, core equality rights are those “few basic rights essential to a contemporary democracy,”[38] which are supported by a social consensus in their favour. Contested rights are categorized as rights around which no social consensus exists. He argues that courts should uphold only core equality rights and leave contested rights to be determined by legislatures and executives. Judicial review becomes problematic in his theory when it strays beyond policing the “minimal requirement[s] for democratic government.”[39] Review of contested rights places courts outside of their legitimate role, as their analysis becomes dependent upon shifting public views of what democracy requires.

Hasen acknowledges that mitigating the self-interested actions of political actors must be a goal of election law, which is a core concern of structural theory.[40] Problems of self-interest should be addressed not by focusing exclusively on political competition or structural concerns, but through what he calls the “collective action principle”.[41] This principle embodies the idea that governments should not be able to prevent citizens from deliberately exercising their right to act collectively within a democracy. At its root, Hasen’s theory espouses judicial restraint. His theory does not trust courts to achieve anything more than adequate policing at the extremes of established equality jurisprudence. Courts should wait for social consensus to develop on contested equality rights issues before plunging into the “political thicket”.[42]

II. Structural Themes in the Canadian Law of Democracy

A. Jurisprudence on Section 3 of the Charter

In this section, I consider the instances in which the Supreme Court of Canada has raised the problem of self-interested behaviour in its law of democracy jurisprudence. I also consider the main scholarly literature on the Canadian law of democracy that engages with structural theory. I argue that the same basic concerns that animated the emergence of structural theory in the United States exist in Canada, though like American courts, Canadian courts have not fully grappled with the problem of self-interested manipulation of the democratic process.[43] Being less than fully engaged with this problem means that the courts are unlikely to be overseeing the democratic process to an extent sufficient to ensure its proper functioning. It also implies that the doctrinal framework used by the courts on section 3 of the Charter will not capture the political motivations and consequences underlying election laws.

The clearest example of structural concerns in the Canadian jurisprudence comes from the dissent in Harper v. Canada (Attorney General).[44] At issue in Harper were several provisions of the Canada Elections Act[45] that together limited spending[46] by third parties (meaning entities other than political parties), imposed disclosure and registration requirements on third parties, and banned outright third party advertising on election day. The Alberta Court of Appeal found the provisions unconstitutional and not justified as reasonable limits under section 1 of the Charter. It objected in particular to the provisions limiting spending.[47] In a 6-3 decision, a majority of the Supreme Court of Canada found the provisions to be violations of the constitutional protection of free political expression in section 2(b) of the Charter but not to be violations of the right to vote protected in section 3, because they enhanced electoral fairness. The violations of section 2(b) were saved under section 1 as reasonable limits, and the law was upheld.

In the dissenting opinion, Chief Justice McLachlin and Justice Major advanced a view of the state that would fit comfortably within a structural paradigm. They came very close to arguing that the legislation is unconstitutional because it is an incumbent protection device designed to keep the electoral playing field exclusively for political parties. They argued that the legislation works for political parties but not citizens (whom they equated with third parties). They did not take the next step and attribute an improper, unconstitutional purpose to the legislature in devising the rules at issue in Harper, but their minimal impairment analysis under section 1 verged on reaching the same conclusion. They wrote:

It is not an exaggeration to say that the limits imposed on citizens amount to a virtual ban on their participation in political debate during the election period. In actuality, the only space left in the marketplace of ideas is for political parties and their candidates. The right of each citizen to have her voice heard, so vaunted in Figueroa is effectively negated unless the citizen is able or willing to speak through a political party.[48]

I am inclined to favour the majority’s conclusion in Harper that some limits can be imposed on third party spending because of the need to ensure the opportunity for meaningful participation for all individuals in a democracy.[49] The spending limits were sufficiently low, however, that on the particular facts of the case, one should have serious doubts as to the purpose of the legislation and the validity of the provisions, as the dissent suggests. The dissent highlights the need for Canadian courts to at least acknowledge the potential for structural breakdowns. The dissenting justices point out the potential for self-interested election laws to be passed by a House of Commons composed of representatives who generally benefit by retaining control over what is spent and who can speak during election campaigns. The spending limits were low enough that political parties were to some extent squeezing citizens and interest groups out.

Several of the provisions at issue in Harper were “election period” regulations.[50] Because, in parliamentary systems, the government of the day calls the election to begin and end within a short, fixed time period that is generally not known in advance, what election law permits or does not permit during the election period is of paramount importance in those democracies.[51] The potential for conflict between the principle of election period regulation and the preservation of rights to political speech came to a head before the European Court of Human Rights (ECHR), which struck down very restrictive British election period rules on third party spending in Bowman v. United Kingdom.[52] The British government argued the spending cap was essential to ensuring a fair election; the ECHR instead found the legislation to be counter to the guarantees of free elections and freedom of political speech in the European Convention on Human Rights.[53]

Election period regulation raises two distinct structural issues related to the potential for manipulation by self-interested incumbents. First, a governing party may utilize election period regulation to disadvantage its direct electoral competitors, which are other political parties. For example, Canadian political parties face a cap on total spending during the election period. The limit depends on the number of districts contested by each party and the number of electors in each district.[54] The limit was set at approximately $20 million in the last election cycle for the three national parties. The spending limit put the parties on a relatively equal playing field, assuming they could raise or borrow the money to spend the maximum. With the combination of public financing and private donations, parties other than the Liberals and Conservatives are now able to spend the amount. The NDP spent the maximum in the last two elections for the first time in its history. If the governing party were to legislate a much higher spending level during the election period, we could wonder whether it did so to gain an electoral advantage. A higher spending limit would confer an advantage on the parties best able to raise private funds, unless there was a corresponding increase in public financing for all. Election period regulation can be as easily distorted to ensure partisan results as other parts of election law.

Second, election period regulation can be exploited to privilege political parties at the expense of third parties and voters. This structural feature is recognized in Bowman and in the dissenting opinion in Harper. The regulatory approach limiting the role of third parties in an election period is generally justified as an attempt to equalize the electoral playing field so that wealthy interests are not permitted to dominate and determine the outcome of elections by spending vast amounts of money.[55] There is the potential, however, for election period regulation to be so strict as to result in the monopolization of the election period by political parties. This was the case with the spending restrictions in Bowman and to a lesser extent in Harper. The dissent in Harper did not fully develop these ideas, but hinted at their relevance to Canadian democracy.

There is also some writing verging on a structural view in Figueroa v. Canada (Attorney General),[56] where the Supreme Court struck down the requirement that a party must field fifty candidates or more in a general election in order to be a registered political party. Only registered parties had the right to issue tax receipts for donations received outside of the election period, to transfer funds unspent by candidates to the party, and to have their candidates list their party affiliation on the ballot. Candidate deposits were also only party refundable to candidates who won less than 15 percent of the vote, which was often the case for those from small parties. Collectively, these rules were “a matter of life and death” for small parties.[57] The “50 plus” rule worked against small parties who did not have the resources to field a larger slate of candidates, including the Communist Party of Canada led by Miguel Figueroa. The majority in Figueroa focused largely on the need to ensure that the individual voter may participate in the electoral process and considered whether the impugned provisions had that effect. The concurring opinion, however, recognized the role played by political parties in the creation of the provisions at issue, rather than only looking at their consequences. Most importantly, it acknowledged the potential for “manipulation” of electoral rules by political parties.[58] The concurring opinion unfortunately did not expand on this point. The implication raised by the concurring opinion was that the major parties had colluded to pass rules harming their smaller competitors.

A further structural strand can be found in the case law on access to the franchise. It is this area where the Supreme Court has most robustly policed democratic rights. The Court struck down federal legislation disenfranchising felons in Sauvé No. 1[59] and then ruled that narrower legislation disenfranchising only certain felons passed in response to Sauvé No. 1 was also unconstitutional.[60] Again, the Court came close to expressing a structural rationale. The reasoning in Sauvé No. 2 can be seen as going beyond concerns for the individual harm to particular disenfranchised voters. The Court raised the additional concern about the legitimacy of the House of Commons, selected by an electorate that excludes felons, deciding that felons should remain disenfranchised.[61] Raising the legitimacy concern was a major step forward in the jurisprudence because it showed the potential for the democratic process to be flawed. The Court did not, however, take the next step and explain why disenfranchising unpopular groups is likely to be suspect: there is an ongoing risk that incumbents will diminish their accountability to the electorate by disenfranchising voters likely to be hostile to them.

B. Canadian Scholarship

The structural concern with self-interested legislative behaviour has been raised in the Canadian literature,[62] though I would argue the implications of the problems posed by self-interested behaviour have not yet been fully explored. Recent political science scholarship has endorsed the view that prior to 1993, the major parties colluded to exclude smaller parties and to jointly extract benefits from the state to make up for losses due to dwindling social bases of support.[63] Heather MacIvor argues that during this period, parties sought to “promote the security of the ‘ins’ and minimize challenges from the ‘outs.’ The shared interests of the cartel take precedence over the interests of society and the health of democracy, and competition among its members is muted.”[64]

The Court’s jurisprudence has been debated in the growing literature as a choice between an egalitarian and a libertarian model of the law of democracy.[65] Colin Feasby and Heather MacIvor have separately advanced the argument that the Court was adopting an egalitarian model, particularly in its campaign finance jurisprudence,[66] largely in contrast to the more libertarian approach taken by the US Supreme Court.[67] Manfredi and Rush have critiqued the egalitarian argument as an incorrect reading of the case law. They argue that cases such as Libman and Figueroa are at most qualified endorsements of the egalitarian model.[68] Their claim is that there is a great deal of convergence between Canadian and American jurisprudence, with the same issues dividing each supreme court internally.[69]

Feasby has presented the most developed structural approach, what he terms “process theory lite”.[70] In his article, “Freedom of Expression and the Law of the Democratic Process”, Feasby provides a description of the process theory of John Hart Ely, and then a brief overview of the arguments of Issacharoff and Pildes. His goal in the article is to reconcile process theory with both the Supreme Court of Canada’s normative analysis and the egalitarian model that, he argued in earlier articles, was evident in the jurisprudence. He believes that process theory is “germane to the review of the laws that govern the democratic process.”[71] He claims that process theory is consistent with the egalitarian model evident in cases such as Figueroa, Libman, and Harper.[72] Feasby argues that, as most cases are decided at the section 1 stage, process theory should inform the section 1 test in order to deal with the problem of self-interest. In his formulation of the test, courts should defer to legislative objectives, but not to legislative means.[73] He asserts that this is the proper balance between policing self-interested behaviour without judicial overreach. Feasby’s discussion of process theory, while focused more on Ely than on Issacharoff and Pildes, raises key structural concerns in the Canadian context.

C. Current Democratic Practice

Despite the hints of structural concerns in some of the jurisprudence and scholarly pushes in that direction, the Supreme Court currently does not have a framework for understanding and remedying self-dealing. Concerns regarding the lack of structural reasoning by the Court are not merely theoretical. If faced with an electoral map of the type at issue in Carter, it is unclear whether the Court would recognize the harm caused by partisan motives and effects or be equipped doctrinally to fix such a breakdown.

Current democratic practice around campaign finance directly raises these issues. Feasby details how the campaign finance reforms introduced by the Conservative government in 2006 worked to the clear advantage of Conservatives.[74] The Liberal Chrétien government had introduced a public financing scheme for parties based on a subsidy for each vote won in the previous election, while eliminating corporate funding of parties. The regime still permitted corporate donations to ridings or candidates in the amount of $1,000. Individual donation limits were set at $5,000, down from the previous $10,000. The Chrétien plan envisioned public funding of parties with some limited corporate funding directed toward other outlets, in combination with a reduced role for large donors overall.[75] The Conservative government cut the remaining outlets for corporate money, but also capped individual donations at $1,000 adjusted for inflation ($1,100 in 2011). The result was a mix of public and private funding, but with private funding determined entirely by small donations. As the party best able to raise small amounts of money from many individual donors, the new campaign finance regime was a success for the Conservative cause. More reliant on corporate donations, the Liberal Party was comparatively disadvantaged.

The Conservative government continued its attempt to achieve partisan ends through the campaign finance system in 2008 when, closely following its re-election as a minority government, it attempted to eliminate the subsidy each political party was given per vote received in the previous election. The government proposed to take away the subsidy, while keeping the donation limits in place. Again, as the most successful party at garnering small donations from individuals, the Conservatives would have been the major beneficiaries of a purely private financing regime with strict limits on maximum donations. The Chrétien era reforms reduced or eradicated the role for large-scale donors, be they corporations, unions, or individuals, but balanced lower contribution limits for individual donors with public funding. The new Conservative regime would make small, individual donations the sole source of funding for parties and candidates, which also happened to be their donor base.

Whatever the merits of the existing system, which have been much debated,[76] the plan to eliminate public financing was a use of state power for partisan ends. The move from public to private funding would have been problematic because of its anti-competitive effects. The amendments would have further weakened the governing party’s electoral competition, compounding the effect of the earlier limitations on corporation and union donations, under the rhetorical guise of otherwise legitimate democratic goals. The opposition parties thwarted this attempted partisan manipulation of the democratic process by banding together to form a coalition that would have replaced the minority Conservative government. The prime minister prorogued Parliament rather than allowing the coalition to defeat the government.

At the time of writing, the Conservative majority government, elected in May 2011, tabled the Economy and Jobs Growing Act,[77] the budget implementation bill, which would amend the funding formula. The bill is different from the 2008 proposal, as it would gradually phase out the funding. Beginning in April 2012, public party funding will be $1.53 per vote, dropping to $1.02 in April 2013, and down to $0.51 in 2014 before being cut altogether. The elimination of the public subsidy will skew the campaign finance regime in the Conservative Party’s favour. Similar problems are likely to come up going forward, as future governments seek to replicate the success of the Harper Conservatives in amending election laws to enhance their political prospects. Judicial oversight of the law of democracy is likely to be unsatisfactory if the need to address the problem of self-dealing continues to be peripheral to the Court’s decision making. Treating the law of democracy as a clash between egalitarianism and libertarianism addresses competing principles that have guided judicial reasoning. The egalitarian-libertarian lens does not, however, provide a way of understanding self-dealing.

III. The Argument for and Critiques of Structural Theory

A. The Value of a Structural Approach

A structural approach to the law of Canadian democracy is necessary in order to understand current political behaviour and the need to remedy legislative self-dealing. A structural approach suggests that distortions of the democratic process are likely to occur because of incentive structures facing elected representatives and that these manipulations cause constitutional harm. Election law creates the potential for the incentives of representatives to be misaligned with the incentives of voters. This misalignment is an example of the problem of agency, where a principal selects an agent to work on her behalf, but with the risk that the agent may serve his own interests rather than the principal’s. Theories of representation generally view elected representatives either as delegates of the popular will or as trustees exercising independent judgment, or some hybrid of the two.[78] The delegate view is that representatives must act on the wishes of their constituents, even if these are in conflict with the representative’s own opinions. The trustee view sees representatives exercising their own judgment, rather than following the dictates of their constituents. On either view, however, the presumption is that representatives will act on behalf of their constituents. Yet the interests of the governors can deviate from those of the governed. When the interests of representatives do not align with those of the electorate, one can reasonably worry that the interests of representatives will tend to be served at the cost of those of their constituents.

Incumbents have a direct self-interest in the rules of the electoral game in a way that they do not in other areas, raising the chances of misaligned incentives. Those with the highest stake in determining election law are incumbent representatives.[79] Those very individuals are permitted to largely determine the electoral “rules of the game”.[80] Representatives are more likely to pass legislation for the sole purpose of their own private benefit in the law of democracy than in other areas with different incentive structures, such as health care or the environment. Federal elected representatives in Canada are potentially constrained by the section 3 jurisprudence and independent, impartial institutions such as electoral boundary commissions and Elections Canada. Representatives still have significant discretion, however, in crafting the law on campaign finance, party funding, and election period regulation through the Elections Act, and on apportionment and redistricting through other legislation. They are able to structure elections and the constraints on political actors to work to their advantage. The situation is akin to corporate directors or CEOs setting their own levels of compensation. Because election laws can be outcome determinative, the incentives are likely to be high for elected representatives to use their discretion to engage in distortion of the democratic process. Representation-reinforcing or process-based understandings of constitutional law have long understood the need for some oversight of those who govern.[81] The long struggle of African-Americans to achieve equality, and of Aboriginals in Canada, is a testament to the potential for interests to be diminished through distortions of election law.[82]

A structural approach applied to Canada suggests that the diminishment of political accountability through the manipulation of elections laws should be viewed as a constitutional harm. When political actors distort the democratic process, the goal will often be to reduce competition in order to enhance their electoral prospects, as with the map in Carter, the rules struck down in Figueroa, and the recent campaign finance reforms. Diminished political accountability is harmful because it reduces the likelihood that elected representatives will govern in the interests of those they represent. If there is less chance they will be held to account by voters, then elected representatives will be free to govern according to their own preferences. Political accountability should be preserved or enhanced and laws which detract from it should be seen as constitutionally suspect.

B. Critiques of Structural Theory

Structural theory, however, has been critiqued along a few consistent lines in the US literature. I address the major ones here, with the goal of arguing that structural theory is preferable to rights theory as an approach to the law of democracy in Canada. Structural theory has been mainly criticized for: 1) failing to see that individual rights and not systemic structural concerns are what is at issue in election law cases; 2) ignoring separation of powers concerns by fostering excessive judicial involvement in politics under the cover of judicial regulation of the democratic process; and for 3) failing to provide a baseline against which to measure whether the democratic process has been distorted.

1. The Individual Rights Objection

Critics of structural theory are troubled by the emphasis on structural concerns rather than individual rights. Hasen takes issue with Issacharoff and Pildes when they label the balancing of interests that occurs in typical equal protection jurisprudence as “sterile”.[83] Hasen argues that courts are not experts in political science able to determine the ideal type of democratic system. He also claims that the empirical data demonstrates adequate levels of political competition, and therefore Issacharoff and Pildes are unjustified in elevating political competition above the normative emphasis on equality rights in traditional election law jurisprudence.[84] As I interpret this critique, Hasen and other rights theorists worry that by prioritizing structural concerns, structural theory risks obscuring the need to rigorously protect core individual equality rights.

I believe the criticism that structural theories ignore individual rights is largely misguided. As a general matter, the rights versus structure debate has been overblown in the American literature. There are important points of convergence, notably over the need to regulate self-interested behaviour.[85] Structural theories do not claim that individual rights are irrelevant, or that it is group rights or the interests of the state that should be dominant in election law cases. Rather, I interpret them to be arguing that in assessing the meaning of the individual right, courts should be cognizant of the systemic impact of a seemingly small change in election law.[86]

Political party financing, regulation of election period spending, restrictions on third party advertising, redistricting laws, and apportionment rules all form interlocking pieces of the democratic puzzle. No rule can truly be considered in isolation. Altering one rule puts pressure on other aspects of the system.[87] By considering each issue in isolation, a court risks removing a piece of the puzzle without turning its attention to the effect on the puzzle as a whole. This is largely an institutional feature of the judicial system, which is set up to consider one case at a time. Structural theorists merely argue that the court should apply a broader lens and pay attention to the effects of its potential decision on the functioning of the other pieces of the democratic puzzle.[88]

The rights versus structure debate is also overblown in another respect. Rights theorists argue that structural theory ignores textual guarantees of individual rights in the Constitution in favour of constitutional values such as electoral competition only found outside of the text.[89] These values, they argue, are the province of democratic theory, not of courts adjudicating whether the constitutional rights of an individual have been violated. Individuals risk having their rights subsumed under the goal of protecting democratic values like robust electoral competition.

Yet structural theories again assert a less controversial claim than rights theorists accuse them of making. Courts must pour content into the guarantees of individual rights in text, constitutional doctrine, or constitutional values. In determining the meaning of the right to vote, for example, courts need some guide to assess what the particular content of due process, equal protection, effective representation, or meaningful participation actually means. Structural theories do not deny the applicability of individual rights claims. They assert instead that in interpreting individual rights, the right of each individual to hold the government accountable is an overarching concern that should inform the content of the specific constitutional text, doctrine, or values. Issacharoff argues that reading democratic values into the text is an inescapable enterprise, despite the claims of rights theorists.[90] In his view, there is no explicit textual basis for any election law jurisprudence in the United States because the terms “due process”, “equal protection”, and “republican form of government” provide little real guidance, despite the claims by critics of structural theory.[91] The phrase “[e]very citizen has the right to vote” in the text of section 3 of the Charter similarly conveys little on its own to a court seeking to resolve disputes about campaign spending limits or the like.

The debate about individual rights versus structural concerns has been to some extent misleading. As I see it, the main area of disagreement between rights theorists and structural theorists is not about whether one should respect individual rights, but about where we draw the line to distinguish legitimate from illegitimate state action.[92] I would argue that the fundamental question within that debate is how to assess where the need for electoral accountability has been undermined to the degree that constitutional norms have been violated. That is a much smaller scope for disagreement than is commonly acknowledged.

2. The Separation of Powers Objection

Structuralism has been accused of fostering judicial involvement in politics in violation of the separation of powers. In their comparative study on Canadian and American election law, Manfredi and Rush argue, for example, that Issacharoff and Pildes’s process-based theory is actually animated by a normative claim about the “right answer” in a given case.[93] They argue there is a normative vision underlying the emphasis on process. In their reasoning, judicial overreach will occur if process-based theories win the day because courts will be empowered by this hidden normative vision to interfere with legislative decisions in breach of the separation of powers.

Richard Hasen formulates a similar critique by arguing that process theories have an “implicit normative agenda”,[94] namely political competition, which Issacharoff and Pildes substitute for other values such as political equality. As a result of this normative bias, the view of politics as markets will lead to “intrusive judicial involvement”[95] and “judicial hubris”.[96] Hasen argues that Issacharoff is a “revolutionary” for writing that electoral maps drawn by legislatures should be considered unconstitutional.[97] Hasen believes that if Issacharoff’s views are taken to their logical conclusion, they can lead to no other end than the elimination of single-member plurality (SMP) electoral districts through undemocratic judicial fiat.[98] He argues that Issacharoff’s theory provides “no limiting principle to separate permissible from impermissible state regulation of elections.”[99]

This is a serious charge if correct. If courts are empowered to reach too far into the political realm, this is a problem of democratic legitimacy. The US Supreme Court’s two most prominent election law cases in recent years, Bush v. Gore[100] and Citizens United,[101] have been heavily criticized as unhelpful intrusions by the Court into the law of democracy.[102] Given the arguable failure of the Court in these cases, critics of the structural approach have reason to ask why a theory of the law of democracy that encourages judicial oversight of the democratic process should be advanced.

Pildes, and Charles separately in a recent article, argue that this critique misconstrues the effects of judicial oversight.[103] By breaking up lockups, courts will allow the process to function as it should. They argue that getting the fundamentals right means courts will face less pressure to intervene in the democratic process downstream. Put another way, the assumption appears to be that if the legislature is composed fairly, and the rules of the electoral game are set fairly, less manipulation will occur later on. As Charles formulates the argument, by paying attention to the democratic framework and intervening to end structural breakdowns at that level, courts will have less of a need to intervene in day-to-day politics.[104] The implication is that, while free to roam in the land of democratic process, courts will have less legitimacy to trespass into the land of politics on most occasions.

This counter-argument by the structural theorists would seem a difficult one to prove. They are claiming in one respect that a legislature whose composition is determined on fair terms will be less likely to act in a way that is unconstitutional. Pildes gives the example that there will be less need for court protection of “first-order issues of equality and liberty” because with fair competition parties will be held accountable for violating them.[105] This seems correct to an extent. If barriers harming small political parties are removed, and small parties are represented in the legislature, they will exercise their power to make it less likely that laws will be passed which put them at a competitive disadvantage.

At best, however, this seems plausible under limited conditions in my opinion. If a small party is elected to a legislature, a majority committed to re-introducing new barriers to entry could still outvote it. Courts operating under a structural approach may rule these barriers unconstitutional, but extensive judicial review would still seem necessary. The argument that less judicial review will be required if the fundamentals are right runs counter to the observation running through structural theory that parties engage in a continuous process of evading whatever new rule is put in their path. Even if the fundamentals are right, we will likely still require judicial intervention downstream.

Arguably, Aboriginal voters in Canada have been a discrete and insular minority largely excluded from majority electoral coalitions since Confederation due to their small numbers in all but a few electoral districts.[106] They have often been the subjects of discriminatory action by governments of all stripes. In that scenario, even conditions of fair competition respecting the right of Aboriginals to participate meaningfully in the electoral process would not lead to a reduction in the need for judicial review of rights violations.[107] Fair competition does not necessarily increase the electoral influence of a numerically small minority to a level sufficient to dissuade parties from engaging in politically popular anti-minority legislation.

I believe there is a more persuasive argument, often left implicit in structural theory, to rebut the argument that structural theory violates the separation of powers. The recognition that a legislative majority can manipulate electoral rules to organize the game in its favour weakens the argument against the legitimacy of judicial review. A majority that enacts an electoral law to perpetuate the disadvantages faced by its competitors would be entitled to little deference on separation of powers grounds. A majority constituted on unfair electoral grounds is not a legitimate majority.[108] Legitimacy in a democracy is partly determined by elections. If unaccountable to the actual preferences of the electorate because of the manipulation of electoral rules, a legislative body’s enactments cannot be equated with the preferences of the electorate. The primary rationale against judicial review of election law is therefore significantly weakened.

Consider a concrete example. Under legislation that banned women from voting, a legislature composed under those conditions would not be an accurate representation of the preferences or interests of its population. If the legislature is illegitimate, and a court acts to increase accountability and to ensure that the elected representatives actually represent the population, then the counter-majoritarian difficulty is eased in the specific context of the law of democracy. Judicial review can still potentially cross the line established by the separation of powers. Structural theory, however, invites judicial review where it is most needed—when the legislative majority has not been constituted in a democratically legitimate fashion. As Pildes frames it, structural theory seeks not more judicial review, but “more focused and better justified” judicial oversight.[109]

3. The Objection Against Political Competition

The last major critique of structural theory is that it provides no meaningful measure by which to assess the legitimacy of state action. Hasen, for example, argues that structural theory is too “shallow”.[110] He asserts that Issacharoff and Pildes are inconsistent in their application of the principle of political competition and, in any event, the principle itself is inadequate as a measure of democratic legitimacy.[111] The heart of the critique is that Issacharoff and Pildes provide no meaningful rationale to determine when institutions such as legislative committees charged with redistricting, or rules such as campaign finance regulations, violate the principle of political competition.[112] On this view, democracy entails equality rights and substantive outcomes, and theories of election law should further these concerns rather than focusing on procedural attributes such as political competition.

Hasen is correct that structural theory emphasizes political competition and accountability as necessary aspects of a democracy, rather than substantive visions of what political equality means. This feature of structural theory should be regarded as a virtue, however, because it actually responds to the criticism addressed earlier in the article that structural theory would lead to excessive judicial review. Embedded in structural theory is the recognition that there are limits to what review or oversight of the legislature is democratically legitimate. A structural approach is intended to function as a theory of the law of democracy, not a complete democratic theory in and of itself. It is a theory of when judicial review or oversight by an intervening institution, such as an electoral boundary commission, is necessary to check defects in the democratic process.

Theories of the law of democracy must provide a rationale for interventions to end breakdowns in the democratic process. The rationale must be robust enough to justify oversight of the “political thicket” where needed, without legitimizing judicial involvement in everyday politics. For the purpose of determining the appropriate boundaries of judicial review, the adoption of a minimal view serves this end better than a more robust view that may require even greater levels of intervention. A minimalist view of democracy focused on competition and accountability sets a reasonable boundary line for when and on what subject matter courts or other institutions may intervene in democratic politics. If a more robust theory of democracy is applied to justify intervention in the law of democracy, one runs the risk of setting the line too far on the side of judicial review and ignoring the counter-majoritarian dilemma.

If we use a more robust set of norms than ensuring accountability through electoral competition, then there is the need to regulate a broader set of practices and outcomes. Theories focused on the law of democracy must acknowledge not only the breakdown potential in the legislature and elections, but also the fear of overreaching by the judiciary. The fear of judicial overreach identified by rights theorists is a viable concern, especially now after Bush v. Gore and Citizens United. The adoption of a minimal set of required democratic norms in structural theory reflects this concern for potentially excessive judicial intervention. The emphasis on competition acknowledges that when parties shield themselves from electoral accountability, they violate the minimum expectations we should have of a democracy. A structural approach to the law of democracy is a legal theory based in democratic theory, not a complete democratic theory tangentially related to legal concerns such as the appropriate contours of judicial review.

IV. Breakdowns in the Democratic Process

A. Understanding Breakdowns

A structural approach to the law of Canadian democracy suggests greater focus is needed on those instances where self-dealing leads to breakdowns in the democratic process. In this section, I develop the concept of a breakdown in the Canadian context from a structural perspective and delineate the main types of breakdowns. Breakdowns in the democratic process become likely when the process is structured in a way that the interests of incumbent elected representatives are likely to deviate from those of the individuals they represent. The risk is that elected representatives will select election laws that function to their benefit, rather than to the benefit of the electorate. The distortion of the preferences of the electorate can occur in myriad ways. For instance, when a group of elites controls the political system by virtue of excluding its opponents from the electorate, we can see this as a breakdown. Apartheid South Africa, as well as Canada, the United Kingdom, and the United States before the franchise was expanded to women and un-propertied men are all examples. Issacharoff and Pildes reinterpreted the well-known “White Primary Cases” not as examples of different approaches to judicial review, as had been the dominant trend in the literature, but as distortions of the political process by White elites.[113] Whites were a majority in the American South, not a minority, but White political elites operating in a de facto one-party region of the country colluded to prevent African-Americans and poor Whites from forming coalitions to dispel their rule. They did so by manipulating the rules on primaries to disable electoral competitors.

The comparative study of the law of democracy in mature democracies has entered a new phase accompanied by a new generation of concerns. Jurisprudential victories in voting rights litigation have eliminated many of the overt restrictions on the ability to cast a ballot or run for elected office that had been directed against specific racial or ethnic minorities, women, and particular economic classes. In Canada this was achieved largely through the Charter, with its explicit guarantee of the right to vote. Despite no clear textual guarantee of the right to vote, the United States reached the same result through the reinterpretation of earlier textual guarantees, such as the 15th Amendment, or broadening the equality rights doctrine under the 14th Amendment.[114] Restrictions on the franchise for felons in the United States are one example of disenfranchisement that persists.[115] While restrictions on the franchise still exist, the general trend has been toward opening access to the franchise.[116]

Outstanding issues in the law of democracy largely concern fair methods of aggregating votes, such as in redistricting, or regulations concerning the role of money in politics and what limits are justified on political speech.[117] In the redistricting context, issues of aggregation are more subtle and perhaps less obvious than the question of individual access to the ballot box. The analogy would be to equality rights jurisprudence. Courts have a relatively easy time resolving doctrinal confusion over the basis for prohibiting direct discrimination in areas like employment and housing. Indirect discrimination claims are much more complex. They involve delving into unconstitutional motives and assessing the effect of a facially neutral law on an aggregation of individuals most likely to be disadvantaged.

The law of democracy in Canada is in the indirect discrimination phase of its evolution, to continue the analogy. Determining the constitutionality of legislation prohibiting all citizens of Japanese ancestry from voting in federal elections is straightforward for courts, for example, despite earlier uncertainty in both Canada and the United States as to whether such “political questions” should be off the judicial table.[118] Assessing whether a facially neutral piece of election law, however, has an indirect but unconstitutional impact upon Japanese Canadians is a different endeavour entirely. The same could be said for assessing whether the electoral and legislative processes that led to that type of legislation violate norms of political equality. A theory of breakdowns in the democratic process engages these second generation concerns about how facially neutral rules can distort the expression and/or fair aggregation of voters’ preferences.

B. Types of Breakdowns

In their original formulation of the idea of “lockups” in the democratic process in the United States, Issacharoff and Pildes identify two types.[119] Lockups may be precommitment pacts among elites to deny non-elites the ability to engage in the political process, such as the prohibitions on African-American political participation. Lockups may also raise barriers to entry into the process for potential challengers. Think here of rules that require large upfront deposits for primary candidates or leadership aspirants that will likely discourage unknown challengers from entering the fray, or of rules restricting what interest groups can spend on political expression.

Precommitment pacts and barriers to entry are important types of problems for the democratic process, but may not encompass the full range of democratic breakdowns. The theory of Issacharoff and Pildes operates at a relatively high level of generality, rather than seeking to delineate to a precise degree how anti-competitive behaviour occurs. Structural theory could more squarely address the underlying issue of explaining political behaviour. In particular, we need to expand our understanding of the consequences of the exercise of self-interest. Self-serving legislative behaviour may take a variety of forms, and will occur in specific areas of the law of democracy more than in others.

A broader understanding of what types of breakdowns may occur in Canadian democracy would therefore be useful and I attempt to provide one here through a typology of breakdowns. There is a variety of potential types of breakdowns that are identifiable from the perspective of a structural account focused on competition and accountability. I identify three general types: partisan, incumbent, and interest entrenchment breakdowns.

The recognition of partisan, incumbent, and interest entrenchment considerations in the design of election laws and institutions, and of harms flowing from them, is of long-standing concern in the literature in the law of democracy, especially in the United States. The US Supreme Court recognizes the potential harms produced by partisan gerrymandering, for example.[120] Considerations of partisan harm, however, have been made in relation to specific activities, such as redistricting, rather than tendencies in political behaviour across the law of democracy as a whole. I intend these categories to apply across the various aspects of the law of democracy, as general types of political behaviour that should be of concern and not tied to a specific practice such as gerrymandering.

The three types that I propose overlap to some extent, as they all may involve self-interested manipulation of election laws to evade electoral accountability, yet they are conceptually distinct as they each address different motivations held by political actors and allow courts to consider what kinds of interests are being illegitimately entrenched. Outlining the broad categories of problematic political behaviour, as I attempt to do here, is analytically useful as it allows us to distinguish between different types of motives held by political actors and different consequences of election laws. It is also of importance for judicial review, as it provides a framework within which courts can analyze laws or rules justified by legislatures on principled grounds, but which are fundamentally about furthering partisan, incumbent, or other interests.

In the US literature, Michael Klarman argues for an approach to developing forward-looking categories to classify problematic political behaviour that should be rectified by courts. Klarman advances a general theory of judicial review, which he calls the “anti-entrenchment” view, as a response to the counter-majoritarian critique of judicial review.[121] He argues that courts should prevent illegitimate entrenchment of power by legislative majorities. Legislatures can seek to entrench their power against current political majorities, which he calls legislative entrenchment, or against future majorities, labelled as “cross-temporal” entrenchment.[122] Klarman applies these two categories of entrenchment to a variety of different aspects of the law of democracy, from ballot access to campaign finance. Klarman’s category of legislative entrenchment is sufficiently broad to hinder its predictive use.[123] It is preferable, in my opinion, to look directly at the types of interests that lead to failures in the democratic process, rather than broadly at legislative goals. Different implications may flow from whether the breakdown is partisan or incumbent, which is not incorporated into Klarman’s scheme.

Klarman’s “cross-temporal” category and my own approach raise what Adam Cox calls the intertemporal dimension of the law of democracy, which he argued has been largely ignored.[124] Cox elucidates the extent to which considerations of time operate implicitly in the reasoning of courts when they assess the constitutionality of election laws. He argues that while voting rights are properly seen as aggregated and not individual rights, the basis of aggregation has been conceived of too narrowly. In his view, aggregation of voters’ preferences can be considered not just spatially, as in aggregating votes within an electoral district, but intertemporally. He raises the issue of the time frame within which the fairness of a voting rule should be considered.[125] For example, the constitutionality of an electoral map’s impact on an aggregation of minority voters may need to be judged not only in relation to other voters under the current map, but also in regard to the position of minority voters under previous electoral maps at earlier points in time. Whether, and how, to weigh such claims is developing into a larger aspect of voting rights litigation.[126]

Klarman raises these temporal issues, but the distinction between legislative and cross-temporal entrenchment is not sufficiently developed. All legislative decisions have implications for future majorities, as do many other types of legal rules. Rather than viewing intertemporal issues as a separate problem, as Klarman does, it is better to view the intertemporal dimension as arising in each potential type of breakdown. One can push Cox’s notion of the intertemporal dimension of election laws in a new direction by placing it within a structural view of the law of democracy that is concerned with when the democratic process fails. Partisan, incumbent, and interest entrenchment breakdowns can be understood as occurring when there is an attempt to perpetuate political control by one aggregation of voters, group of elected representatives, or set of interests beyond the time when the electorate’s preferences would otherwise distribute power differently. By manipulating electoral structures, representatives can seek to limit the impact of shifts in the public’s preference for elected representatives. A governing party elected at time 1 may anticipate that the preferences of the electorate will change at time 2 to its partisan disadvantage. In order to prevent the loss of its power, the governing party utilizes its control over the state to enact laws designed to keep it in power even after the preferences of the electorate shift at time 2. Even if voter preferences remain constant, electoral rules can be manipulated to ensure that the result differs from election to election.

1. Partisan Breakdowns

A partisan breakdown can occur when a government passes legislation distorting the democratic process in order to provide it with an advantage over its competitors. On a structural theory of the law of democracy, where political actors are likely to seek to enhance their electoral chances in relation to their competitors, partisan breakdowns are a serious concern. By manipulating electoral laws, political actors possessing state power can create underlying conditions more favourable to their own partisan interests. The attempted use of election law for partisan gain is one recurring theme in the law of democracy in Canada. The problem is structural in nature, and not tied to any particular political party. Whoever is in power will be likely to have an incentive to use the state for partisan advantage. Governments, of course, take policy decisions that reward their ideological and partisan supporters all the time. Such responsiveness can act as a bridge between electioneering and governing and often leads to greater accountability. We want the policy priorities of parties to be closely linked to those who support them, so there can be a contest between parties. Unjustified partisan distortion of the democratic process occurs when the government alters electoral structures in an attempt to ensure its re-election or diminish its likelihood of defeat.

In large case studies in political science on choice of electoral system, there is convincing evidence that the move to proportional representation (PR) in European democracies was the result of self-interested manipulation of the democratic process. On one understanding, Conservative governing parties feared their diminished electoral prospects against new socialist parties as the franchise was expanded to include workers.[127] On another, not only democracies in Western Europe between the World Wars but also new democracies in Eastern Europe in the 1990s adopted PR as a result of strategic behaviour by political actors seeking partisan gains in the face of tremendous uncertainty.[128] Under either scenario, the defining consideration in the choice of electoral system was partisan advantage.[129]

Partisan gerrymandering by the majority party against the minority party is a classic example of a partisan breakdown in contemporary politics. In the United States, redistricting is generally conducted by state legislatures for state and federal districts, though some states have now adopted redistricting commissions.[130] The constitutionality of partisan gerrymandering has vexed American courts[131] and led to intense scholarly debate.[132] Redistricting by the legislature puts those who have the most self-interest in the design of electoral districts, elected representatives, in charge of designing the very districts they must contest on election day. A legislator tasked with redistricting may have significant incentives to draw a map favourable to her own re-election and her party’s success.

The introduction of independent, non-partisan electoral boundary commissions in Canada in 1964, a veritable “electoral boundary revolution”,[133] eliminated partisan gerrymandering for federal districts. Partisan redistricting remains a potential problem at the provincial level where legislatures are permitted to draw electoral maps or be involved extensively in designing the redistricting process.[134] The Carter decision raised briefly in the introduction also appears to have entailed a partisan breakdown in provincial redistricting in Saskatchewan. The map imposed by the Saskatchewan Conservative government resulted in less voter equality than the previous map, overrepresented rural, northern voters who were part of its core base of supporters, and under-represented the urban voters who were more likely to vote for other parties. Collectively, these points raise a strong suspicion that the map was the product of partisan motivation and would have partisan effects. Even where commissions design electoral maps, both provincial and federal governments retain the ability to overrule the commission in part or in total by legislating a new electoral map if they do not like the results dictated by a commission. Legislatures have delayed the implementation of a commission-designed electoral map federally,[135] and provincially in British Columbia, the Legislature criticized the commission until it amended its proposals, often with justifications of dubious merit.[136] This power can potentially be abused if a government delays the implementation of an electoral map, or modifies the map, to ensure greater electoral success. As the discussion of recent democratic practice in Section II indicates, campaign finance and reform also often contains partisan motivation and effect.

2. Incumbent Breakdowns

An incumbent breakdown can occur when a majority of existing representatives cooperate across party lines to pass legislation that disadvantages either a legislative minority or political parties not yet represented in the legislature.[137] The sweetheart gerrymanderers that emerged in the United States during the rounds of redistricting in the 2000s fit with this type.[138] In sweetheart gerrymandering, both parties agree to support an electoral map that limits competition against incumbents of both political stripes.

The provisions of the Elections Act struck down in Figueroa can be understood as incumbent breakdowns. Those provisions placed small parties and parties without sitting MPs at a disadvantage in comparison to the larger parties represented in the legislature. Generally, any rules which make it easier for large parties to engage in electoral competition but harder for small parties, or which provide added resources to large parties represented in the legislature can be understood as incumbent breakdowns. These types of rules raise the barriers to entry into the political realm, thereby deterring challenges from new parties.[139]

Distortions of the democratic process favouring incumbents, and political parties in general, are prevalent in the rules regulating political parties in Canada. Chris Bredt and Laura Pottie argue that “[m]uch of the current electoral regime is clearly designed to protect and promote established parties and/or incumbents.”[140] Candidates and political parties who receive a certain threshold of the vote obtain partial reimbursement of their expenses.[141] By making eligibility for reimbursement of electoral expenses contingent on electoral performance, smaller parties are disadvantaged in relation to larger, more successful ones. Broadcasting time allocated to parties is also dependent on electoral performance.[142] Such rules were implemented by legislatures composed of elected representatives who stood to benefit directly. They raise the barriers to entry for new parties and decrease the incentives for small parties to compete. There is little substantive difference between these rules and the threshold rules that assisted the large parties that were struck down in Figueroa.

Public funding for political parties can also potentially be classified as an incumbent-protection device, depending on how it is implemented. In Australia, public financing is conditional on the party receiving at least 4 percent of the first preference votes in their preferential voting system.[143] Independents can receive the funding as well, assuming they gain 4 percent support in a district, but the threshold ensures that many small political parties are excluded from the public financing regime. Given the connection between popular support and funding in the legislation, the most successful parties receive the majority of the subsidy.[144] For parties receiving less than 4 percent of the vote, the lack of funds is likely to be damaging enough, yet even those small parties receiving the subsidy face a public funding regime that rewards the large parties first. Similar critiques of incumbent protection can be levelled at the UK party funding regime,[145] as well as the Canadian one.[146] Canada’s public funding scheme, based on a per vote subsidy, assists existing parties and incumbents and rewards the larger parties above the smaller ones. A preferable alternative might be a base level of funding provided to each party, rather than a subsidy based on voter support. The existing standard has the merit of tracking popular expression of support for a party, but aiding the largest parties most of all may harm competition.

Even though small parties rarely gain seats in Parliament, additional barriers to entry for small parties are not a trivial result. Small parties play an important role in disciplining the major parties and in keeping them accountable. The threat of electoral competition from small or new parties provides incentives for major parties to internalize the values of their small electoral competitors in order to co-opt their supporters or to prevent defection by their own supporters. Major parties can be expected to move along the political spectrum in order to neutralize the actual or potential emergence of electoral competitors. If parties do not adapt, the newer entity can become ascendant. The replacement of the Progressive Conservative Party with the Reform Party can be explained in these terms.

Political parties as a whole also receive benefits that other entities do not. The interest groups or third parties whose role in the democratic process was at issue in Harper compete with political parties for influence with the electorate. Political parties are granted numerous advantages by election laws that third parties are not. They receive access to voter information, to broadcasting time during elections, and can provide their donors with tax deductions. Third parties receive none of these advantages.[147] The limits on third party advertising that split the Court in Harper are part of a broader trend in Canadian election law that disadvantages third parties in relation to political parties.

Justifications offered for rules that function as incumbent-protection mechanisms often include appeals to democratic stability or the provision of strong government. In Figueroa, the minority judgment authored by Justice Lebel justified the “50 plus” district restriction because it ensured “[t]he promotion of cohesion over fragmentation.”[148] Even if we accept stability and strong government as legitimate aims in election law (though I would be inclined to view these claims skeptically as potential covers for self-dealing), much of this function is fulfilled by the electoral system itself. Single-member plurality districts of the type used in Canada generally result in majority government and two-party competition for power, a phenomenon known as Duverger’s Law.[149] Stability is therefore provided by the choice of electoral system and subsequent election laws are likely to add little to this probable outcome. A structural approach to the law of democracy suggests that we should be skeptical of appeals by incumbents to abstract values such as democratic stability when the particular law under consideration aids incumbents.[150]

The history of restrictions on the right to vote, now prohibited under section 3, also show the tendency for incumbent-protection distortions. For example in British Columbia, Asians (mainly Chinese, Japanese, and those from the Indian subcontinent) were excluded from voting until 1948, while Aboriginals gained the full right to vote without having to sacrifice their official Indian status only in 1960.[151] Their disenfranchisement in the province was partly a result of discriminatory attitudes, but the underlying reason was that together Asians and Aboriginals would have held a majority of the BC electorate (over 60 percent) until quite late in the province’s existence.[152] Representatives of the minority British and those of other European origins acted to disenfranchise Aboriginals and Asians so they could assume political control without interference.[153] Unable to win a legislative majority when faced with the whole electorate, they instead sought to use public power to alter electoral rules so that their interests would remain paramount.

Urban vote dilution is a notable incumbent breakdown currently in effect in Canada.[154] Under the Electoral Boundaries Readjustment Act, federal electoral districts can deviate from population equality among the districts in a province by 25 percent above or below the provincial average, or by an undefined greater amount in extraordinary circumstances.[155] Deviations from one person, one vote were upheld in Carter. The tendency among boundary commissions has been to design urban districts with larger populations than rural districts. Rural votes are on average worth more than urban votes and, at times, quite substantially more.[156] The result is a House of Commons skewed away from representation by population. Overrepresented rural incumbents will have strong incentives to resist moves toward population equality, as they would be in danger of losing their seats or seeing their influence diminish with the addition of more urban MPs. This is likely to be the case regardless of their political stripes.

The common thread with incumbent breakdowns is that most of those currently elected benefit from the existing electoral rules of the game. They are therefore reluctant to change these rules and, as a consequence, political coalitions and actors who are disadvantaged by the current system are likely to remain so, absent some new political dynamic. Like partisan breakdowns, incumbent breakdowns flow from the misalignment between the interests of representatives and those of the governed. Incumbents develop an interest in preserving the status quo or in manipulating it in a way that decreases their likelihood of being removed from office. Incumbent breakdowns occur because incumbents of different political affiliations all have a common interest in remaining insiders.

3. Entrenchment of Interests

The first two types of breakdowns involve furtherance of the interests of political parties. Breakdowns in the democratic process can also occur where other types of interests are entrenched. These interests may be religious, racial, or regional, or may be tied to some other demographic group. Interest entrenchment breakdowns are the hardest type of breakdown to assess. The line between what is an illegitimate freezing of political power and what is simply a legitimate legislative precommitment pact required to bring parties with different interests together is a fine one. To some extent, all governments seek to entrench their policy preferences over time. Yet constitutional convention in parliamentary democracies generally prevents one Parliament from binding future Parliaments. A new Parliament retains the ability to amend or repeal legislation passed by its predecessors. This convention institutionalizes the recognition that new political majorities should have leeway in enacting their own policy preferences. Path dependency remains a significant factor in how this plays out in practice, as does legislative inertia.[157] Once a policy is in place, it is often hard to displace even by a newly elected and ideologically opposed government. The principle that earlier legislative majorities should not be able to formally bind future majorities, however, remains operative. Interest entrenchment breakdowns are attempts to deviate from this majoritarian feature of democracy.[158] Such a breakdown occurs when incumbents seek to pass legislation at a point of political power that will insulate them from political accountability if and when the electorate’s preferences change.

To come into being or to expand, states may need to strike bargains between geographic regions of the new state or between the majority and minority groups, whether to accommodate minority interests or to achieve the political support sufficient to move forward as one entity. Institutional arrangements, such as quotas on legislative representation, power sharing between ethnic groups, regional appointments to courts, and rules on cabinet formation may be adopted to achieve the necessary agreement at a point in time given the current distributions of political power, wealth, and demography, among other factors. At some point these accommodations may devolve into a breakdown of the democratic process that prevents the interests of today’s voters from being realized through politics.

Consociational democracies with formal power-sharing arrangements, such as in present-day Iraq with alteration in political power between religious groups,[159] or Lebanon, which froze power based on demographics in 1940 between Sunnis, Maronites, and Shia,[160] are likely to be particularly at risk. Institutional arrangements that seek to balance competing principles of representation, such as representation by population and regional representation, may also become suspect over time. For instance, in the bicameral US Congress, the Senate has two representatives per state regardless of population while the House of Representatives tracks population. As the populations of fast-growing states outpace those of the less populous states, these arrangements may become problematic.[161] Additionally, where institutional arrangements are the result of majoritarian institutions or elite accommodations that excluded or disregarded the affected minority, the political action is less likely to be a legitimate precommitment pact, and more likely to be an interest entrenchment breakdown.

Interest entrenchment breakdowns often have an intertemporal dimension that may distinguish the type further from partisan and incumbent breakdowns. Interest entrenchment breakdowns often occur over longer time frames than partisan or incumbent ones, which tend to be directed at the next election cycle. This intertemporal aspect arises because demographic changes, such as movement in the proportion of a population divided among religious groups, generally happens over greater time spans.

Examples of interest entrenchment breakdown in Canadian democracy are the rules determining representation for the provinces in the House of Commons. In 1915 and 1985, the rules determining representation in the House for the provinces were altered to entrench prevailing levels of representation for provinces faced with losing MPs because of relative or absolute population decline. In 1915, the Senate clause was introduced, ensuring that no province may have fewer seats in the House than it has Senators, and was later constitutionalized.[162] The result was to increase the representation of the Atlantic provinces from what it would be under representation by population.[163] The Representation Act 1985 introduced the “grandfather rule” guaranteeing that no province could have fewer MPs than it had in 1986, the 33rd Parliament.[164] The practical effect of the grandfather rule is to protect the entitlements of the prairie provinces.[165] The Senate and the grandfather rules are deviations from the principle of majority rule that would be the result of representation by population. Seven out of ten provinces benefit from deviations, to the detriment of Ontario, British Columbia, and Alberta. The population of these three provinces is more than half of Canada’s total, concentrated mostly in Vancouver, Calgary, Edmonton, and Toronto, but they have a smaller percentage of representation in the House than their populations warrant.

The guarantees agreed to in 1915 and 1985 of a minimum number of electoral districts for the less populous provinces can be seen as attempts to decrease the continually growing influence of voters in the largest provinces. This was done to preserve the electoral power of the smaller provinces against demographic change in an urbanizing country where population growth was concentrated in certain provinces.

There are certainly arguments for ensuring the effective representation of voters in all provinces in a federation.[166] The legitimacy of guaranteeing a minimum number of seats for the less populous provinces would be on more stable ground, however, had the rules been the result of legislative or constitutional debate by representatives who were disinterested in the result. It is difficult to see a majority of voters today agreeing to the Senate or grandfather rules. In each case, Parliaments that represented certain interests entrenched the overrepresentation of those interests in future Parliaments to the detriment of future electoral majorities. The Senate clause is constitutionally enshrined and therefore unlikely to change, but the grandfather clause was passed pursuant to ordinary legislation and could therefore be replaced by a parliamentary majority. The entrenchment of regional interests by those two rules, however, means that the body of legislators with power to alter the rules is unlikely to do so, due to the overrepresentation of those interests in that body.

The current government has, on four separate occasions, introduced bills to address this problem by keeping the allotment of the less populous provinces constant while adding seats to Ontario, Alberta, and British Columbia. These efforts, however, have floundered in the House of Commons due to opposition by MPs from provinces whose relative influence would decrease.[167] At the time of writing, new legislation adding seats to the most populous provinces has again been introduced by the government.[168] The intuition in a structural account that election law reform will be difficult to achieve if against the interests of existing representatives has unfortunately been validated in the debate over representation by population.

The example of representation by population thwarted by regional interests was the result of strategic behaviour. Interest entrenchment may also occur as the result of otherwise legitimate bargains that become less so over time. The Voting Rights Act[169] (VRA) of 1965 has undoubtedly become the centerpiece of US election administration. The VRA protects minority voters through a variety of measures, including what is known as pre-clearance. Election administration is generally devolved to the states, but the VRA requires that state election laws be pre-cleared by federal officials prior to implementation in states traditionally hostile to minority voting rights. Even scholars supportive of minority voting rights, however, have begun to question whether this entrenchment of racial and ethnic interests remains constitutional, given demographic changes and the improving fortunes of some communities.[170] The debate over the VRA cannot be done justice here, but the point is that while interests may legitimately be entrenched at one juncture in time, shifting conditions at a later date may undermine the legitimacy of that entrenchment.

C. Implications for Judicial Review

A full consideration of the myriad complexities involved in the judicial review of the law of democracy is beyond the scope of this article, but some initial conclusions flow from this analysis. Judicial review of the law of democracy in Canada should be centered on preventing the three types of breakdowns of the democratic process identified here. At times, discerning partisan motivations and partisan effects of an election law, for example, will be straightforward, as in Carter. The Court would have had ample grounds to reject the gerrymander as counter to section 3 and not justified by section 1 of the Charter, and should have done so. Where partisan motivations are hidden, partisan effects may alone be sufficient to violate section 3, on a structural reading of the provision, if political competition would be reduced sufficiently so that the ability of voters to hold the government to account will be impaired. An electoral map that was not motivated by partisan reasoning, but which would harm political competition because of how voters are distributed, is problematic as well from this point of view.

More difficult cases, however, arise when partisan motivation and effects are present, but the action taken by political actors would be otherwise constitutional.[171] Elected representatives of course operate with a mix of public and private motivations. They may have improper motives and seek a desired effect, but may still pass legislation that would otherwise not raise troubling constitutional questions. This problem is evident in the example raised earlier of the Harper government’s plan to eliminate the public subsidy for political parties.[172] To do so would have a clearly partisan impact and would appear to be motivated at least partly by partisan considerations. Yet there are also defensible, principled reasons for preferring private to public funding of parties, whatever the motivation behind the legislation. Prior to the introduction of the subsidy, the Court had never held that public funding was constitutionally required, and there was little indication in the jurisprudence that such a change was forthcoming.

A court reviewing elimination of the subsidy under section 3 should investigate the presence or absence of partisan motivations and effects, and weigh its findings against the constitutionally permissible scope of action. Where the action would otherwise be constitutional, but political competition would be so reduced as to harm the ability of voters to hold the government to account, I would argue that there is a serious constitutional problem. On this reasoning, the elimination of the party subsidy should be seen as unconstitutional if political competition is severely reduced, as appears to be the case. As the problem emanates from the incentives facing legislators, it is unlikely to be fixed in the legislature itself, and should, therefore, be remedied by courts.

Governments may amend election legislation in ways that demonstrate that despite partisan consequences, they are not seeking to reduce competition and accountability, thereby fixing some of the constitutional deficiencies. After being re-elected in 2011 with a majority, the Harper government reconfirmed its commitment to eliminating the party subsidy, but modified earlier approaches by stating that the phase-out would be gradual to give all parties time to adjust to needing new sources of funding.[173] These amendments would lessen the partisan impact and, if implemented, should be viewed favourably by courts as facilitating political competition in a manner that earlier legislative proposals did not.

This example suggests that the crafting of remedies can also be guided by the typology. If a partisan breakdown is at issue, courts can require greater competitive balance between parties. If incumbent protection is before the court, then remedies should be crafted that decrease the burden on small parties or those not represented in the legislature. Where interest entrenchment is being considered, then courts can investigate the likely impact of the law on other interest groups, especially as projected over time.

Conclusion

A structural theory of the law of democracy provides a descriptive account of why breakdowns occur based on a view of election law as potentially subject to self-interested manipulation by elected representatives and a normative understanding of why such manipulation is problematic. Applying this structural theory, I developed a theory of breakdowns in the democratic process in Canada focused on partisan, incumbent, and interest entrenchment breakdowns. Because of the incentives that exist to manipulate the democratic process to achieve electoral gains, the history in democracies of such manipulations, and the problematic nature of some current practices, breakdowns are likely to pose an ongoing dilemma.

A structural account of the Canadian law of democracy suggests avenues for future research. I have argued that breakdowns in the democratic process exist in the areas of redistricting, apportionment, campaign finance, choice of electoral system, regulation of political parties, and the rules governing third parties. All of these areas could be the subjects of more detailed accounts of how each particular breakdown is perpetuated. While the literature has addressed the 2003 and 2006 campaign finance amendments to some extent,[174] the later dynamics of campaign finance and the other areas of the law of democracy should be the subject of further analysis.

The role of courts in the law of democracy remains a live issue. At the level of judicial doctrine, a structural account suggests a role for courts in reviewing the law of democracy to counteract the likelihood of manipulation of the democratic process along incumbent, partisan, or interest entrenchment lines. Feasby’s proposal that in reviewing the law of democracy under section 1 courts should show deference to legislative objectives, but not to legislative means, is a valuable contribution to the debate. I am inclined to be more skeptical than Feasby not just of legislative ends, but also of legislative objectives. Governments purportedly seeking to enhance political stability or cohesion will often be engaged in attempts to justify actions that diminish their accountability to the electorate. More work is needed that engages with the doctrine applicable under section 1.

There is also a set of considerations that exist prior to the doctrinal specifics of section 1 analysis that has not been explored in the literature. Institutions such as electoral boundary commissions, Elections Canada, and now citizens’ assemblies on electoral reform have been tasked with oversight of the democratic process to varying degrees.[175] One understanding of these institutions is that they prevent breakdowns in the democratic process on redistricting, election administration, and choice of electoral system, respectively, by taking decisions out of the hands of self-interested legislators. These non-partisan, independent institutions offer the promise of creating election laws that are free from partisan, incumbent protection or interest entrenchment considerations. With a decreased likelihood of breakdowns, courts should perhaps show more deference to the decisions of these electoral management institutions than they would to legislatures, as long as there are no violations of administrative or procedural fairness in their decision making. For instance, in reviewing redistricting, a court could apply something like strict scrutiny to legislation such as the Electoral Boundaries Readjustment Act, as it is determined by potentially self-interested legislators, but show more deference to the electoral maps produced by boundary commissions. A structural approach to judicial review in Canada should be further explored to develop these themes.