The Civil Rights Case for Voting Reform & Evaluating Key Options
[download PDF from ERRE website – note that figures referenced in the text below are found in the PDF]
Canada’s Charter of Rights and Freedoms articulates the key values that govern our nation. Two central tenets of the Charter are our rights to effective representation and to equal treatment under the law. Our current Single Member Plurality (SMP) voting system violates these principles by effectively disregarding half or more of the votes cast and denying effective representation to these voters. Since numerous other voting systems exist that ensure that the vast majority of voters can be represented by an MP of their own choosing, and since the academic evidence is clear that adopting an inclusive voting system involves “no trade-off … between governing effectiveness and high-quality democracy” (Lijphart), there is no reason to continue excluding so many voters; it is well past time to “Make Every Vote Count”.
As to which voting system will best suit Canada, we recommend that any system adopted:
- enable voters to choose specific candidates,
- maximize the number of voters who vote see their choices reflected in Parliament,
- encourage parties to put forward more diverse slates of candidates, and
- preserve our current proportionate distribution of MPs from different regions.
Fair Voting BC continues to strongly support the recommendation of our BC Citizens’ Assembly on Electoral Reform – the Single Transferable Vote. Amongst the Mixed Member Proportional models, we most strongly support a candidate-centred version in which voters determine the order of election of list candidates. Finally, we also strongly recommend the newer Rural-Urban Proportional Representation model that features multimember ridings wherever these are feasible, but can accommodate single-member ridings in the more remote parts of the country. With a small top-up layer, RU-PR is a highly flexible and responsive voting system that can be tuned over time through the public and non-partisan Electoral Boundaries Commission process to remain well-matched to voters’ concerns as these evolve in the future.
The Civil Rights Case for Voting Reform
Problems With Our Current Voting System: There are many reasons why voters are frustrated and upset with how politics works in Canada, and many of these can rightly be attributed to how our voting system works. As many other submissions will attest, our Single Member Plurality voting system (colloquially known as First-Past-the-Post, though there is in fact no post) leaves upwards of half the voters without an MP they have voted for, which has the practical effect of denying these voters a voice in parliament. This creates significant distortions in the makeup of Parliament, including significant under-representation of some political perspectives in many regions (eg, low to non-existent representation of Liberal supporters across the prairies, Conservative supporters in many cities, NDP supporters in Atlantic Canada, Bloc Quebecois supporters in Montreal, and Green supporters almost everywhere), over-representation of regionally-concentrated voices (eg, the former Reform Party), suppression of important and valid minority perspectives, especially at the national level (eg, supporters of the NDP and Green parties), monopoly control with plurality support (the majority of our parliaments), and a parliament that under-represents women and most ethnocultural groups.
Our National Aspirations: These problems created by our current voting system are not consonant with our highest national aspirations. Our Charter of Rights and Freedoms asserts that we are a free and democratic society. Equal treatment of all citizens is the foundation of true democracy, and we have an obligation to continually strive to be more inclusive in our core democratic practices.
The current government has officially recognized that not all votes count equally. It has promised to “Make Every Vote Count” and the Minister has stated that “We require an electoral system that provides a stronger link between the democratic will of Canadians and election results”. This would nearly double the number of voters represented in Parliament – a step as comparatively large as when women first won the vote.
Charter Requirement for Effective Representation: A strong argument can be made that our Charter requires this form of equality. From the beginnings of our nation, the purpose of our electoral system was (in the words of Sir John A. Macdonald) to ensure that “different interests, classes and localities should be fairly represented”.
More recently, Chief Justice Beverley McLachlin has reminded us in her ruling in the Saskatchewan Provincial Electoral Boundaries reference case that “Ours is a representative democracy. Each citizen is entitled to be represented in government. Representation comprehends the idea of having a voice in the deliberations of government.” She further clarified that the purpose of our Charter right to vote “cannot be less than to guarantee to citizens their full democratic rights in the government of the country.”
In interpreting our Charter rights, McLachlin explains that “the Canadian tradition [is] one of evolutionary democracy moving in uneven steps toward the goal of universal suffrage and more effective representation.” There is no doubt that some of our previous electoral practices, hallowed as they may have been at the time (eg, restriction of the vote to property-holding males, exclusion of women, of aboriginals, of Asians, of teachers, of civil servants), would not be considered compliant with our Charter were someone to be so foolish as to argue that they should be reintroduced.
In our view, excluding half the voters from representation in Parliament, when there are obviously numerous electoral systems that could represent virtually everyone, cannot be considered compatible with the Charter. And as McLachlin also reminded us, “inequities in our voting system are [not] to be accepted merely because they have historical precedent. … Departures from the Canadian ideal of effective representation may exist. Where they do, they will be found to violate s. 3 of the Charter.” Our current voting system clearly departs from this Canadian ideal.
The Nature of Parliamentary Representation: Some object that voters who did not vote for their local MP are not unrepresented, arguing that most MPs offer exemplary service to all constituents. But we are not speaking here of non-partisan constituency service, no matter how conscientiously performed. The principal role of an MP in a Westminster Parliament is (as originally articulated by John Stuart Mills) “to consider, refine and pass legislation”, and the vast majority of voters cast their votes primarily based on a candidate’s party affiliation and policy positions, not on their reputation for constituency service.
When MPs cast legislative votes, they almost always do so along party lines, so voters who did not vote for the MP will typically disagree with how they voted. Indeed, a federal court in New Brunswick ruled in 2004 that “a minority group [is right to] fear that it will not be adequately represented by its member of Parliament … the reality … is that an elected representative who is faced with the conflicting interests of the majority and a minority will often have to choose to represent the interests of the majority”. The court stated that the Supreme Court has rejected the claim that “any individual [ie, an MP] is capable of sympathetic reasoning in relation to other individuals [ie, voters who do not support that MP] … [and] can be sensitive to, concerned about, and capable of responding to the needs of other persons within this society” and concluded that “citizens with distinct interests [are entitled to] an effective voice in the legislative process.”
Claiming that an MP can represent someone who does not support that MP reflects a deep misunderstanding of the nature of political representation. As political theorist Hanna Pitkin wrote in her influential book, “The Concept of Representation” (1967), “For representatives to be authentically democratic,
- they must be authorized to act;
- they must act in a way that promotes the interests of the represented; and
- people must have the means to hold their representatives accountable for their actions.”
From the perspective of a voter who did not vote for and does not support their MP, that voter has not authorized the person to act, the voter feels the MP is acting against their interests, and, given our voting system, the voter has no practical way to hold the MP accountable for their actions, since the voter has already cast a vote against them and found that it had no effect. Accountability of an MP to a voter therefore fundamentally requires that a voter have voted for that MP in the first place so that the removal of that support in future can have an effect. From Pitkin’s perspective, half the voters are denied authentic political representation and, as US political scientist David Plotke wrote in “Representation is Democracy”, “the opposite of representation is exclusion”.
Therefore, the primary purpose of adopting a new voting system should be to move beyond merely having a right to vote (where one’s vote as often as not is not reflected in the makeup of Parliament) to ensure that all those groups in our society that are currently effectively disenfranchised become fully and properly included in our democratic processes.
Rural-Urban Proportional Representation In BC
We will not repeat material on more well-known voting systems (STV and MMP) that is found in Fair Vote Canada’s submission, but we do wish to call attention to and express support for the recently-proposed Rural-Urban Proportional Representation (RU-PR) model that we co-developed with FVC.
- corrects the disproportionality inherent in the single-member seats of Kingsley’s mixed model
- requires far fewer top-up seats than MMP (~12-15% of all seats, vs about ~35-40%)
- allows for the smallest possible single-member (rural) ridings if that is considered an desireable feature – only ~15% larger than our current ridings
- uses a simple one-part ballot (similar to our current ballot in single-member ridings and to an STV ballot in multimember ridings), and can work well either with ranked or ‘mark an X’ ballots; see Appendix 1
- is a highly flexible voting system that can give future Electoral Boundaries Commissions discretion to tune the balance between multimember and single member ridings and to choose the size of top-up regions according to local voters’ desires and preferences
In summary, we believe that the RU-PR model combines the best features of STV (strong voter choice and excellent baseline proportionality in multimember districts) with two well-liked features of MMP (single member rural ridings and a (smaller) top-up layer), and we recommend its serious consideration.
Connecting Our Values to Choices
In this section, we evaluate five systems (STV, MMP, RU-PR/STV+, AV and SMP) on a variety of measures; where useful, we report separately on the STV+ variant of RU-PR (we use RU-PR alone to refer to a variant in which the number of single member ridings is relatively large (~25% of all seats)).
Representation and Proportionality:
The three measures described below are illustrated in graphs (Appendix 2); we use colours (green, orange and red) to indicate good, moderate and poor performance, respectively. Estimated ranges of variables are shown as lighter shades. Details supporting the estimates can be found at fairvotingbc.com and election-modelling.ca.
Representation by a Specific MP for Whom a Voter Has Voted: In a representative democracy, we would argue that the most important metric to use when assessing different voting systems is how many voters are able to elect an MP they explicitly support – this is a candidate-centred metric that is not commonly reported, but which could be considered to be a measure of ‘direct representation’. We consider values above 80% to be very good to excellent; our current voting system has a value of ~50% (poor). STV/STV+ (and to a lesser extent RU-PR) are excellent on this metric; MMP scores moderately because a significant fraction of voters are represented indirectly through a candidate of the same party as the one they named on their ballot.
Representation by a Voter’s Preferred Party: We can also consider the proportion of voters who help elect an MP from their preferred party, even if a voter doesn’t specifically name an elected MP on their ballot; we could consider such votes contributing to ‘indirect representation’, and the sum of direct and indirect representation produces this metric of representation by one’s preferred party. All three proportional systems score well on this metric. MMP, in particular, does better than on the previous metric because a vote for a specific candidate on an open list can be transferred to help elect a different candidate from the same party.
Party Disproportionality Index: The most commonly cited metric represents the discrepancy between a party’s vote share and seat share at the national level (though such a metric can conceal large regional variations). Values of less than 5% are normally considered excellent, and all three proportional systems are largely below this threshold, with RU-PR/STV+ scoring the best overall.
Local Representation & Threshold for Representation
The following table compares various measures of the geographic coverage provided by several different voting systems, along with the threshold for election.
|System||Smallest Riding Size (Relative to Current)||Typical Riding Size||Top-Up Region Size||Threshold for Election (% of Vote)|
|RU-PR/STV+||1.1X – 1.2X||3-4X||7-20X (STV+),
|MMP||1.7X (single member)||1.7X||8X – 14X||~4-6%|
|STV||2.0X (two member)||~4-5X||n/a||~10-15%|
The RU-PR/STV+ model offers the smallest riding size amongst the proportional models; MMP and STV require somewhat larger minimum riding sizes, though a two-seat STV riding would have two MPs. Minimum riding size is sometimes considered important in designing a proportional voting system that addresses logistical difficulties in serving more remote regions.
The threshold for election is typically somewhat less than the vote share required to guarantee election (~1/N, where N is the number of seats in the riding for STV or for the region for the other systems). The RU-PR model, with a region size of 20, offers the smallest threshold for representation (~3%), while STV without a top-up layer offers the largest at about 10%. Green Party supporters can elect MPs in several parts of the country with RU-PR/STV+ and MMP-14 (regions of 14 seats, typically 8-9 constituency and 5-6 top-up seats), but tend to be below threshold in many places with STV and MMP-8 (regions of 8 seats – 5 constituency and 3 top-up seats).
In some countries (e.g., New Zealand), a threshold of ~5% (regions of ~10 seats) is considered to represent an appropriate tradeoff between being open to minority perspectives while avoiding micro-parties.
The voting systems described above all have easy-to-use ballot designs (see Appendix 1). The ballot used in STV+ and RU-PR elections would be identical to that used for STV – it would list one or more candidates per party, organized by party (or as independents), typically with ~10-12 names listed for a 4-5 seat riding. Voters could either rank the choices or mark a single X.
The ballot used in MMP elections would normally be a two-part ballot, with the first part identical to our existing ballots and the second regional part listing all the candidates running in the constituencies across the region (possibly with additional region-only candidates). There would typically be about 25 names on the regional portion of the ballot in an 8-member region, and about 40 names in a 14-member region.
Impact on Electing Women and People from Other Under-Represented Groups: A voting system can improve representation of women and other under-represented groups in two key ways:
- Improving the electoral success of parties that typically nominate more such candidates
- Using multimember districts and explicitly showing the parties’ slates on ballots for these districts
All PR systems address the first mechanism. STV and related systems explicitly encourage multimember nominations, which are directly reflected on the ballot. With MMP, the second part of the ballot in an open list system reveals the collective results of the constituency nominations – if parties have not succeeded in nominating a diverse set of candidates through some coordination of the nomination contests, it will be apparent in the regional list.
- The need for voting reform should be seen primarily as a civil rights issue aimed at ensuring equal treatment of all voters and delivering effective representation to the greatest extent possible
- The Committee should recommend that Parliament adopt some form of proportional voting system
- The Rural-Urban Proportional Representation model should be seriously considered due to its flexibility in addressing the wide variety of local circumstances across the country and its ability to be fine-tuned on an ongoing basis through the Electoral Boundary Commission process
- All three proportional voting systems discussed (STV, RU-PR/STV+ and MMP) give almost all voters an MP from their preferred party and provide a close match between national vote share and seat share by party. STV and STV+ (and to a slightly lesser extent, RU-PR) maximize the number of voters who have voted for specific candidates and should be seriously considered if the priority is to strengthen the bond between voters and individual MPs