The motion creating the Special Committee on Electoral Reform
mandated it to “identify and conduct a study of viable alternate voting systems
to replace the first-past-the-post system,” our current voting system, using
the five principles set out in the motion as a guide: effectiveness and
legitimacy; engagement; accessibility and inclusiveness; integrity; and local
representation.[139]
Two fundamental questions are at the core of the Committee’s study
of alternative electoral system options:
- “Why reform the current system?”, or, what is the
problem that the Committee is “trying to address in which a different change
might actually work?”
- And second, “What might be the consequences if we adopt one system or
another? What would it do?”[140]
The Committee heard wide ranging testimony and received numerous
submissions on the merits and potential drawbacks of the current voting system
as well as various possible alternative voting system options. Underlying the
two fundamental questions at the core of the Committee’s study of electoral
system reform – why change, and what change would do – are the values and
principles set out in the Committee’s motion.
The five principles set out in the motion creating the Committee
are elaborated
as follows:
- 1) Effectiveness and legitimacy: that the proposed measure would increase
public confidence among Canadians that their democratic will, as expressed by
their votes, will be fairly translated and that the proposed measure reduces
distortion and strengthens the link between voter intention and the election of
representatives;
- 2) Engagement: that the proposed measure would encourage voting and
participation in the democratic process, foster greater civility and
collaboration in politics, enhance social cohesion and offer opportunities for
inclusion of underrepresented groups in the political process;
- 3) Accessibility and inclusiveness: that the proposed measure would avoid
undue complexity in the voting process, while respecting the other principles,
and that it would support access by all eligible voters regardless of physical
or social condition;
- 4) Integrity: that the proposed measure can be implemented while
safeguarding public trust in the election process, by ensuring reliable and
verifiable results obtained through an effective and objective process that is
secure and preserves vote secrecy for individual Canadians;
- 5) Local representation: that the proposed measure would ensure
accountability and recognize the value that Canadians attach to community, to
Members of Parliament understanding local conditions and advancing local needs
at the national level, and to having access to Members of Parliament to
facilitate resolution of their concerns and participation in the democratic
process.[141]
Of these principles, numbers one, three, and five speak most to
facets of an electoral system, or goals that an electoral system should embody.
The first principle, “effectiveness and legitimacy” is framed in
terms of how votes are translated into seats in the House of Commons, and
indicates that a proposed electoral system should ensure that votes are “fairly
translated” into seats in a way that “reduces distortion” and “strengthens the
link between voter intention and the election of representatives.” As explained
by Professor Byron Weber Becker, “Distortion is introduced when representation
in government is significantly different from the level of popular support
expressed in the election.”[142] Indeed,
electoral system reform has been a subject of interest at the federal level for
almost a century, since the time when federal elections were contested by more
than two political parties, and has been studied numerous times at the federal
and provincial levels.[143]
At the root of most of these studies is the notion of the legitimacy of the
current electoral system’s method of translating votes into seats – the link
between voter intention and representation in the legislature. In response to
the overall query of “why reform the current system,” two subsidiary questions
necessarily follow:
-
First, when more than two candidates run for election in a
particular riding, should the victor have to obtain a majority (more than 50%)
of votes cast, or just a plurality (more votes than any other candidate, but
not necessarily more than 50%) of votes cast, as is currently the case?
-
Second, should the electoral system, which currently treats each
riding as its own contest, and thus does not take into account support for a
political party or interests across ridings or across a region, in some way
seek to translate this aggregate, proportional vote share into representation
in the House of Commons or provincial legislatures?
The third principle speaks both to the elements of a system, in
that it should “avoid undue complexity in the voting process,” which could be
both in terms of how voters cast their ballots and what calculations are used
to translate votes into seats, as well as to the inclusiveness of the voting
process, in that a system (or voting overall) should “support access by all
eligible voters regardless of physical or social condition.”
Finally, the fifth principle emphasizes local representation and
accountability as fundamental. Indeed, in a country as geographically diverse
as Canada, the relationship between local representation and how votes are
translated into seats is already rather complex. Our parliamentary system is
based on representation by population (that ridings should basically contain
about the same amount of voters), though there is huge variance in that regard.
For example, the most populous riding in Canada is Brantford–Brant
in Ontario, with a population of over 132,000 (including over 95,000 eligible
voters).
By contrast, Nunavut is both the least populous (it has a population of just
under 32,000) and the largest riding in Canada, at over 1,750,000 km2.
By contrast, the riding of Papineau, Quebec is only 9 square kilometres in size
(with a population of approximately 110,000 individuals). As noted by one
witness in Whitehorse, the three northern territories are “overrepresented
population-wise but very under-represented when geographical area is
concerned.”[144]
By contrast, it appears that the values set out in the second and
fourth principles should apply regardless of electoral system, though the
choice of electoral system may have an impact on how easy or difficult it is to
implement the principles (as certain systems are associated with certain
values). The second principle, “engagement,” calls on the Committee to identify
measures that “encourage voting and participation in the democratic process,
foster greater civility and collaboration in politics, enhance social cohesion
and offer opportunities for inclusion of underrepresented groups in the
political process.”
As discussed particularly in Chapter 9 “Voter Engagement and Participation”,
increasing involvement in the greater political process is a goal shared by all
members of the Committee. The Committee recognizes that fulfilling the
objectives of this principle requires ongoing work and commitment. The fourth
principle “integrity,” focuses on the reliability and verifiability of the
electoral system, and the preservation of the secrecy of the vote. Public trust
in the electoral process is high in Canada, and electoral reform should not
undermine it.[145]
Numerous witnesses observed that different electoral systems
emphasize different principles set out in the Committee’s motion. As noted by
Pippa Norris, “you can't get all of these values in any one particular option.
They are all trade-off values.”[146]
Essentially, as posited by Professor Thomas Axworthy, the choice of an
electoral system for Canada is about determining what values ought to be
emphasized, how, and to what extent:
… [T]here is no perfect electoral system. There
are advantages and disadvantages to all of them, and it is really a question of
values, of differing perspectives, that will inform your own debate. There's no
technical solution to the issue of electoral reform. It is basically a
political process of deciding your purposes and values and what you value most.[147]
In other words, as observed by Amanda Bittner, “[a]ll systems have
trade-offs … and at the root of each is a normative idea about how politics
should be.”[148]
Another witness, Richard Kidd, expressed it thusly:
No system is perfect. If we could find a perfect
system, every country in the world would be using it right now. All systems
have their pluses and their minuses, and the big challenge that's facing you is
to try to figure out a system where the pluses outweigh the minuses, or they do
the things that you want them to do.[149]
The fact that there is no perfect system is not a cause for alarm
or inaction. Jonathan Rose suggested:
While … others have argued that there is no one
perfect system, I want to quote Richard Katz, who argued that there is a
perfect system. He argued that the best electoral system, depends on “who you
are, where you are, and where you want to go.”[150]
Indeed, the principles provided for in the Committee’s mandate have
proven most useful to demonstrate how no one system incorporates them all, and
to focus the Committee on evaluating how the principles should interrelate with
each other to create a more robust electoral system for Canada. As witness Matt
Risser suggested, while no electoral system may be perfect, some systems may be
more in accord with expressed values and principles than others:
I did want to make one point, though, in reference
to a point you made about there being no perfect system, because this is
something the committee says a lot. I just want to say that just because
there's no perfect system, which there isn't, obviously, it doesn't mean that
some systems aren't better than others.[151]
In order to evaluate the relative strengths and weaknesses of
various electoral systems, what follows is an overview of electoral system
families, and the key principles and elements that distinguish them from each
other.
At its most basic, an electoral system sets out how votes get
translated into seats in a legislature. In majoritarian electoral systems, the
winning candidate is the individual who garners the most votes in an electoral
district. A political party’s seat share depends on the individual district
results across the country; a party has as many seats in the legislature as
candidates elected. By contrast, proportional systems seek to match a political
party’s vote share with its seat allocation in the legislature. A system is
referred to as “mixed” (the most discussed is mixed member proportional
representation (MMP) when it combines elements of both majoritarian and
proportional systems in terms of how votes are translated into seats. The
objective of mixed systems is to achieve more proportionality by using
compensatory seats to reflect a party’s overall vote share while at the same
time maintaining local ridings with single representatives.
Some important characteristics that differentiate electoral systems
from one another are the following:[152]
-
Number of candidates per constituency (district magnitude): In
each constituency, is one candidate elected or are multiple candidates elected?
The number of candidates per constituency, also called “district magnitude,” is
what distinguishes majoritarian systems from proportional systems. Majoritarian
systems (“a family of systems that includes first past the post, the Australian
alternative vote system, and the French two-round system”[153])
have a district magnitude of one, “which means there is one winner per
district.” Proportional systems, on the other hand, elect multiple candidates per
constituency (in mixed systems, the compensatory seats are based on multiple
member regions). Indeed, “If you have one winner per district, you cannot be
proportional, because the winner takes all. You cannot divide
the one seat among multiple contenders, so it is not a proportional system.”[154]
In other words, if there is only one seat per constituency, it is not possible
for that one seat to reflect the share of votes cast for different
parties/candidates.
-
Ballot: Does the elector place a mark beside a single
candidate’s name (making more than one mark spoils the ballot), or must the
elector rank or order candidates or parties from a list on the ballot? It is
important to note that ranked ballots are a tool that can be used either in
majoritarian (such as the Alternative Vote (AV)) or proportional electoral
systems (such as the Single Transferable Vote (STV), in open lists in an MMP
system, etc.)
-
Threshold for determining winners: What is the percentage
of votes needed for a candidate or party to obtain a seat? For example, in
single-member constituencies is a plurality of votes sufficient (more than any
other candidate, but not necessarily more than 50%), or is a majority (more
than 50%) required? In multiple member constituencies, what is the minimum threshold
or quota to be elected (for example in a constituency with three members, a
threshold of 33% would guarantee a seat)?
-
Procedure to determine winners: How many steps are there,
and therefore how simple or complex is it to determine how many seats each
party has won and which candidate has won which seat? For example, in
proportional systems, a first step is to determine a political party’s overall
vote share (within the country or a region of the country) and then the second
step is to allocate seats based on the vote share (which could be determined
through a set list of candidates, or based on individual votes for specific
candidates). In majoritarian systems using ranked ballots, what is the process
used to determine which candidate wins?
Discussion of these characteristics, how they relate to the
principles set out in the Committee’s mandate, and how they can be applied to
create different electoral systems permeated the testimony heard by the
Committee.
Canada’s single-member plurality electoral system, commonly called first-past-the-post
(FPTP), the winning candidate is the individual who garners the most votes
(though not necessarily a majority) in an electoral district. A political party
has as many seats in the legislature as it has candidates elected. In other
words, its seat share in the legislature is the result, the sum total, of the
individual contests that take place across the country. With regard to forming
government, the leader of the party that secures the largest number of seats in
the House, and can therefore hold its confidence, is generally invited by the
Governor General to be the prime minister and form government.[155]
While the focus of the
Committee’s study was on alternatives to replace FPTP, the following attributes
were noted as being the perceived strengths of Canada’s FPTP system:
Thomas Axworthy, in observing that there is no crisis in Canadian
democracy, noted how strongly Canada’s government and electoral system compares
internationally:
When we look at the various assessments
internationally, we see that the World Bank, for example, which sponsors a
worldwide governance indicator project, indicated that in 2014 Canada had
ratings of 96% in accountability, 91% in political stability, 95% in government
effectiveness, 98% in regulatory policy, 95% in the rule of law, and 94% in the
control of corruption. That's absolutely in the top ten of attainment.
Professor Norris's own electoral integrity project
had Canada again as probably—and she can correct me on this—at the top of the
majoritarian practitioners of electoral systems, with a rating of around 75% to
80%, ahead of the United States and so on. Again, it was in that absolute top
rank.
This international
assessment about the value of Canadian government practice and electoral
practice has led, as we all know, famously to the human development index of
the United Nations, where Canada has always been in the top 10 and sometimes
has been number one. I think in 2014 we were number nine.
The strength of our
government system and our electoral system has certainly had a positive impact
on those achievements in the human development index. That is because—pride of
position here—the Westminster system, with its combination of a concentration of
power to get things done and an accountability related back to what David
Smith, the brilliant scholar from Saskatchewan, calls “the people's House of
Commons”—that combination of [the]people’s sovereignty as represented in the
House and the concentration of power for effective government—is really the
secret of the Westminster system when it is working correctly. For most of our
history, it has been working correctly in Canada.[156]
In examining options for reform, former Chief Electoral Officer
Jean-Pierre Kingsley noted “the relative simplicity of the system or the ballot
that we would replace, if we replace the present system” and observed that
“nothing will be viewed as being as simple as the present system, because we've
been at it for 149 years.… This is part of the DNA of being Canadian and being
born Canadian.”[157]
Currently Chief Electoral Officer Marc Mayrand noted how the current system is
“relatively simple and easy to understand.”[158]
As well, the ballot counting process is relatively straightforward.
Mr. Mayrand added that changing systems “may make it difficult to publish the
outcome of the election on election night by completing the counting of ballots
manually at voting sites, as is currently the case.” He added that “Canadians
are accustomed to learning the results of elections quickly, and any possible
delay should be considered carefully by the Committee.”[159]
A number of open mic participants expressed their view that the
current system is simple, works well, has held up for 150 years, and should be
maintained. In the words of one participant, “first past the post has served us
well for nearly 150 years. It is simple, and it is easy to understand. Please
don’t change it.”[160]
Another participant observed, “if the system isn’t broken, don’t fix it.”[161]
The importance of local representation was raised numerous times as
a key value for various witnesses. Former Chief Electoral Officer Jean-Pierre
Kingsley referred to
“the rapport, the link, between the elector and the elected, both for the
representation
of the electors, collectively and individually, and for the accountability of
the elected representatives.” He added that “Canadians are well accustomed to
that rapport, that link. It has to be weighed very carefully if there's going
to be any change.”[162]
The importance of local representation was also reflected in
responses to the Committee’s online consultation, where 72.5% of respondents
either strongly agreed or agreed with the statement that “Canada’s electoral
system should ensure that voters elect local candidates to represent them in
Parliament.”[163]
FPTP is more likely to produce majority governments.[164]
As noted by political scientist Christian Dufour, FPTP “delivers the most
important elements. It delivers governments that are strong, but that can also
be voted out, and that's not nothing. In the context of globalization, which is
dangerous, the powerlessness of democracies is something to be avoided. Our
system ensures that governments often enjoy a majority.”[165]
This point was echoed by Kenneth Dewar in his submission to the Committee. He
stated,
The “single member plurality” system has
historically provided Canada effective government, resulting often in strong
majority governments (usually based on a plurality of votes) and occasionally
in effective minority governments.[166]
In addition, Kenneth Carty observed how Canadians are used to
majority governments, and change would require adaptation.
I think it's fair to predict that under most other
electoral systems, majority governments of the sort that Canadians have been
generally most used to will disappear. We have so little experience with
multi-party governments in this country that we don't have any clear idea how
they'll work either in the short term or in the longer term or what they will
look like.[167]
Finally, related to the notion of being able to produce majority
governments is the understanding that the FPTP system is also conducive to
being able to vote governments out once they are no longer in favour, what Bryan
Schwartz called “alternation”:
I'm in favour of alternation. I like the idea of
different people, different voices having a turn. I like the idea of policies
being evaluated and given a fresh thinking. I like the idea that one team of
patronage seekers doesn't always win. I like the idea that people who disagree
get a turn in office, and they can live with the problem.[168]
As elaborated below, the primary critique of FPTP is that in
ridings with more than two candidates, and electoral contests with more than
two political parties, FPTP fails to accurately represent the will of voters,
both at the riding level and in terms of overall vote share in the House of
Commons. A further critique is that by failing to accurately represent the will
of voters, FPTP may discourage people from voting, possibly leading to voter
apathy and dissatisfaction with the system, and resulting in lower voter
turnout. Finally, a number of witnesses suggested that the nomination process
in individual ridings,
where some ridings are considered “safe,” could “contribute to the challenges
faced by women, Aboriginal people and minority groups in being nominated as
candidates and elected as members of the House of Commons.”[169]
Roderick Wood, law professor, who had served as a commissioner with the Law
Commission of Canada when it published its report on electoral reform in 2004,
summed up the various perceived shortcomings of FPTP as follows:
… the existing system results in
disproportionality, the creation of artificial majorities, regional imbalances,
and what the Jenkins commission referred to as the creation of electoral
deserts, in which whole regions of Canada may have little or no representation
in the [G]overnment. It results in the under-representation of women,
minorities, and first nations peoples. It gives a sense of the lost vote—“Why
should I vote? It's not going to be counted. It's not worth anything”—and may
even lead to strategic voting, the feeling that you have to vote for a less
preferred candidate because otherwise your vote simply wouldn't count. It can
also lead to what is viewed as a hyper-partisan adversarial political culture
in the country.[170]
From Confederation in 1867 until 1921, federal elections were
contests between two political parties, and it was understood that in this
environment Canada’s FPTP electoral system worked fairly well in terms of
translating votes into seats:[171]
As long as federal elections were contested by
just two political parties, the first-past-the-post system produced parliaments
in which there was a pretty good match between the distribution of seats in the
House of Commons and the popular vote for political parties.
The majority governments that these parliaments supported on all but one
occasion were led by leaders whose party members won a majority of seats in the
House and whose candidates won over 50% of the popular vote.[172]
However, towards the end of the First World War new political
parties entered the fray, and in the December 1921 federal election, three
political parties elected members to the House of Commons. Since then all
federal elections have been contested by three or more political parties.
This shift in political reality, with multiple parties contesting
seats in the House of Commons, began to raise questions about the legitimacy of
the FPTP electoral system whereby candidates began to be elected with a
minority of the votes cast in a riding.
As observed by the Special Committee on Proportional Representation (PR) and
the Subject of the Single Transferable or Preferential Vote in its 1921 report:
It must be apparent to all that the present system
of election in single member constituencies meets fully the purpose intended
only when not more than two candidates are nominated. Recent experiences in
elections in Canada have brought home to our people the fact that where three
or more candidates present themselves in single member constituencies, the
candidate declared elected may, and often does, represent merely a minority of
those voting in the constituency.[173]
This question of the ongoing appropriateness of FPTP in an
environment with more than two political parties vying for power, and more than
two candidates running in any riding, was reiterated by Peter Russell, who indicated
that “The FPTP system, which Canada has had at the federal level from
Confederation until today, no longer fits the political circumstances of the
country and has not done so since 1921.”[174] He
added that since 1921 “we've had a multi-party system, mostly four or five
parties, and that kind of party system is really torpedoed, undermined, by the
first past the post system.”[175]
Jean-Pierre Derriennic, in his book titled A Better Electoral
System for Canada (which formed the basis of his testimony and submission
to the Committee), described the distortions caused by FPTP when multiple
parties and candidates run for election:
In Canada, three, four, or five parties can get
MPs. We are not in a two-party system, and the plurality voting system has much
more detrimental effects, as may be seen in the outcomes of the last election
and the previous ones.
With more than two big parties, the plurality
voting system becomes a machine for making parliamentary majorities out of
electoral minorities. On October 19, 2015, the Liberal Party received 54% of
the MPs and 100% of the decision-making power in the House of Commons. It got
39.5% of the popular vote, but much of that vote went to defeated Liberal
candidates. Therefore, probably 39.5% of all Canadians are rather happy with
the Liberal Party’s victory, but only 26.1% voted for one of the 184 elected
Liberal candidates. Given a 31.5% abstention rate, these Liberal MPs have 100%
of the legislative power and support a government that has 100% of the
executive power, despite being elected by only 17.9% of all Canadians of voting
age.[176]
Deciding whether or not FPTP continues to be legitimate is a
question of values and principles, rather than “empirical” facts, as argued by
Emmett Macfarlane, who stated that “[t]he implication is that the first-past-the-post
system is, obviously, illegitimate because it's producing undemocratic results,
but that's a value proposition, not an empirical statement.”[177]
Finally, respondents to the Committee’s online consultation overall
expressed lukewarm support for FPTP. In response to the statement “Seats in the House of Commons should be filled by the candidates who
receive the most votes in their ridings, even if they receive less than 50% of
the total votes cast,” a majority disagreed (51.7%), while just over a third
agreed (34.5%):[178]
Seats in the House of Commons should be filled by the
candidates who receive the most votes
Scale: 1(Strongly Disagree) – 5 (Strongly Agree);
NA
A number of witnesses identified a variety of secondary problems
associated with how FPTP translates votes into seats, and seats into
government.
Brian Tanguay[179]
and Craig Scott (former MP)[180]
commented on how FPTP can result in “artificial” or “false” parliamentary
majorities that can have a negative impact on Parliament and on governance (in
terms of the power given to the executive branch).
Nathalie Des Rosiers, who
served as the President of the Law Commission of Canada during the period of
its study of electoral reform, opined that the preference for stability under
FPTP was not sufficient to overcome the system’s lack of representativeness:
The preferences of some systems come at too high a
price. That was our conclusion regarding the first-past-the-post system. Its
preference for stability was too costly, as it deprived us of a more adequate
representativeness in terms of ideas and people.[181]
As well, Lise Ouellette, who served as co-chair of the New
Brunswick Commission on Legislative Democracy between 2003 and 2004, observed,
“Here we are 12 years later, and I think that change is even more necessary
federally than it was, or than at the provincial level” due to the “risk of an
unrepresentative federal government or of an unrepresentative Parliament, be it
geographically, ideologically or demographically.”[182]
Eric Maskin, Nobel laureate, set out five problems with FPTP in his
appearance before the Committee, beginning with the lack of a majority of votes
for a candidate in a riding and culminating in a sense of voters feeling
disenfranchised, that their vote is wasted, and potential candidates being
discouraged from standing for election:
The first problem is
that it's often the case that the MP representing a particular electoral
district is a minority MP, in the sense that most voters in the district didn't
vote for that person.
Second, first past the post often leads to a
serious discrepancy in Parliament, by which I mean that the majority party
often receives much less than a majority of the votes.
For example, in 2011 the Conservative Party had 53.9% of the seats but only
39.6% of the vote. There are many other examples of such discrepancies.
Third, the candidate elected in a district can
often be wrong….
Fourth, a voter is in effect disenfranchised if
she votes for an unpopular candidate, a candidate who is not likely to win the
seat. If candidates A and B are the candidates who have a serious chance of
winning, and I vote for candidate C, then in effect I have no say in the choice
that really matters. I'm wasting my vote. I could vote strategically—that is,
even though I prefer C, I could vote for A or B—but strategic voting itself is
problematic for reasons that perhaps I can come back to in the question period.
Fifth, unpopular candidates or parties may be
discouraged from standing. For example, suppose I'm a candidate on the right
but one who disagrees with the Conservative Party on some important policy
points. I may hesitate to stand for office, because if I do stand, I run the
risk of splitting the vote on the right, and by doing so I may help to elect a
left-wing candidate. For that reason, I may deliberately not stand, and through
that decision I'm not only depriving myself of a political candidacy but I'm
also depriving the electorate of another political voice.[183]
Numerous open mic participants expressed what it felt like for them
to vote strategically rather than for a candidate that they preferred in order
to avoid electing someone else. For example, Mary Cowper-Smith explained:
I have voted in every federal and provincial
election since I was old enough to vote, and almost every time I either felt my
vote was wasted or I felt compelled to vote strategically.
As a voter, I have felt frustrated and cheated.[184]
As well, the perception of votes being “wasted” or “lost” under
FPTP (primarily in comparison with proportional electoral systems) was raised
by various witnesses, in submissions to the Committee, and was noted in
responses to the Committee’s online consultation.[185]
Matt Risser explained the perception of a vote being “wasted” in the following
way:
When we talk about wasted votes, we should really
distinguish between input fairness and output fairness. Canada has input
fairness — nobody disputes that; every vote is counted fairly—but output
fairness is that every vote counts fairly.[186]
Jean-Pierre Derriennic used the following figures from the 2015
federal election to describe his understanding of “lost votes” and the sense of
cynicism that can result:
Less than half the voters elected all of the MPs
of all parties in 2015: 47.6% of all valid votes. Conversely, 52.4% of all
valid votes failed to elect anyone. They were lost votes. Votes for
second-place candidates were not completely useless, being the best means to
prevent election of the winning candidate, but votes for candidates who came
neither first nor second were wasted. And those were 23.2% of all valid votes.
This is clearly a very serious problem. There are
several reasons for low election turnouts and the cynicism that many of our
fellow citizens feel towards democratic institutions. One of them is the
electoral system. Many Canadians feel that their vote is useless, that nobody
represents them in Parliament, and that politicians can get into power with the
support of minorities and then rule while ignoring the needs of large parts of
the population. This feeling has a basis in reality and comes from their
experience as disappointed voters. To correct this feeling and to end this
cynicism every vote should count.[187]
Additionally, a number of witnesses raised the issue of “policy
lurch” occurring under FPTP, as a negative consequence of alternation in
governance between parties with different policy views and approaches. For
example, James Bickerton explained:
[T]here has emerged in Canada a relatively recent
problem, at least in terms of its severity, that has been referred to as
“policy lurch”. I say it is relatively recent, because for decades prior to the
1990s Canadian governance was shaped by centrist, brokerage-style politics that
moderated the policy shifts that are the normal expectation of a change in government.
However, the more ideologically polarized environment that has emerged in
Canada since that time has given rise to concerns about more severe instances
of policy lurch that are evident in other first-past-the-post jurisdictions
with more ideologically polarized party systems. Indeed, it has been cited as
one of the main reasons for New Zealand's decision to change its electoral
system.
To illustrate the problem, the current Trudeau
government has spent much of its first year in office, and will no doubt do the
same for a good part of its second year, undoing many of the changes introduced
by the previous government, at which point in time they will begin taking steps
to prepare the way for the next federal election campaign. Yet a relatively
minor shift in votes of five or six percentage points in that election could
result in a new government that engages in another round of policy lurch,
undoing much of the undoing that this government has been doing. This kind of
roundabout “now it's our turn” policy-making can hardly be thought of as
beneficial for stable long-term governance that is built on a solid foundation
of a reasonably broad societal consensus.[188]
Arend Lijphart commented on the lack of policy coherence that
results from alternation between governments:
For one thing, as we all know, fast decisions are
not necessarily wise decisions. Also, a great deal of coherence in policy is
lost in the alternation between governments of the right and governments of the
left, and then back again to governments of the right. This was the main reason
why the famous British political scientist Samuel Finer, who had been a strong
supporter of FPTP, changed his mind and advocated PR in an influential book
published as early as 1975. Finally, policies supported by a broad consensus
are more likely to be successful and to remain on course than policies made by
a so-called decisive government against the wishes of important sectors of
society.[189]
Furthermore, a consequence of FPTP’s local focus is that it “tends
to favour parties with regional, rather than national, appeal.”[190] And, as noted above, there is a tendency for
FPTP to result in “regional deserts,” where entire regions of the country are
either not represented by the governing party or the opposition. Jean-Pierre
Derriennic described the regional challenges of FPTP as follows:
The plurality voting system usually increases the
gains of whatever party gets the most votes in the country and decreases the
gains of other parties. It does the same in each region, thus making the
differences of opinions and interests between people in different parts of the
country appear greater than they really are. Concern for Canadian unity is not
as serious now as it was in 1993, but the voting system is still exaggerating
antagonisms between regions: in 2015, the Conservative Party got no MPs in the
four Atlantic Provinces despite receiving 19% of the votes; in Alberta, the
Liberal Party and the NDP together received 36% of the votes and only 14% of
the MPs.
This is one of the most dangerous consequences of
our electoral system. In a very large country where people, natural resources,
and climate are diverse, it is unwise to keep a voting system that exacerbates
antagonisms between regions. This is probably the strongest argument for
electoral reform and is well known.[191]
Finally, numerous witnesses associated the lack of women’s,
minority (including racial minorities and Canadians with disabilities) and Indigenous
representation to FPTP. As Brian Tanguay said:
In terms of producing a Parliament that is a
mirror of the nation, the present electoral system does a very poor job indeed.
It poses significantly high barriers to the election of women, minority, and [I]ndigenous candidates … something that
detracts significantly from the effectiveness of this body as a forum for the
generation of new ideas and policies to cope with the challenges posed by this
rapidly changing world.[192]
Pippa Norris suggested that it is “far more difficult for women and
other minorities to get elected on the first past the post in single member
districts than it is under the party list.”[193]
As discussed in Chapter 7, the underrepresentation of women and minorities is
not necessarily due to a particular political system, but rather attributable
to how the nomination process works and how political parties operate. For
example, as noted by Antony Hodgson:
With the way we do nominations in this country, we
end up putting one person forward from each party. On average, that is biased
in favour of the “male, pale, and stale” stereotype.
I am very pleased to see that this is not true here at this table, but
statistically there certainly is that bias. I think young people in particular
are not represented in government as much as they should be.[194]
The challenge with ensuring diversity of candidate representation, as
one witness suggested, is that it is hard to manage the individual nomination
processes in all
338 ridings in Canada.[195] By
contrast, it is more obvious in list-based electoral systems for voters to see
the relative numbers of women and minority candidates.
The five principles set out in the
Committee’s mandate have enabled the Committee to focus on certain electoral
system options to the exclusion of others.
For example, pure List Proportional Representation (List PR)[196]
would not be suitable for Canada because it does not focus on electing local
representatives but rather on political parties (though moderate variants with
provinces divided into regions with small open lists could be considered). As
noted by Brian Tanguay:
[W]hat is not acceptable in Canada? Well, “list
PR” is not acceptable. Also the Israeli system is not acceptable, nor that of
the Netherlands. Anything that simply offers voters a single choice for a party
is not acceptable.[197]
Pure List PR was not an option often raised for consideration by
witnesses and individual participants,[198]
and received limited support in the Committee’s online consultation.[199]
Two-round voting, within the majoritarian electoral system family,
also was not discussed to any extent by witnesses or individual participants[200]
and received limited support from respondents to the Committee’s online
consultation.[201]
This system, used in France, has been associated with significant cost (of
holding two rounds of voting) without much attendant benefit in overall
representation.[202]
The individuals who engaged with the Committee (either through
submissions, testimony, or in the e-consultation) and recommended reform
overwhelmingly favoured the addition of some element of proportionality to
Canada’s electoral system.
The options commonly discussed[203] include
(organized in order from majoritarian to proportional to mixed systems):
- The Alternative Vote and
other ranked ballot variants within the majoritarian system family (note that
most individuals who engaged with the Committee and discussed this option were
against it);
- The Single
Transferable Vote and other candidate-focused multiple member riding proportional
options;
- Mixed Member
Proportional Representation (note that most individuals who favoured reform
expressed support for this system) ; and
- Variations between
the above to account for the significantly different geographical realities
found across the country (such as the use of ranked ballots with a mix of both
single-member and multiple-member constituencies, depending on population
concentration).[204]
Brian Tanguay suggested that if the Committee has “the broad
alternatives available to voters and discuss and debate them fairly and
transparently, that will serve this Committee and the [G]overnment well.”[205]
It is the Committee’s hope that the following pages present the primary
electoral system reform options “fairly and transparently.”
Finally, as electoral system options are set out below, it is
important to keep in mind the impact that reform will have on Canada’s overall
governance ecosystem, as noted by Maryantonett Flumian, President of the
Institute on Governance:
This [is] my fundamental message: whatever
recommendations your deliberations take you to, rest them on our entire
governance ecosystem. People want some change. People want evolution. Our
system has to evolve in order to maintain that primordial connection directly
with citizens, which I think is fundamental to our democratic system of
governance, but understand the whole system.[206]
One electoral system reform option proposed to the Committee would
be to introduce ranked ballots in single-member constituencies. The problem or
issue that this reform would try to solve is that of candidates in ridings
being elected without a majority of support from voters, for example as
expressed by Eric Maskin:
Under the current system, the first-past-the-post
system, there are many, many cases of MPs being elected without absolute
majorities. What's worse is that we don't know, because we aren't finding out
from voters, whether there are other candidates whom a majority would have
preferred. That's why switching to a voting system under which voters can
express themselves more fully is a way to ensure that the right MPs get
elected.[207]
On the ballot, instead of marking an “x” or equivalent, voters
would rank the candidates running in their electoral district in order of their
preference. Results would then be tallied to determine which candidate is
preferred by the majority of voters in the district. The Committee heard
testimony about three possible counting methods that could be used to determine
the result. These variations are described below.
It is important to note that in Australia voters are required to
“complete all boxes on the ballot paper for the vote to be formal and included
in the count.”[208] In
other words, if a voter in Australia does not rank all of the candidates on the
ballot, his or her vote would be considered “informal” and would not be
counted. Should the use of ranked ballots be adopted in Canada (in either
single or multiple member constituencies), it is not recommended that voters be
forced to rank all candidates listed. As was noted in testimony, this
requirement would add complexity to the voting process, would limit voter
choice, and would not significantly alter the results.[209]
As elaborated below, the main arguments put forward in favour of
using ranked ballots in single-member constituencies is that election results
would be more legitimate by more closely reflecting voters’ preferences (in
relation to the first principle set out in the Committee’s mandate); that the
proposed measure would be relatively simple to understand and to implement (as
it would involve no change to riding boundaries and candidates could continue
to seek election as under FPTP); and that it would encourage moderation and
consensus-building (as candidates and political parties would be incentivized
to be voters’ second, if not first, preference).
The fundamental critique of introducing ranked ballots in single-member
constituencies is that the proposed measure would do nothing to correct the
perceived issues with majoritarian electoral systems: namely, that it would not
take into account support for a political party or interests across ridings or
across a region, and in some way seek to translate this aggregate, proportional
vote share into representation in the House of Commons or provincial
legislatures. As well, by favouring moderation and consensus, it was suggested
that the use of ranked ballots in single-member constituencies would
effectively discriminate against smaller parties and minority viewpoints,
resulting in less representational diversity. This, in turn, could actually
increase distortion between voter preferences and outcomes. Finally, it was
argued that moving to ranked ballots while maintaining single-member
constituencies would result in such minor change that it would not be worth the
effort.
The Committee heard of three main variants that could be used to
tally the rankings to determine the winning candidate in an electoral district:
AV, also referred to as Instant Runoff Voting, the Borda count method, and the
Condorcet Method. As described below, different tallying methods may lead to
somewhat different results. It is important to note that any of the following
tallying methods could be applied to the use of ranked ballots in multiple
member constituencies.
The most commonly known ranked ballot system in single-member
constituencies is AV, currently used to elect members of the House of
Representatives in Australia, and previously used at the provincial and
municipal levels (along with STV) in certain areas in Alberta, British
Columbia, and Manitoba.[210]
Essentially AV works as follows: on the ballot, voters rank
the candidates running in their electoral district in order of their
preference. To be elected, a candidate must receive a majority of the eligible
votes cast. Should no candidate garner a majority on the first count, the
candidate with the fewest first-preference votes (lowest-ranked) is dropped,
and the second-preference votes on the ballots where that candidate ranked
first are assigned to the respective remaining candidates. This process
continues until one candidate receives the necessary majority.[211]
In the 18th century, French mathematician, political
scientist, mariner and physicist Jean-Charles Borda devised a preferential
voting system, called the “Borda count” to correct the perceived issues with
the plurality system used to elect members to the French Academy of Sciences.[212]
The Borda count system for ranked ballots assigns points (the number of points
awarded is based on the total number of candidates) “according to its
preference position.” As explained by witness Russ Husum:
The Borda count
method is simple to use, and for the reasons that follow, it gives a more
accurate result than simply dropping people off if they are the lowest
first-preference candidate.
First off, no candidate is dropped. Second, every
preference level of every ballot is used to calculate the total. Third, every
preference on a ballot is given a value according to its preference position.
For example, if you
had six candidates, the first vote would be worth six points to a candidate.
Then the second vote would be worth five points, then four, then three. If
there were eight candidates, the first would be worth eight points and then
seven points, and so on.
Let's say you have five candidates running. A
first preference vote is worth five points to each candidate. Let's say Mary
Smith gets 10,000 first-place votes. She gets five times 10,000. If she gets
5,000 second-place votes, she gets four times 5,000. Those are totalled up for
each candidate, so in the end you get a more accurate total then simply
dropping people off.[213]
Mr. Husum argued that using the
Borda count to tally voters’ ranked ballot preferences would address certain
perceived shortcomings with the AV method of counting:
There are some concerns with the regular ranked
ballot counting, and the Borda count method takes care of them. I'll go over
three of them right now.
Sometimes, when you drop
the first candidate—the lowest candidate with the least first-choice
preferences—you can drop the candidate who is actually most preferred.…
Also, the regular
accounting method can sometimes inadvertently pick a majority winner when in
reality they are not the most preferred candidate.
Finally, one of the
criticisms of the ranked ballot is that second and third preferences that are
reassigned should not be worth as much as the first preference.[214]
One can compare the plurality vote, AV, and the Borda count to how
students are ranked in school. Under our current plurality system, where only
one preference is indicated, candidates are essentially ranked based on who
gets an “A”, while ignoring any lower grades. AV tweaks this by first counting “A”s,
then if necessary looking at “B” grades, and so on. By contrast, the Borda
count operates like a grade point average, whereby the winning candidate is the
one with the highest total number of points.[215]
Of note, a modified Borda count system, applied to multi-member constituencies,
is used in the Republic of Nauru (which became independent from Australia in
1968).[216]
It is a popular counting method for granting sports awards, such as Major
League Baseball’s Most Valuable Player award and US college football’s Heisman
Trophy.
Returning to 18th
century France, the Marquis de Condorcet, disagreed with Borda (his
contemporary) in terms of ranking preferences. He instead proposed an
alternative system whereby a ranked ballot would be used to compare candidates
in head to head matchups. The winning candidate is the one who comes out ahead
in the various paired contests. Eric Maskin from Harvard proposed a Condorcet
system called “Majority Rule.”[217]
Professor Maskin explained “Majority Rule” as follows:
Under Majority Rule, voters now have the
opportunity to do more than just vote for a single candidate: they're allowed
to rank candidates. Candidate A is best, candidate B is second best, and so on.
The winner is the candidate who is preferred by a majority, according to the
rankings, to each opponent. The candidate is the true majority winner. The
candidate would beat each opponent in a head-to-head contest.
I have a slide to
illustrate this. Let's imagine that the electorate divides into three different
groups: 40% of the electorate likes candidate A the best, then B, then C; 35%
put C at the top, then B, then A; and then the remaining 25% like B best, then
C, then A. This is just an example. It's not meant to correspond to any
real-life situation.
What
happens under Majority Rule? Under Majority Rule, candidate B beats A by a
majority because the group in the middle, the 35% group, prefers B to A, and
the group on the right, the 25% group, prefers B to A. That's a majority.
That's 60%.
Candidate B also beats C by a majority because the first group, the 40%
group, prefers B to C, and the third group, the 25% group, prefers B to C.
That's 65%, so B is the true majority winner.
Let's contrast that with
what happens under first past the post. Under first past the post, you just
vote for a single candidate. Presumably the people in the first group will vote
for A, the people in the second group will vote for C, the people in the third
group will vote for B. A is the winner because 40% is the highest vote total,
and so we get the wrong candidate elected. A is elected under first past the
post, but a majority, 60%, prefer B. For that matter, in this example, a
majority also prefers C to A, so A is really quite a terrible choice from the
standpoint of majority will.
Majority Rule solves all
five problems that I described because the winner represents a majority of
voters.[218]
He added that “Under Majority Rule, voters have no incentive to
strategically vote anymore. They have every incentive to vote according to
their true preferences.”[219]
Maskin then described the difference between “Majority Rule” and
AV:
I think the easiest way to see the difference
between alternative voting, which is sometimes called instant runoff voting,
and what I was talking about, Majority Rule, is to use the example that is on
the screen.
As I showed you in that example, candidate B is
the majority winner because B beats A by a majority, and B also defeats C by a
majority.
However, if we use alternative voting, instant
runoff voting, then we'd look only at first-place votes, so 40% vote for A, 35%
vote for C, and 25% vote for B, we notice that B, who is actually the true
majority winner, is eliminated under alternative voting. That's because under
alternative voting, if no candidate gets a majority of first-place votes, you
eliminate the candidate who has the fewest first-place votes, and that's B in
this case.
This example encapsulates the difference between Majority
Rule and Alternative Voting.[220]
While he proposed Majority Rule as being preferable to AV, he
argued that either would be an improvement over FPTP:
I think either alternative voting or majority
rule, or some other similar variant in which voters have the opportunity to
express themselves by ranking rather than just voting for a single candidate,
would be a considerable improvement over the current first-past-the-post
system.[221]
One issue acknowledged by the Marquis de Condorcet himself was the
possibility that no one candidate would emerge as the overall Condorcet winner
in head to head matchups. In that case, which Professor Maskin argued would be
rather remote in the Canadian political context, a tie-breaking mechanism could
be applied.[222]
Finally, Professor Maskin noted that as the Majority Rule voting
method requires computational capacity, it was not a system that could be
considered until recently:
I think an important
reason that until fairly recently Majority Rule, rather than alternative
voting, was not on the table is simply that counting ballots under Majority
Rule was somewhat more complicated. You have to look at all pairwise
comparisons. With modern computers, that's not a problem, but before modern
computers were around, it certainly was.[223]
Some suggested that introducing ranked ballots in single-member
constituencies (no matter the tally method) would increase the perceived
legitimacy of election results by better reflecting the overall preferences of
voters in a riding and requiring candidates to obtain a majority of support in
their riding. As noted by André Blais (in comparison to FPTP):
Basically, the system is not too different from
first past the post, but a party that is the second choice of many would get
more seats. That would be the biggest difference. It's up to you to decide
which is the party that is the second choice in a given context, and then
you'll see which party is most likely to be favoured at a given point in time.
That's the main difference. It's more legitimate,
in the sense that every candidate who is elected gets at least 50% of the vote.
In my view, that's more legitimate. It is still not proportional and so on in
many different aspects, but it is, in my view, more acceptable.[224]
And as observed by Professor Derriennic:
Every MP would be elected by a majority, there
would be less risk of a party winning with less of the popular vote than that
of its main rival. Citizens could vote the way they wish, without fear of
wasting their vote or having to vote strategically. Each party’s real popular
support would be known, and the big parties would want to heed the opinions and
needs of supporters of other parties.[225]
As well, by maintaining local representatives in single-member
constituencies, a change to ranked ballots would be considered a relatively
“innocuous” innovation in comparison to other proposals presented to the
Committee, for example as noted by Royce Koop:
The alternative vote doesn't really affect what I
was talking about with local representation. There would certainly be a local
representative. That would be preserved, so it would be a real plus of the
alternative vote as well.
We would perhaps see that people feel like they
have more input into the choice. Because of the ranking nature of the ballot,
more votes are included in the overall result. We might see increased
democratic satisfaction as a result of that, but beyond that I am not sure.
It wouldn't be a huge change. It is a relatively innocuous change to the
electoral system compared to some of the other alternatives that we are talking
about today.[226]
The Hon. Paul Okalik commented on how introducing ranked ballots
while maintaining single member constituencies would be a clear and simple
innovation:
If it's the desire and the will of the committee
to move forward with a different model, I stress that it should be as simple
and as clear as possible for all concerned. The alternative vote model would be
my preference, as it maintains the clarity and simplicity to the voters and is
in keeping with their wishes.[227]
James T. Arreak, Chief Executive Officer of Nunavut Tunngavik Inc.,
spoke about how introducing ranked ballots would favour consensus, and in that
sense would be consistent with Inuit culture:
One alternative to the first-past-the-post system
is the ranked candidate system, with each elector numbering candidates in order
of preference, and then the votes of candidates with fewer first preferences
being tabulated and redistributed until one candidate is the ranked choice of
at least 50% of the electors. This system has the virtue of overcoming one
defect of the first-past-the-post system: in a first-past-the-post contest, a
person can be elected having extreme positions that may appeal to a minority of
voters that are heartily rejected by a majority. The ranked candidate system
appears to be more in keeping with the premium placed on consensus-building and
the preference for inclusiveness that is characteristic of Inuit culture.[228]
Finally, Joel Howe, also speaking in Fredericton, New Brunswick,
noted how ranked ballots would encourage moderation:
With ranked ballot, for example, you allow for
many parties, but they must each jockey to be voters' second or third choice.
This means they cannot simply pander to their existing limited base if they
want to get elected. This is the incentive toward moderation that a 5% or 10%
threshold under PR can't hope to provide.[229]
The greatest perceived shortcoming of the use of ranked ballots in single
member constituencies is that it is a majoritarian system that can effectively
squeeze out smaller parties while benefitting “big tent” political parties. For
example, as posited by the Hon. Ed Broadbent:
A ranked ballot system can have the effect of
eliminating particularly very small parties.
They can be ranked out of the system. The advantage of either MMP or strict PR
is that every vote will count and you don't need to have a ranking to make it
count.[230]
For example, one participant made the following
statement:
If we adopted a preferential vote system, how
would we make sure that our country did not always elect a centrist party like
the Liberal Party? That is to say, going forward, a party that benefits from
being a second choice for everyone could win every time. What sort of systems
and fail-safe measures will we have in place to protect the country from that
happening all of the time?[231]
Brian Tanguay argued that introducing a ranked ballot in single
member constituencies would only replicate problems found in FPTP:
I personally am not a fan of the alternative vote.
Although it does give greater choice to the voters, it seems to replicate all
the problems that we find in first past the post. The ranked ballot by itself
would not address the issue that we certainly heard from citizens at the time
of the law commission and at the time of the Ontario referendum. That system
would not address the flaws in the current system that are so in need of
resolution.[232]
In a similar vein, Nelson Wiseman noted how introducing a ranked
ballot itself could also cause distortion: “The ranked ballot, which I do
not oppose, may also cause distortion. A party could receive 40% of all the
first-choice votes and not win a single seat.”[233]
Indeed, in his presentation to the Committee, Byron Weber Becker
demonstrated how he assessed various electoral systems. A “well-behaved” system
is one that, in his assessment, “reduces distortion[234]
and strengthens the link between voter intention and the election of
representatives.” In other words “in a well-behaved system, the number of MPs
awarded is proportional to the number of votes earned.”[235]
According to Mr. Becker, AV “misbehaved” more than FPTP.[236]
He explained why by using an analogy to the “tragedy of the commons”:
I think that alternative vote makes a lot of sense
at the individual riding level. Let me say that I can appreciate why it would
be attractive at the individual riding level, but I think there are also some
problems at that level.
In each individual riding,
the decision is made independently of all the other ridings, the same as with
first past the post. It's when you aggregate all of those individual decisions
that it breaks apart and becomes a disadvantage for Canada as a whole. I have
sometimes compared it to the economic theory of the tragedy of the commons,
where a village has a common pasture and everybody grazes their cow on that
common pasture, and it works out wonderfully, as long as everybody obeys the
rules. But then some bright soul says, “Ah, I can graze two cows on that
pasture.” They make a locally optimal decision just for themselves. It's like
the individual riding saying that it's best for it to use alternative vote. If
everybody does that, the pasture gets over-grazed and everybody fails. If all
of the ridings use alternative vote, then the system as a whole becomes very
disproportional and Canada as a whole suffers.[237]
Henry Milner further argued that preferential ballots in single
member constituencies would result in less diversity in the House of Commons:
I've never understood the advantage of
preferential systems per se. It seems to me that we know the disadvantages,
which are that these systems make it difficult for parties that are not within
the mainstream — even harder than it is under our system — to get elected, so
you have less diversity.[238]
In his appearance before the Committee, Harold Jansen noted that
his research into the historical use of AV and STV suggested that electoral
results under AV did not differ much from FPTP. By contrast, the use of STV, a
proportional system, did have a significant impact:
My research into the uses of preferential
balloting in Canadian provincial elections has led me to conclude that the
alternative vote is probably not the best option for Canada.
The historical experience with AV suggests that it results in election outcomes
that differ little from those we would encounter under the first past the post
system, and it wouldn't do anything to address the most serious shortcoming of
first past the post: the failure to produce a legislature that accurately
reflects the preferences of Canadians. In Alberta and Manitoba, the system had
no impact on proportionality, which is how political scientists measure the
correspondence between seats and votes. It had no impact whatsoever.
Even if we look at the district level,
the alternative vote produces results that differ little from first
past the post. In the entire experience of Alberta, with over 30 years of using
the alternative vote, fewer than 3% of all the seats contested would have
turned out differently under AV than first past the post. In Manitoba, the
number is less than 2%.
In other words, in 97% to 98% of the cases, the person ahead on the first count
ends up winning, and that person would have won under the first past the post
system.
…
By contrast, I'd argue the STV system used in
Edmonton, Calgary, and Winnipeg performed much better.[239]
Professor Jansen added that the Australian model of AV, whereby
voters must rank all candidates on the ballot, differs from the historical
Canadian us of AV in British Columbia, Alberta, and Manitoba:
When we imagine how the alternative vote might
work in Canada, we often cast our eyes towards Australia, but one of the things
we often overlook in the Australian case, is that in Australia, voters are
legally compelled to rank each and every candidate. That’s a big difference. In
[Manitoba, Alberta and British Columbia] we didn’t do that.[240]
In a background paper, Professor Jansen warned of the consequences
of mandating voters to rank all candidates on the ballot:
In Australia, the choice between optional and
compulsory preferences is intimately linked to the competitive position of
parties. Labor, often the victim of preference exchanges between its opponents,
has favoured repealing compulsory preferences, and has done so in Queensland
and New South Wales.[241]
Michael Gallagher emphasized how AV, by not being proportional,
would achieve little while causing smaller parties to suffer:
The alternative vote is not a kind of PR
and the outcomes it produces are not that different from first past the post,
really, so in some ways I think it would be a huge amount of effort to achieve
very little if Canada had a really strong deliberative process and then simply
moved to the alternative vote. It wouldn't make a great deal of
difference.
Yes, I think so, because the results of Australian
elections tend to be just as disproportional as elections in Britain or Canada,
for example. You don't get very close proportionality, and in particular the smaller
parties really lose out systematically.[242]
In a paper published in 2016 titled “The Consequences of the
Alternative Vote,” authors Lydia Miljan and Taylor Jackson note the political
implications of the Australian model of AV. That system was introduced in 1918:
… by a non-Labour government in response to the
formation of the Country Party, a competing right-of-centre party that was
establishing a growing regional influence. Implemening the AV electoral system
allowed the non-Labour parties to exchange voting preferences, particularly in
rural districts. This helped to prevent the Labour party from benefitting when
the non-Labour vote was split.[243]
Miljan and Jackson argue that the adoption of AV and the
requirement to exhaust the ballot had the effect of changing the outcome of
elections, in favour of Australia’s political right and against the political
left, for half a century. Miljan and Jackson note,
[N]on-Labour parties have tended to be the
beneficiaries of this system. In her examination of the effects that AV had on
Australian elections from 1919 to 1951, Rydon (1956)
found that of the 73 seats that required a redistribution of preferences to
determine the winner, 58 of them were won by non-Labour candidates and only 15
were won by Labour candidates….[244]
This suggests how a new electoral system can have a long-lasting,
predictable effect on the outcome of elections. It can favour one party and
hobble another over several election cycles in a manner that was foreseeable by
the legislators who designed the
new system.
Finally, in their paper, Miljan
and Jackson provide a detailed analysis as to how AV would have affected
Canadian federal elections since 1997. The most striking result is that the
Liberals would have won a larger number of seats in every election: 15
additional seats in 1997, 17 additional seats in 2000, 25 additional seats in
2004, 22 additional seats in 2006, 11 additional seats in 2008, 13 additional
seats in 2011, and 31 additional seats in 2015.[245]
As discussed above, a primary facet of a proportional electoral
system is a district magnitude greater than one. The greater the district
magnitude (number of members per constituency), the greater the
proportionality, as the entire range of voters’ party preferences would be more
accurately reflected. The trade-off that comes with increased district
magnitude is less local representation. Given the preferences expressed
throughout the Committee’s study for strong, accountable local representation,
any introduction of proportionality in Canada would need to maintain a district
magnitude that would not unduly dilute local representation. Indeed, as
explained by Pippa Norris, proportionality and local representation can
co-exist:
They can go together,
and what matters is the size of the district magnitude. If you have a small
district—Spain has three-to-five, Ireland has five—then essentially the
individual voter can find, in particular, not just an MP but a couple of MPs,
perhaps from different parties, to represent their constituency concerns or to
lobby for them or to do any other sort of service work.
If you get a large district, however, that
dilutes. Many countries will have districts of, say,
16-20, and there is no constituency service when you get to a very, very large
constituency. The classic cases are in Israel, where you have the whole country
as one constituency, and in the Netherlands. In those countries, there are very
weak links indeed between the members of the Knesset in Israel and individual
voters. At that stage it's broken.
It really depends on
how you draw your boundaries as to how you actually create an incentive to have
constituency service. It's not about an either/or system, PR versus single
member.[246]
Indeed, Laura Stephenson noted that in terms of district magnitude,
“[a]ny number greater than one would lead to more proportionate outcomes than
our current system, and many systems around the world have districts with low magnitudes.”
She added that “[e]xperts would suggest that between three and seven is an
ideal number.”[247]
One tool that has been developed to measure an electoral system’s
relative disproportionality between votes received and seats allotted in a
legislature is the Gallagher Index, developed by Michael Gallagher (who
appeared before the Committee). As noted by Byron Weber Becker, the Gallagher
Index “combines both over and under-representation for each party into a single
number.”[248]
According to Professor Becker, a Gallagher Index of less than 5 is considered “excellent”.
As well, Professor Becker developed the “Gallagher Index Composite” for the
Committee’s study, comprised of the “average of the Gallagher Indices for each province
and territory, weighted by its number of seats.”[249]
Professor Becker noted that:
This corrects for a problem in calculating the
Gallagher Index for the nation as a whole, which can hide regional
disproportionalities such as the significant over-representation of
Conservatives in the Prairies offsetting the over-representation of Liberals in
the Maritimes.[250]
Professor Becker submitted that the most recent FPTP election had a
Gallagher Index score of 12.0%, and a Composite Gallagher Index score of 17.1%.[251]
In his submission, Professor Becker provided the Committee with a chart of the
relative distortion and Gallagher scores under various potential electoral
systems (titled Summary of nine electoral systems’ properties):[252]
The primary argument raised in favour of proportional electoral
systems is that they more fairly translate votes cast for political parties
into seats in the legislature. Indeed, this element was emphasized by various
witnesses who testified before the Committee. For example, as explained by
Arend Lijphart:
The main aim of proportional representation is to
get proportional outcomes so that parties, or groups of representatives, are
representing roughly equal representations of the voters. PR systems differ in
terms of how proportional they are. They may use systems that are not
completely proportional and that raise barriers for smaller parties, and so on.
When you look at outcomes of PR systems, there is not one that is completely
100% proportional.[253]
And as echoed by Harold Jansen:
The one thing that the electoral system does is
alter the math of the translation of votes into seats. PR systems like STV or
MMP or list PR do that much more accurately than first past the post, or the
alternative vote. I'd argue that this is really the fundamental basis on which
a decision about electoral reform should be made.[254]
Henry Milner added:
The more proportional the system, the more equal
every vote is in terms of its ability to get somebody elected. So the less
proportional a system is, the less equal each vote is in terms of its effect on
getting somebody elected.[255]
Kenneth Carty observed that “fair representation,” equated with
some type of PR, was a desired objective of multiple provincial reform
initiatives:
Let me say that the thousands of Canadians who
took part in the recent provincial reform debates in the citizens' assemblies
in Ontario and in British Columbia, in the Commission on Legislative Democracy
in New Brunswick, on the parliamentary commission hearings all across Quebec,
have been very clear about what they want in an electoral system. They've
identified pretty clearly what they think are the three values that they most
highly value. One was fair representation, and by that most voters meant
something like proportional representation was the value that was to be put
near the top of the list.[256]
Professor Carty added that “strong, identifiable, local
representation” and “more choice on the ballot” were the other two values most
highly regarded by individuals who participated in provincial electoral reform
initiatives over the past 15 years.[257]
The value noted by Professor Carty also came through in responses
to the Committee’s online consultation. Indeed, 71.5% of respondents either
strongly agreed (59.1%) or agreed (12.4%) with the statement “Canada’s
electoral system should ensure that the number of seats held by a party in
Parliament reflects the proportion of votes it received across the country”:[258]
Number of seats held by a party in Parliament should
reflect
the proportion of votes it received across the country
Scale: 1 (Strongly Disagree) – 5 (Strongly Agree);
NA
As well, 72.5% of respondents either strongly agreed (48.6%) or agreed
(23.9%) with the statement that “Canada’s electoral system should ensure that
voters elect local candidates to represent them in Parliament”:[259]
Voters should elect local candidates to represent them in
Parliament
Scale: 1 (Strongly Disagree) – 5 (Strongly Agree);
NA
Finally, a majority of respondents either strongly agreed or agreed
with the following two statements connecting proportionality with voter choice:
First, that “Voters should determine which candidates get elected from a
party’s list and the seats in the House of Commons should be allocated based on
the percentage of votes obtained by each political party”:[260]
Voters should determine which candidates
get elected from a party’s list
Scale: 1 (Strongly Disagree) – 5 (Strongly Agree);
NA
And second, that “Canada’s electoral system should produce a
proportional Parliament (where seats roughly match the parties’ vote share)
through the direct election of local representatives in multi-member electoral
districts”:[261]
Canada’s electoral system should produce a proportional
Parliament through the direct election of local representatives in multi-member
districts
Scale: 1 (Strongly Disagree) – 5 (Strongly Agree);
NA
The Hon. Ed Broadbent, in his remarks to the Committee, noted that
experts in democratic institutions tend to favour PR systems over majoritarian
ones:
[W]hen the large majority of experts—not only
those who have made their presentations to this committee, but also those
around the world who have studied democracy and democratic institutions—make up
their minds about electoral systems, they come down on the side of some form of
PR.[262]
Finally, Arend Lijphart’s research highlighted the relationship
between PR and what he calls “consensus democracy,” through the increased
likelihood of coalition governments:
I've gradually come to the conclusion that
proportional representation, or PR, is the better option. This has also been
the trend among political scientists generally. The empirical evidence is now
overwhelmingly strong in support of this conclusion. PR is a crucial ingredient
in what I have called “consensus democracy”, especially in combination with a
parliamentary system of government. It tends to lead to a multi-party system,
which in turn tends to lead to coalition cabinets, and also leads to
parliaments that are stronger and cabinets that are less dominant than in
majoritarian systems. In addition, it tends to be associated with a more
co-operative system of interest groups.[263]
The Hon. Ed Broadbent added on this topic:
[T]he PR system is conducive to more civility in
politics. I had experience following my political life with, for example,
German politicians in both the CDU and the SPD. They both say, as people
familiar with the Scandinavian situation, that with multi-party systems in
which it's taken for granted that you're going to have multiple parties forming
governments, the politicians are more civil with each other before elections
and during elections because they know they're going to have to work with
somebody afterwards. That isn't a trivial point.[264]
As discussed above, the primary shortcoming of highly proportional
electoral systems is the diminution of local representation, which is why such
options are not being considered by the Committee. As well, some witnesses
raised the end of majority governments and the prospect of coalition
governments as being more complicated:
Nothing comes without problems, and there are two
problems in particular [with PR] that might be identifiable. One is that
constituencies as we call them, ridings, would have to be much larger, both in
geographical size and in population because proportional representation
necessitates multi-member constituencies, so ridings would be much larger, and
they already are huge in some cases. In addition, government formation becomes
a much more complicated process because single party government would be very
unlikely. It’s very hard for any party under a really proportional system to
win an overall majority. That’s not necessarily a bad thing; there are pros and
cons in coalition government, but it would become more complicated.[265]
Indeed, the increased likelihood of coalition governments in
proportional systems would be a significant change that would require both
adaptation and education:
[I]n Canada's case, clearly a significant change
in political culture would be required. One of the consequences of adopting a
proportional electoral system is the more frequent occurrence of minority
governments. The population is used to this type of government.
One thing would be different in Canada's case, and
that's the occurrence of coalition governments. There's no tradition in that
regard here. Sometimes coalition attempts were made that could be surprising,
but there was never a real coalition. This could be a significant change in
political culture.
Work must be done not only for the public, but
also by the witnesses who cover political life, and that requires information.
A factor that was also measured is the importance of having citizenship
education courses.…[266]
Of note, respondents to the Committee’s online consultation, who
overall preferred some element of proportionality, were open to the idea of collaborative
governments. Indeed, 53.5% of respondents either strongly agreed (31.8%) or
agreed (21.7%) with the following statement: “Canada’s electoral system should
favour the following outcome: no single political party holds the majority of
seats in Parliament, thereby increasing the likelihood that political parties
will work together to pass legislation.”[267]
No single political party should hold the majority of seats
in Parliament, increasing the likelihood of political parties working together
to pass legislation
Scale: 1 (Strongly Disagree) – 5 (Strongly Agree);
NA
Another concern that was raised during the Committee’s study is
that purely proportional electoral systems would lead to the proliferation of
small political parties at the expense of large national parties. This
apprehension was expressed by Kenneth Carty:
My view is that with a highly proportional
electoral system, there’s a major risk that we would lose our national
political parties. I think the electoral incentives would powerfully favour
regional and sectoral parties at the expense of national ones. Our national
parties might easily break into pieces of different kinds and different shapes
depending on the party.[268]
He added that the erosion of national parties could lead to the
erosion of national politics:
I believe under relatively highly proportional
systems, even moderately proportional ones, the big national parties would be
disadvantaged. In fact, it would be to the advantage of different parts of
these national parties to kind of go their own way, as the Conservatives went
three ways in 1993. Without first past the post, they would never have come
back together. I think that over time we would have, in fact, the erosion of
national parties because there would be electoral incentives in different
regions, among different groups, to produce their own candidates and not be
tied by a national platform. I believe the real risk of proportionality is the
erosion of national parties, and I believe, national politics.[269]
He concluded that the success of Canada’s current electoral system
has largely been in preventing regional/ideological cleavages from dividing the
country:
The great success of the Canadian party system, in
my judgment, has been in some sense preventing the enormous variation in the
cleavages, in the divisions of Canada, from spilling into our Parliament in a
way that would make us a dysfunctional country.… One of the strengths of the
way our system has worked is that it has in fact forced the parties in some
sense to work hard at preventing that expression of so much division, in a
country that’s constantly changing.[270]
Finally, related to Professor Carty’s notion of the erosion of
national political parties is the argument that greater proportionality would
favour the proliferation of smaller, and possibly “extremist” political
parties:
No doubt proportional representation, particularly
extreme proportional representation as they have in Israel, gives organization
and voice to very small groups of sentiment, some of which can be quite
extreme. While our current FPTP privileges the regional representation,
proportional representation privileges ideological representation.[271]
Other witnesses provided a less drastic assessment. For example,
André Blais suggested that increased proportionality would result in more
diversity of viewpoints, possibly adding to polarization in the House of
Commons:
I don't think the evidence is that clear on
exactly what the consequence would be. Well, there would be a consequence in
that there would be a wider array of viewpoints, and some of them would
probably be more extremist than they are now, so there will be more diversity
but also perhaps a little bit more polarization at the beginning in the House
of Commons.[272]
As well, Brian Tanguay did not share the view that introducing
proportionality would lead to the fragmentation of national political parties,
or favour the proliferation of extremist political parties:
I'm not as convinced as Professor Carty that it
would be the death knell for national political parties. I don't see Canada
being as riven by what political scientists call cleavages, as, say, Belgium
is. The model proposed by the law reform commission would have a built-in kind
of threshold. You'd need, probably, at least 10% of votes in a region to get
one of those list seats.
To me, the worry that there would be a proliferation
of fringe or extremist parties and that the national parties would fall apart
seems exaggerated. I just don't see—and I share, actually, your views so
eloquently stated in the preamble to the question—that we grew up or lived
through the near death of the country, all under first past the post. I don't
think that a mixed member proportional system would exacerbate regionalism. I
don't think it would be any worse than it is now.[273]
One proportional option for electoral system reform that was raised
by numerous witnesses is the Single Transferable Vote (STV), as it is
candidate-focused and preserves local representation:
The single transferable vote is known in the
Anglo-Saxon world, and it is for good reason known as the Anglo-Saxon PR.
Ireland, Malta, upper house Australia, it's well known in the Anglo-Saxon
world. It preserves local representation.[274]
As well, STV has a history in Canada. Provincially, from the 1920s
to the 1950s, Alberta and Manitoba both adopted STV for elections held in urban
ridings and the AV for elections held in rural ridings.[275]
Additionally, in the late 1910s to early 1920s, a number of municipalities in
Alberta, British Columbia, Manitoba and Saskatchewan adopted STV systems.[276]
Finally, British Columbia’s Citizen’s Assembly recommended “BC-STV” as the
system of choice for the province in its December 2004 report.[277]
Essentially STV aims to achieve a moderate level of proportionality
while giving maximum choice to voters and maintaining the local connection
between MPs and constituents. As explained by Michael Gallagher in his
appearance before the Committee:
One type of partial representation is proportional
representation by the single transferable vote. This aims to do a number of
things simultaneously. First, it attempts to achieve a reasonable closeness
between the share of votes cast and share of seats cast for each party. Second,
it tries to give a maximum choice to voters — more choice than open-list
systems. It avoids having voters waste their vote by casting it for someone who
has no chance. Third, it aims to retain the close territorial connection
between voters and MPs, or TDs, as deputies are known in Ireland. It aims to do
all of those things.[278]
STV works as follows:[279] voters
in multi-member electoral districts (Ireland’s districts contain three to five
members) rank candidates on the ballot. They may rank as few or as many
candidates as they wish. Indeed, this is the practice in Ireland, whereas in
the Australian Senate voters must rank all candidates.
In most variations of this system, winners are declared by first
determining the total number of valid votes cast, and then establishing a minimum
number of votes that must be garnered based on the number of seats to be filled
(the “vote quota”). Candidates who receive the number of first-preference votes
needed to reach the quota are elected.
If there are still seats to be filled, a two-step count occurs. In the first
step, any votes in excess of the quota for elected candidates are redistributed
to the second choices indicated on the ballots of the elected candidates, using
a weighted formula (this is called “excess transfer”). Candidates who then
reach the quota are elected. If no candidates reach the quota in this way, a
second step takes place in which the candidate with the fewest first-preference
votes (lowest-ranked) is dropped, and the second-preference votes on the
ballots where that candidate ranked first are assigned to the respective
remaining candidates. Such extra counts continue until enough candidates reach
the quota to fill all available seats. As noted by Professor Gallagher, “the
surplus distribution is the most complex part of STV.”[280]
Some variations of STV do not involve excess transfer, but only the
elimination of the lowest-ranked candidate and the reassigning of the second
preferences on the ballots for that candidate. This simplifies the counting
process.
In Ireland the counting process may take several days:
Counting proceeds until
all the seats are filled. The counting is a multi-staged process.
It takes much longer than a first past the post count. In Ireland we had an
election earlier this year. It was on a Friday, and the counting of the votes
didn't start until 9 o'clock on Saturday morning. Most of the seats were filled
by midnight on Saturday, but some went into Sunday. There was one constituency
in which the outcome was very close and there were a few recounts, so it didn't
end until early on Wednesday morning. Counting is not an instantaneous process —
it can be several days before the full result emerges.[281]
Professor Gallagher then described the impacts of Ireland’s STV
system in terms of representation, turnout, party cohesion, and constituent
links with MPs:
Firstly, in terms of the accuracy of
representation, it does give fairly accurate representation. It doesn't give
extremely high proportionality like the South African system does, but it gives
pretty average levels of proportionality by the standards of most European
electoral systems. It's much more proportional than non-PR systems such as
Canada uses or such as Britain or France use. On that criterion, it performs to
the satisfaction of people here.
In terms of government
stability, over the years there has not really been a problem there. Most
governments these days are coalitions, but they can be just as stable as
single-party governments. We've had 29 elections in the history of the state,
so something like three years between elections. Having said that, the last
election in February did not produce a very stable-looking government. We have
a minority government, with only 58 seats out of 158. It took two months to put
it together. Its lifespan is rather uncertain. At the moment we wouldn't rate
highly on current government stability, but over the entire period this has not
been a problem.
One of the strengths of
PR-STV, as I mentioned before, for its proponents is that it gives voters a lot
of choice. They can really say exactly what they feel. They're not compelled to
vote just for, to name the Irish parties, Labour or just for Fianna Fáil or
just for the Greens. They can vote number one for Green Party, and if the Green
Party candidate is eliminated, then they can give a second preference to
Labour, a third preference to Fine Gael and their vote isn't wasted, it still
counts. They can choose on the basis of any criterion they want. They can vote
on party lines or some people will vote on geographical lines. They want a
candidate from this part of the constituency, a candidate whose home base is
somewhere near here. For that reason they might give their first preference to
a local candidate from one party and their second preference to a candidate
from another party.
Do turnout levels engender
high participation? Not particularly in Ireland. Turnout is not especially
high. It was around 65% for the election earlier this year. But people who
study turnout say that it is affected by lots of different factors. The
electoral system might have only a minor role. The only other country in Europe
to use PR-STV is Malta, and that has a very high turnout, over 90%.
In terms of the cohesion
of parties, as I said before, this internal party competition doesn't really damage
party cohesion. In this country the solidarity of parliamentary groups is very
high. It's very rare for MPs to defy the party whip. For good or for bad,
that's the way it is. MPs nearly always vote the party line, they just don't
vote different ways. Whatever the local pressures might be, the parliamentary
parties are very cohesive.
Next is links with
constituents. It's quite interesting that this arises in the Canadian context
because this is quite a controversial point in Ireland. Links with constituents
are extremely strong in Ireland. Links between TDs-MPs and their constituents
are very strong. MPs spend a lot of time dealing with their constituents,
representing their constituents, meeting their constituents, taking cases to
central civil service bureaucracy on behalf of constituents.… For sure, there
doesn't seem to be any reason to be concerned that PR-STV would weaken
constituency links, if anything quite the contrary. Academics, as I say, take
that view.
The main point about PR-STV in this regard is that MPs now have a strong
electoral incentive to respond to constituents' demands.[282]
Professor Gallagher concluded his introductory
remarks with a suggestion as to how STV could work in Canada:
At the moment you've got 338 MPs, so if Canada had
PR-STV there might be around
70 to 90 multi-seat ridings, each returning anything from maybe three to seven
MPs, or it could be more. Just looking at a few particular provinces, we see
that Newfoundland and Labrador currently has seven single-seat ridings that
might become one three-seat riding and one four-seat riding, for example.
Prince Edward Island currently has four single-seat ridings that would become
one four-seat riding. New Brunswick currently has 10 single-seat ridings that
could become two five-seat ridings. It could be that really large geographical
areas like Labrador, the Northwest Territories, Nunavut, and Yukon would remain
as single-seat ridings.[283]
As suggested by Professor Gallagher’s testimony, the perceived
strengths of STV are that it introduces proportionality (albeit moderate
proportionality given smaller district magnitudes), it emphasizes voter choice,
and it is still founded on the link between candidates and constituents.
The primary shortcomings associated with STV are the perceived
complexity (and time required) to determine winners, and the notion that STV
results in candidates from the same party campaigning against each other.[284]
On the latter point, Professor Gallagher posited that in Ireland “this internal
party competition doesn't really damage party cohesion.”[285]
As well, in testifying about his experience as part of the B.C. Citizen’s
Assembly, Craig Henschel noted that internal competition could also have
positive elements:
There are two aspects to that. The assembly
members really like the idea of more competition, and Canada is paying more
attention to voters to get the vote. We also recognize that, because of the
preferential ballot, if you want to get elected, you can't say horrible things
about the other candidates, because you may need their support. You might need
support from their followers. The tenor of elections, the tone, should improve
even though the competition increases.[286]
Finally, various witnesses testified regarding the ideal district
magnitude in a system such as STV that would enable both proportionality and
local representation. Ireland’s district magnitude range of three to five
members per district is set out in its Constitution. Under BC-STV, each
electoral district would have had between 2 and 7 seats. Indeed, as noted by
Laura Stephenson, “Any number greater than one would lead to more proportionate
outcomes than our current system, and many systems around the world have
districts with low magnitudes.” She added that “Experts would suggest that
between three and seven is an ideal number.”[287]
Jean-Pierre Derriennic recommended, in his presentation based on
his recently published book, A Better Electoral System for Canada,[288]
that Canada introduce a
variant of Ireland’s system in Canada. In his system, called “moderate
proportional representation,” Canada would be divided into districts of three
to five seats. As explained in his submission to the Committee:
Implementing this reform would not be difficult
following these principles: the total number of MPs or their number for each
province would not change; contiguous districts would be joined together
without changing their present limits; in the new multi-member districts the
ratio of MPs by inhabitants should be kept as equal as possible.
Prince-Edward-Island would form one district with four seats. In other
provinces, the correct number of MPs would be met by combining districts with
three, four or five seats. The main danger of proportional representation, too
many parties having MPs, would be avoided.[289]
As well, his system would include ranked ballots, which he argues
should be a feature of any electoral system, whether proportional or
majoritarian in nature:
Ranked ballots should be, as a rule, a feature of
any electoral system, because citizens should have the right to vote sincerely
without having to suppose how others will vote and without being manipulated by
opinion polls and rumours.
In single-member districts, ranking preferences
rather than expressing one choice makes the results more legitimate, because
all MPs are elected by a majority of voters.
Ranking preferences is possible also when choosing
between lists of candidates in order to get a proportional result. It can be
done by using the Single Transferable Vote, as in Ireland, or by ordering
preferences between closed lists of candidates.[290]
However, to “avoid weakening the parties” (though intra-party
competition which occurs in STV), Professor Derriennic proposes the following:
Citizens would vote as in Ireland: the ballots
would list the names and party affiliations of the candidates, and the voters
would mark their order of preference, which may be complete or not. Ballots
would be counted not as in Ireland but rather through ranked ballot voting
between lists, as described above in this chapter: each party’s number of votes
is calculated by adding together all the first preferences for candidates of
the same party: votes for parties with too few first preferences to be entitled
to one seat are transferred according to the second or next preferences[291]; when the parties remaining in
competition are each entitled to at least one seat, they are assigned the seats
proportionally. The seats gained by a party are assigned to its candidates
according to the personal votes each of them got.[292]
Professor Derriennic argues that his system would be less
complicated than STV as there would be no need to calculate quotas and
redistribute excess transfers.
In his testimony before the Committee, former Chief Electoral
Officer Jean-Pierre Kingsley proposed that some level of proportionality be
introduced in Canada by clustering urban ridings into multiple member
constituencies while keeping remote, rural ridings under FPTP. Voters would
continue to vote once, whether in a single- or multiple-member constituency:
That being said, here is my suggestion. Since
Canada is so vast, we would keep the first-past-the-post system for remote,
rural or large ridings. About 40, 50 or 60 members would be elected using
that system.
As for urban areas, we
could cluster four or five current ridings and ensure that four or five members
are elected by the voters based on the vote results. I will not defend the
following to death, but according to my way of thinking, a voter would vote for
a party or a candidate. The candidates would be selected by the new cluster
association of the four or five ridings. So the people would be choosing.
As for gender parity,
let's say that there are five seats to fill. I would ask that three men and
three women be elected, and that the party choose, at a local level, one man,
one woman, one man, one woman, one man, one woman, and so on, so that it would
always be one, two, one, two, one, two.
In short, the voter would
choose. They would vote, as they currently do, for a candidate or a party. It
would be the same thing. There would be only one vote. From there, it would be
determined, for instance, that 60% of people voted for a given party, and that
there are three seats. So we would be talking about 20%.[293]
He added that independents would continue to be able to be elected
under his proposal:
In the example I’ve given, you could also have
independents. Their chances of being elected … would probably be the same
as they are right now. We can’t devise a system—at least, not readily—in which
independents would rule the day, but it is important for that phenomenon to be
able to express itself under our system, and there are various systems that
would allow that quite readily.[294]
The determination of whether a riding would qualify as rural or
remote would be made by looking at each province individually: “[E]ach province would have to be looked at
individually to see what people think of and accept as being rural, and what
people think of and accept as urban.”[295]
In response to Jean-Pierre Kingsley’s proposal, Fair Vote Canada
suggested an alternative model that it called Rural-Urban Proportional (with
the objective of making the system more proportional). Fair Vote Canada
describes the system as follows:
- 1) Multi-member ridings in the urban areas (which
could be elected with a ranked ballot - STV - or an open list)
- 2) Single member ridings in the rural and small
urban areas (which could also be elected with a ranked ballot – or by
first-past-the-post)
- 3) A small layer of regional top up seats to make
the overall results in the region proportional (an idea borrowed from Sweden,
where these are called “adjustment seats).[296]
In its submission to the
Committee, Fair Vote Canada suggested a variety of Rural-Urban PR models which
featured:
-
Single member or 2-3-member ridings in areas
where population density
is low,
-
the benefits of multi-member ridings where
population density is higher,
-
a top-up share of about 15 percent of total
seats, achieved by increasing the size of the House or the size of each riding
by 15 percent, a minor change.[297]
In their view, this would
provide a potential tailor-made solution for Canada:
Different applications use different approaches to
give each voter an effective voice. Features of the model can be adjusted to
good effect in each region of the country to provide a made-in-Canada solution
that provides the desired level of proportionality while still managing
differences in riding sizes between urban and rural areas and remaining
sensitive to local concerns and preferences.[298]
The Committee heard a significant amount of testimony concerning
mixed systems, particularly the mixed-member proportional (MMP) system. Mixed
electoral systems combine elements of plurality/majority systems with PR. Under
this system, some MPs are elected by a plurality or majority vote in single member
districts (often through traditional FPTP elections), and some are elected from
party lists in a proportional compensatory manner. The goal of such systems is
to maintain local representation while providing greater overall
proportionality in terms of popular vote to seat count in a legislature.
Essentially, in MMP systems such as those used in Germany, New
Zealand and Scotland, voters get two votes. The first vote (“constituency
vote”) elects an MP in constituencies via the standard FPTP method. The second
vote (“party vote”) determines the total number of compensatory MPs each party
will get. In most systems, the party
vote is primary; namely, a party’s share of the seats in the legislature is
determined by the party vote, and the number of constituency seats it wins is
then subtracted from this total. The remaining seats are filled from party
lists.[299]
David Moscrop outlined one of the primary perceived benefits
associated with MMP when he told the Committee:
MMP allows for direct local representation and
lives up to the commitment many Canadians have to fairness understood as a
proportional translation of votes into seats.[300]
As such, to some, MMP could address the principle of “effectiveness
and legitimacy” as it aims to translate votes into seats in a way that “reduces
distortion,” while maintaining the link between voter intention and the
election of representatives.
A majority of participants who advocated for electoral system
change proposed the adoption of an MMP system, suggesting that it maximizes
voter choice. Leslie Seidle said, “I think that the mixed
member model has a lot going for it because it can be structured to allow quite
a bit of voter choice.”[301] MMP allows voters to split their
vote—meaning vote for a candidate of one party in their riding and vote for
another party in the compensatory vote. Such an option would help resolve the
“wasted vote” argument that prevails under FPTP. Lee Ward added that MMP is:
The only system that empowers the voters is one
that ensures, to the greatest extent possible, every individual's vote—their
first choice, their real choice—will help elect their representative in Parliament.…[302]
Allowing greater voter choice on the ballot could
help address the problem of strategic voting. Craig Scott stated that:
In New Zealand, around 30% take up that option of
cross-voting. It means that the local candidates are more likely to be able to
attract votes for who they are, what they've done, what they can bring
nationally from the local level, without having to worry about the strategic
vote. I think this is an extremely important feature of MMP.[303]
Further, some witnesses noted that moving to an
MMP system would keep the electoral system relatively simple. Katelynn Northam
stated, “the local representation factor seems very familiar and similar to
what [we] know with the current first-past-the-post system. It feels relatively
simple and accessible on the ballot.”[304]
MMP has never been used to elect
representatives at the provincial or federal level in Canada. In March 2004,
the Law Commission of Canada, following a three-year study on electoral reform,
recommended Canada move to an MMP electoral system.[305] The
Law Commission suggested Canada adopt an MMP system for the following reasons:
-
Reduce the discrepancy between a party’s share of the seats in
the House of Commons and its share of the votes.
-
Greater inclusion of new and previously underrepresented voices,
such as small political parties.
-
To elect a greater number of minority group and women candidates.
-
Encourage inter-party cooperation through coalition governments.
-
Reduce the disparities in the value of votes in which winning
parties are often three to four times more “valuable” than a vote for any other
party.
-
Reduce the number of disregarded votes, thus increasing the
number of “sincere” votes, as opposed to strategic votes.
-
Produce more regionally balanced caucuses[306].
In the November 2016 electoral system reform
plebiscite held in P.E.I., following four rounds of counting, MMP
emerged as the preferred option among the five electoral systems under
consideration.[307] MMP was also the subject of a referendum in Ontario (2007) and an
earlier plebiscite in P.E.I. (2005), both of which failed to receive requisite
voter support. MMP was also recommended by Quebec’s Select Committee on
the Election Act and Citizens’ Committee in 2006 and
New Brunswick’s 2006 Commission on Legislative Democracy.
An important feature of the MMP system is
determining how candidates would be elected from the party lists. There are two
primary types of party lists, usually referred to as “closed” and “open.”
With a closed list, the party ranks the names on the list, and
citizens vote for a party, not a specific candidate. Once all votes have been
counted, each party is awarded seats in proportion to its share of the national
vote. Individual seats are then allocated to candidates of each party in the
order in which they are ranked on the party list. Critics of closed party lists
often note that it gives political parties too much control over which
candidates are elected.
Royce Koop observed that the use of party lists would be a “new
experience for Canadians.... It would probably not be thrilling for them in
terms of … being able to hold politicians accountable.”[308] Ms.
Mireille Tremblay echoed this view by stating that because closed lists are
completely controlled by parties “it is likely that [list] MPs would be more
accountable to the party than to voters.”[309]
As will be discussed in greater detail in the following subsection,
a major advantage associated with closed lists is that they can allow parties
to establish lists that will guarantee the election candidates from
historically underrepresented groups such as women, visible minorities and Indigenous
peoples.
With an open list, voters choose a preferred candidate (or
candidates) from the list of the party for which they wish to vote. This means
that voters determine the order in which the candidates on the list will be
awarded seats.[310]
James Bickerton noted that an open list MMP system could resolve the
accountability concerns raised by closed party lists. He stated:
I think there's no reason not to allow voters to
choose between party candidates. Some think that this would generate
competition within political parties between their candidates. Yes, it would,
but from a voter perspective and from a representation perspective, I don't
think that would be a bad thing.[311]
Pippa Norris noted that with an open list, “voters can express a
preference for a particular candidate within a complete list as well, so it
gives them a bit more choice.”[312]
On the other hand, Tana Jukes noted at the open-mic session in Victoria that “open
list MMP … could offer some improvements over our current system, but I am
concerned about the complexity.”[313]
Open list MMP would require voters to familiarize themselves with a greater
number of candidates prior to voting, making the electoral process more onerous
on voters.
Of note, the vast majority (70.1%) of respondents to the
e-consultation indicated that they “strongly disagreed” or “disagreed” with the
statement that political parties should determine which of their candidates get
elected from party lists.[314]
On the other hand, most respondents (59.6%) to the e-consultation strongly
supported or supported the view that voters should determine which candidates
get elected from a party’s list.[315]
As a final point on open and closed lists, it is important to note
they are flexible and many variants of how candidates are elected from lists
are used around the world.[316]
The 2004 Law Commission suggested a middle-ground option, which suggested
providing voters with the option of either endorsing the party “slate” for
their region or indicating a preference for a candidate within the list.[317]
As well, a distinct hybrid option is to use a “best runner’s up” model, also
known as “Baden-Wurttemberg” after the German Land (state or province) in
which it is used. Additionally, as explained by Fair Vote Canada:
One way to simplify balloting for the top-up seats
is to allocate top-ups using a best runners up approach. Under this model, used
in Germany’s Baden-Wurttemberg province, the top-up seats are allocated by drawing
from a party’s defeated candidates in the region, starting with the candidate
who got the most support without being elected.[318]
A number of witnesses held that although electoral reform is not
the stand-alone solution to increasing the representation of traditionally
underrepresented groups such as women, visible minorities and Indigenous
people, party lists could be a useful tool.
If underrepresented groups continue to face obstacles getting elected in
individual constituencies, under closed-list MMP, parties have the ability to
ensure they are elected via party lists.[319]
Mr. Peden noted that this is precisely what has occurred in New
Zealand. He stated that “MMP has resulted in more women and
more Maori elected to Parliament, the majority of them elected as list MPs.”[320]
Many witnesses also held that although balanced
party lists are useful in increasing the diversity of candidates and MPs in a
moderate way, they are ultimately a “Band-Aid” solution to a problem that rests
primarily in the hands of political parties. Melanee Thomas stated that the
election of women and other historically underrepresented groups has not
happened organically.[321]
She added that it is not going to happen “organically under our system and they
are not going to happen organically just because you change to PR.”[322] Amanda Bittner added that “while it is the case that proportional
systems tend to be associated with greater levels of diversity, that link is
still dependent on a commitment from parties to put forward diverse lists of
candidates.”[323]
According to political scientist Joachim Behnke of Zeppelin
University in Germany, the party lists have proven to be the “best opportunity
to force every party to give half of their seats to women,”[324] but
that parties have taken this responsibility themselves. He stated that there
are no legislative quotas in Germany, but that parties have
established informal or voluntary commitments to ensure the representation of
certain groups.
Finally, with regard to diversity and open versus
closed lists, the Committee heard testimony suggesting that voters will vote
for diverse candidates on open lists. Laura Stephenson shared her research on
voters’ likelihood to elect women from open lists:
In any system that involves a list of candidates,
we have to start thinking about the placement of those names on the list. In a
closed system, where the parties have full control over the order in which the
candidates would receive seats, it's important that there is some kind of alternation,
or that at least the under-represented groups aren't placed in winnable
positions. In open list systems this is not as important. In some research I've
done with colleagues, we found that letting people vote in an open list system,
where they get to choose, increased the representation of women, which is of
course good news, right? The disadvantage that women supposedly represent has
not been supported with evidence.[325]
Canada’s geography and Constitution would most
likely require that list MPs be elected through party lists established in each
province and territory. Indeed, this was the conclusion reached by the Law
Commission’s 2004 study.[326]
Prior to election day, each province and territory would draw up a
list of candidates that would be elected via party lists. Some
provinces may require a number of party lists due to high population. Further,
as David McLaughlin suggested, “regional boundaries could be drawn under
an MMP system could very much safeguard and protect communities of interest
where you have significant minority communities located.”[327]
Roderick Wood discussed the Law Commission’s thinking on provincial
and
sub-provincial party lists:
We proposed that the list, except for Quebec and
Ontario because of the size of those provinces, would be on a regional basis,
so you would have your provincial list. What that would mean is that if you
have a province like Newfoundland and Labrador with seven MPs, then there would
be four constituency MPs and three list MPs. Every province would have its own
list.[328]
The question of whether candidates could run as “dual
candidates”—meaning that a person is a candidate in an individual riding as
well as on a party list, was also raised as important issue of consideration.
Louis Massicotte stated:
Under MMP, it is usually possible for a candidate
to stand for a riding and to be on a list, for a very simple reason: the more
successful a party is in a riding, the fewer names it has on the list. As a
result, it is better to try both avenues because when members declare their
candidacy, the final outcome is not known. That is the beauty of democracy.
Otherwise, if you think you will be very successful and run in a riding, but
things change and you are defeated in the riding, you have lost the security
that the list affords.
[D]ouble candidacy is perfectly legitimate,
although it will meet with a great deal of resistance among the public and
among MPs.[329]
Professor Behnke noted that in Germany, the practice of a single person
simultaneously being a constituency candidate and a compensatory candidate is
commonplace:
[M]ost list members or most list-seat members are
also, in many cases, constituency candidates. They have lost in their
constituency, but they have a special relationship to the constituency, so they
are known, and they have an office in the constituency.[330]
Other witnesses, such as Christopher Kam in Vancouver, questioned
whether dual candidacies would be viewed as fair or legitimate. He held:
[I]f you lose an election, you lose an election.
When you have dual candidacy, the members are allowed to contest the district
and the list, and this can almost always ensure their election or at least
insulate them from defeat.[331]
Dual candidacies could, as Professor Kam implies, make it difficult
for voters to oust an incumbent candidate that is not locally popular. Benoît
Pelletier made a similar observation when speaking about Quebec’s consideration
of MMP and dual candidacy:
What was not acceptable to some people was the
idea that someone would be a candidate in the riding and at the same time would
be at the top of a list. When that person was defeated in the riding, it was
the result of a democratic expression of the population that “We don't want
that person” or that “We prefer another person.” The idea that the person could
be an MP or an MNA through a list was not something that pleased
parliamentarians, in particular, first, and some parts of the population
second.[332]
Laura Stephenson raised another common criticism
of MMP when she told the Committee that she does not support MMP because it
creates “two different classes of MPs.”[333] As MPs can be elected to
represent individual constituencies and others are elected through the party
list, some feared that it alters the traditional roles of MPs and raises
questions of accountability. Patrice Dutil outlined some of the key concerns
surrounding the two types of MPs. He stated:
The idea of having two classes of MPs I don't
think will jive with our political culture, where you have one class of MPs who
will cater to the needs of the constituents and another class of MPs who are
always on the list and who are always going to be there. I think Canadians like
to have their Members of Parliament accountable.[334]
Dr. Nelson Wiseman also noted that a divide in
parliamentary duties amongst the two different types of MPs could be
problematic.[335] For example, if constituency MPs carried
out the majority of constituency work, who would list MPs be accountable to?
On the other hand, some suggest that the compensatory MPs could be viewed as
“second-class” MPs because they do not have to undergo the difficult process of
being elected in a constituency.
Roderick Wood, former Commissioner of the Law Commission
of Canada’s 2004 report on electoral reform noted the Law Commission’s
consideration of this concern:
We looked at the argument of the creation of two
classes of MP, the worry being that the list MP, not being voted in, would be
the second-class citizen. We saw that wasn't the case.
In Germany, in New Zealand, they're both MPs and their parties ensure that the
list members have an equitable division in terms of constituency work. In fact,
for voter choice it enhances them, because you can go to your constituency MP,
you also have a regional MP you can go to, and that may be a person from a
different party.[336]
Pippa Norris also suggested that having two types
of MPs does not necessarily pose a problem. She stated:
[I]t does mean that Members
of Parliament would be slightly different in their roles and responsibilities
and in how much they do for constituency service, which is an incredibly
valuable service that takes up a lot of time and is appreciated in any
parliamentary system, versus those who are focused more on committee work or
issues or other types of concerns for Parliament. You just divide the roles a
bit more than you might do under the current system.[337]
Experts from New Zealand and Germany both stressed
that in practice, the two types of MPs do not pose a problem for citizens or
MPs. Robert Peden of New Zealand said “a parliamentarian elected from the list
has exactly the same entitlements and responsibilities as a member elected from
an electorate.”[338] Friedrich Pukelsheim stated:
There's no difference
in their functions and their access to financing and political positions. The
difference is in the understanding that half of them directly represent a
constituency.… They are active, they maintain office hours, visits,
associations, and they try to be visible. In Germany the everyday political
work is very similar between both types of representatives.[339]
Furthermore, Joachim Behnke noted that in Germany,
many MPs elected through party lists were unsuccessful constituency candidates.
As such, many compensatory MPs have a direct connection with voters from their
region.[340]
Finally, some stated that different types of MPs
can provide more effective representation for the electorate as it allows some
MPs to focus specifically on local issues and others to work on broader
regional issues. Ultimately, this could provide citizens with more avenues to
engage with MPs on the issues that are important to them.
It was widely accepted among witnesses that single-party majority
governments would occur infrequently under an MMP electoral system. Brian
Tanguay noted that if MMP were adopted one of the most significant consequences
would be that “coalitions of necessity would become the norm.”[341]
MMP elections generally result in the election of minority or
coalition governments. New Zealand provides an interesting example, as Mr.
Peden noted:
New Zealand has now had seven MMP elections. Each
election has resulted in between six and eight parties represented in
Parliament. Each election has resulted in some form of coalition government or
arrangement between political parties, as is to be expected under a
proportional system. Each government has retained the confidence of the
Parliament throughout the parliamentary term.[342]
Many witnesses and citizens raised concerns about the likelihood of
coalition governments produced by MMP. As Louis Massicotte noted:
In Canada … we do not have a coalition culture.
Coalitions are not viewed favourably by the political class and by part of the
public. Political actors will probably adapt, but that adjustment will not
necessarily be easy.[343]
Similarly, Nick Loenen suggested that Canadians simply would not be
comfortable with “chronic coalition governments.”[344] Peter
Loewen also noted that “blurred accountability and behind-door
compromises that occur after an election and between elections” are some of the
major drawbacks of coalition governments. Professor Loewen added that reform to
a proportional system such as MMP would create a “potentially permanent role
for small regional parties” and that small parties will potentially have undue
influence in government.[345]
In addition to the concerns regarding coalition governments and the
increased representation of small parties, many suggested that multi-party
governance would be beneficial to Canada’s parliamentary democracy. Arendt
Lijphart suggested that when multiple parties are elected to Parliament or are
represented in cabinet, it:
[L]eads to parliaments that are stronger and
cabinets that are less dominant than in majoritarian systems. In addition, it
tends to be associated with a more co-operative system of interest groups.[346]
Jean-Pierre Charbonneau added that coalition governments could
create a culture of collaboration and compromise in federal politics. He
stated:
Coalition does not imply that our governments are
unstable.... Having to make compromises with political opponents, just as with
people whose ideology is closer to our own, actually creates a favourable
political climate. When it comes right down to it, people are fed up with
excessive partisanship and behaviour that devalues the institution of politics.[347]
Although Canada does not have a history of
coalition governments, international experts from Germany and New Zealand
described how political parties, voters and ultimately Parliament can adapt.
Professor Behnke from Germany described the German experience with coalition
governments:
The formation of coalitions is really not so
complicated in most cases, because we have something like pre-coalitions in the
electoral campaign.… In many cases, people say that the flaw of proportional
systems is people not knowing which coalitions they will get, but in reality
this is not the case, because in most cases they get what they voted for.[348]
Some suggested that small parties with too great of an influence on
governing parties could be undemocratic and unrepresentative. Others are of the
view that coalitions are the best avenue for small parties to ensure their
supporters’ views are reflected in government.
One way some countries with MMP systems have addressed the threat
of the election of “fringe” or “extremist” parties is through the use of
thresholds. For example, to be eligible to receive a share of the party vote
seats in New Zealand, a party must garner at least 5% of the national vote or
win one electorate seat.[349]
In an MMP system, an important consideration is the ratio of
constituency seats to list seats. Determining such a ratio requires finding the
appropriate balance between the desire for effective local representation and
proportionality. Ms. Mary Pitcaithly, the Convener of the Electoral Management
Board for Scotland noted that determining the ratio in Scotland was:
[A] political decision. It was entirely the
decision of the Parliament. It was based on the intention that the new
parliament would be proportional but without going as far as 50/50.[350]
In order to implement an MMP system in Canada, one of the following
would have to occur:
-
Maintain the current number
of MPs: the number of constituency MPs would be reduced to allow for the
addition of compensatory MPs. Consequently, electoral districts would become
larger in terms of population and geography.
-
Increase the number of MPs: electoral districts would remain the
same
and a set number of compensatory seats would be added to the existing
338 members.
The ratio of constituency to compensatory MPs varies between
jurisdictions.
In Germany, half of the MPs are elected in individual constituencies and half
are elected via party lists. Whereas in New Zealand, 70 MPs are elected in
individual constituencies and 50 are elected from political party
lists.[351]
In 2004, the Law Commission recommended two-thirds of MPs be
elected in constituency races and the remaining one-third be elected from
provincial or territorial party lists. The Commission noted that avoiding
increasing the size of the House of Commons was a priority in determining said
ratio. David McLaughlin, who oversaw New Brunswick’s Commission on Legislative
Democracy (2003-2006), noted that the same ratio was recommended by New
Brunswick’s Commission on Legislative Democracy in 2006 because it “ensure[d] necessary local representation while
introducing a sufficient degree of proportionality to be meaningful in
translating votes into seats.”[352]
Royce Koop held that if MMP were to be adopted in
Canada, additional members would have to be added to the House of Commons
because “cutting back the number of constituency MPs to make room for
list MPs would hurt the quality of constituency representation.”[353]
The Committee heard compelling testimony from Canadians across the
country regarding the challenges of implementing PR in the territories. Each
territory has a
single seat in Parliament and an extremely widely dispersed population. As
David Brekke in Whitehorse described, the North is “overrepresented
population-wise but very under-represented when geographical area is
concerned.”[354]
Much of the discussion in the territories centered
on the unique circumstances that Canadians in each territory face in terms of
electoral participation and gaining adequate representation. John
Streicker held that “No matter what system you ultimately propose, please do
not lose local representation for the north.”[355] The
Hon. Louis Sebert noted,
that “any consideration of electoral reform should recognize
the circumstances of [the Northwest] territory.”[356]
Some participants and witnesses in the territories
noted that if Canada were to move to an MMP system, that the North could not be
simply excluded due to their small populations. As Dennis Bevington, former MP
for the Northwest Territories, stated in Yellowknife: “I think that for us not
to be involved in mixed member proportional would make us second-class citizens
under voting.”[357] Consequently, some individuals, such as
Andrew Robinson[358] and John Streicker[359]
suggested adding a second compensatory MP to each territory to allow for some
degree of proportionality if an MMP system was adopted.
Recommendation 1
The Committee recommends that the Government should, as it develops
a new electoral system, use the Gallagher index in order to minimize the level
of distortion between the popular will of the electorate and the resultant seat
allocations in Parliament. The Government should seek to design a system that
achieves a Gallagher score of 5 or less.
Recommendation 2
The Committee recommends that, although systems of pure party lists
can achieve a Gallagher score of 5 or less, they should not be considered by
the Government as such systems sever the connection between voters and their
MP.