Electoral reform is on the agenda in at least five provinces and each has
taken a different approach to the process of implemention electoral reform.
The House of Commons has also examined the question and in June 2005 one
of its Committees presented its report. This article considers why it has
been so difficult to reach any consensus on electoral reform and suggests
a way to allow the Canadian electorate to play a larger role in the ultimate
Given that the first-past- the-post system of electing MPs has had very
few overt defenders in the past few years1particularly among politicians,
who are ever-sensitive to the ebb and flow of political fashionone could
be forgiven for thinking that it will be easy to develop a nationwide coalition
in favour of a new and better electoral system. But such a coalition is
unlikely to gel any time in the near future, if the country insists on
using its traditional political processes for seeking change.
The reason for this is straightforward. Politicians are primarily concerned
about getting reelected, and notwithstanding the failings of the current
electoral system, every MP in the House of Commons understands that the
first-past-the-post system (FPTP) has one redeeming virtue: it got him
or her elected. So naked self-interest on its own will not predispose MPs
to unite behind any single alternative to the status quo.
To be sure, most party leaders can point to an alternative electoral system
with respectable antecedents in some other country, that would have produced
more seats for their own party than actually were produced by FPTP in the
most recent federal election.2
But any specific new system can only be beneficial to one or another of
the political parties if it is at the expense of one or more of the other
political parties. This is true whether the alternative under consideration
is the Irish-style single transferable vote in multi-member ridings (STV),
the Australian system of preferential or alternative voting within single-member
ridings (AV), or the German multi-member proportional system (MMP). In
the end, any change to the status quo must inevitably harm the interests
of more sitting MPs than it helps.
The likeliest scenario is that MPs of all stripes will support electoral
reform in principle, while shifting majorities within the House of Commons
will remain opposed to any specific proposal in practice.
As is always the case when the status quo confronts a range of deadlocked
alternative options, neither positive action nor public advocacy is required
in order to keep FPTP in place. FPTP is simply the default solution that
goes on and on, as long as majority support cannot be cobbled together
for any specific alternative.
The irony, is that under such a scenario, no politician need stick out
his or her neck to actually defend the FPTP system from which he or she
is benefiting. Instead, each elected official need only point to his or
her own preferred electoral system as the only truly acceptable solution
to what ails Canadian democracy, and then vote against whatever other alternative
is being placed before the House of Commonsin the name of democracy.
Being freed from the need to defend the status quo is liberation indeed,
since the perverse results of FPTP in Canada are so widely known that they
scarcely bear repeating. A sample of the oddities that this system has
produced at the federal level includes:
1963-1980: the disastrous impact on national unity of the near complete
freeze-out of Conservative MPs from Quebec prior to the Mulroney sweep
of 1984, and the near-elimination of Liberal representation in Western
Canada under Prime Ministers Trudeau and Turner.
- 1993: the reduction of the Progressive Conservative caucus to two MPseven
though the party had won a greater share of the popular vote (16%) than
did the Bloc Quebecois, which captured 54 seats with 13.5% of the national
vote and became the Official Opposition. This election, more than any other,
proves the validity of Andrew Coynes assertion that the party that can
cluster its votes geographically will win many more seats than a party
whose support is spread more broadly and evenly, rewarding regional grievance-mongering
at the expense of a national vision. 3
- 1997: FPTP was responsible for turning Jean Chrétiens very poor 38.5%
showing at the polls into a majority mandate.
- 2000: I had a personal taste of how FPTP distorts electoral results when
I first entered the House of Commons as one of only two Canadian Alliance
MPs elected in Ontario, in an election where my party had won half as many
votes as the Liberalswho were rewarded by FPTP with 100 Ontario seats.
Similar stories occur at the provincial level. Examples include the NDP
victory in British Columbia in 1996, and the Parti Quebecois victory in
1998, in elections where both parties had lost the popular vote to their
Liberal opponents. Equally peculiar has been the grotesque over-weighting
that sometimes has occurred when a party has been awarded every single
seat in a provincial legislature, as the result of an election in which
it has won a much more modest percentage of the vote. This happened, for
example, in the New Brunswick election of 1987, when Frank McKennas Liberals
won just over twice as many votes as their Conservative opponents, and
took every seat in the Assembly.
So public opinion is not a barrier to electoral reform in Canada. The biggest
obstacle to reform, whether at the federal or provincial level, is that
our politicians keep on trying to design and ratify proposed changes, using
the very mechanisms that electoral reform is designed to replace. For example,
Prince Edward Island, New Brunswick and Quebec have all been using traditional
commissions, of one sort or another, to develop proposals for new electoral
The Federal Approach
In autumn 2004 the House of Commons Standing Committee on Procedure and
House Affairs was charged by the Commons with the task of designing a plan
for a national review of the available options for replacing FPTP. The
committee engaged in months of hearings on the methods that other countries
had used to develop their new electoral systems, and took two very expensive
and controversial overseas investigatory trips to Australia, New Zealand
and Europe.4 In June 2005, the committee voted to ignore absolutely everything
that it had spent six months investigating, and decidedover the protests
of a few of its membersthat the best way to achieve electoral reform in
Canada would be to set up yet another House of Commons committee (a special
committee, for what it is worth, as opposed to a standing committee),
which would design the new electoral system.5
Any of the models that the special committee will examine would harm the
reelection chances of a majority of the MPs sitting on the committee. In
the unlikely event that the special committee can nonetheless come to a
consensus that sets aside considerations of self interest, and can somehow
find a way of causing the majority of MPs in the House of Commons as a
whole to similarly vote against their own best electoral interests, Canada
will get the new electoral system that most of its politicians claim in
principle to want. But the far more likely result is a deadlock, either
in the committee or in the Commons.
The depressing truth, which became clear as the Standing Committee on Procedure
and House Affairs undertook its study of different roads to electoral reform
earlier this year, is that electoral reform in the countries we examined
had rarely been achieved by means that Canadians couldor would want toemulate.
In practice, electoral reform has typically been imposed in three different
ways imposition by outside forces, unilaterally by the majority party
or by accident.
Irelands much-admired STV system was imposed by the departing British
in the early 1920s as a way of ensuring that the Protestant minority would
not be frozen out of the Irish parliament, as they might have been under
British-style FPTP system. Germanys much-admired MMP system was imposed
by the victorious Allies in the late 1940s as a way of ensuring that no
marginal party could ever again elbow its way into a position of power,
as the Nazis had done in 1933. Scotlands MMP system was imposed by London.
The innovative STV model now being used for the legislature of the Australian
Capital Territory was imposed by Australias federal parliament. Clearly,
this method of achieving electoral reform is not open to Canadians.
In a number of other jurisdictions, perfectly good electoral reforms have
been imposed for overtly partisan reasons. Australia provides the best
example of how this can happen. Here is how the Australian Electoral Commission
describes that countrys shift from FPTP to AV for elections to its lower
The Commonwealth Electoral Act
was comprehensively rewritten in 1918 ...
and the new Act among other things introduced alternative (preferential)
voting for the House of Representatives; this was in response to the rise
of the Country Party in the aftermath of the First World War, and the consequent
prospect of loss of seats to Labor through a split in the non-Labor vote.6
In short, a governing party with a majority mandate realized that a clearly-defined
change to the electoral system would suit its own partisan interests, and
therefore it enacted that change.
Australia repeated this process when it introduced STV to its Senate in
1948. This time, it was Labor that was in power but facing imminent defeat.
Recognizing that the existing electoral system would exaggerate its decline
in popular support,7 and not believing that it could win the coming election,
the Labor government enacted changes to the electoral system that it concluded
would help to consolidate its parliamentary power base in the Senate.8
Although both the AV and STV models now employed in Australia have their
admirers, it is clear that the means by which they were imposed are not
admirable, and at any rate, could only be imposed in this country if Canada
had a majority government in which the governing party were able to ram
through its changes unilaterally through both the House of Commons and
the Senate. This seems unlikely in the short term.
Finally, electoral reform may be introduced accidentally as was the case
in in New Zealand. MMP was introduced in the 1990s as the unintended consequence
of efforts by the countrys established parties to appear to be favourable
to the principle of electoral reform, while developing de facto roadblocks
to any such reform. In 1992, as a part of this disingenuous effort, an
indicative (non-binding) referendum was held on two questionsshould
the voting system be changed, and which system should be adopted? The second
question had been designed as a FPTP question. Voters could indicate support
for a single option. The assumption of the authors of the referendum was
that support for the various alternatives to the status quo would be so
split that FPTP would win a plurality of the vote even though it lacked
majority support in the country. Thus, the results of the second question
would obviate the mandate for electoral reform that was likely to be recorded
in the first question.
What had not been anticipated was that voter turnout would be very low,
meaning that a disproportionately high percentage of participating voters
would be advocates of electoral reform. Thus, 84.7% of the 55.5% of voters
who bothered to participate cast their ballots in favour of changing the
voting system. As well, the government had not anticipated that there would
be an aggressive advocacy campaign in favour of only one of the three alternatives
that it had placed on the ballot as alternatives to the status quo. The
result of this miscalculation was that a very high proportion of those
who cast ballots did so in favour of MMP. Thus, by accident, a single alternative
to FPTP was identified, and all other options were eliminated from public
debate, at the very same time that an overwhelming (and probably artificially
large) mandate had been given for casting aside FPTP.
With an election looming this combination of results made very difficult
for the government to push the now-dominant MMP option aside. Nevertheless,
an effort to derail electoral reform was made: a binding referendum was
held in 1993, in which voters were given a choice between FPTP and MMP.
Prominent politicians from both the major parties campaigned hard in favour
of the status quo, and worked to boost voter turnout (most notably by holding
the referendum at the same time as a general election). As a result of
these efforts, voter turnout rose to 85.2%. This time, a slender majority
of 53.9% voted in favour of MMP.9
While there can be no doubt that MMP therefore received a genuine popular
mandate, it is also clear that Canadas politicians cannot knowingly engage
in an attempt to achieve unintended consequences of the sort that the New
Zealanders had imposed upon themselves.
The only alternative to the three scenarios outlined above would be to
capitalize on the stated support in principle for electoral reform of all
parties in the House of Commons, in order to establish a mechanism which
is guaranteed to produce a new electoral model, without any prior knowledge
on the part of the politicians (or of anyone else, of course), as to which
model will in the end be adopted. Each party would have to accept the risk
that the model chosen might not be the one that it prefers, but each party
would also have the knowledge that it could potentially come out of the
process as a winner. All parties would, effectively, be stepping behind
what the philosopher John Rawls has referred to as a veil of ignorance.
Lessons from the BC Experience
This veil of ignorance was the key to the success of the worlds most successful
effort, to date, for building a popular mandate for electoral reform. British
Columbias Citizens Assembly, which produced a proposed model for electoral
reform in October 2004, was a truly representative body. Membership in
the Assembly was randomly selected, and no politicians with a partisan
axe to grind sat among its membership. On May 17, 2005, its proposal for
an STV electoral model was put to the voters of British Columbia in a referendum,
and received a 57% mandate, including majority support in all but two of
the provinces electoral districts. This is the most substantial popular
mandate, of which I am aware, that has ever been given to any model of
But even in British Columbia, the old way of doing things helped to snatch
defeat from the jaws of victory. Not only did the government require a
60% majority for the referendum to succeed but the Assemblys recommendation
in favour of STV was submitted to the voters as a take-it-or-leave-it proposition.
People could vote for the new STV system, or for the old FPTP system, but
other options, such as Australian-style AV and German-style MMP, were simply
excluded from the ballot. Voter approval of STV would end any prospect
of another system finding its way onto the ballot.
This led to a situation in which former leaders of the movement for electoral
reform either became neutral, or campaigned openly against the STV option,
in the hope that they would be able to defeat this particular reform model,
while keeping alive the popular pressure for electoral reform, thereby
raising the probability that their own preferred model could be introduced
at some point in the future.
For example, the provincial Green Party had long been a strong supporter
of electoral reform, and probably won much of its support in the 2001 provincial
election based on this policy. During the public hearings part of the Citizens
Assembly process, party leader Adrienne Carr worked hard to convince the
Assembly to adopt MMP. 10 But when the Assembly chose instead to endorse
STV, the Green Party became neutral in the referendum, while Carr and the
partys only elected representative, Vancouver School Trustee Andrea Reimer,
campaigned against STV (and hence, in favour of FPTP). 11
Carole James, the NDP leader, was not quite so brazen in her opposition
to STV. She remained officially neutral throughout the concurrent election
and referendum campaigns, but publicly announced on May 18, one day after
the referendum, that she had voted against STV. I didnt feel that STV
was the direction to go, she explained in an interview. I felt that there
were other models to look at. I think mixed member proportional representation
meets the needs of the population of British Columbia. She went on to
recommend a second referendum, presumably on MMP, to be held concurrently
with municipal elections in November.
The kind of strategic voting encouraged by the Green and NDP leaders was
probably enough, in the end, to cause STV to fail to achieve the 60% threshold
required in order to be put into effect for British Columbias 2009 general
election. With former partisans of reform like Carr and James pulling their
support away from reform, STV won the support of only 57% of voters-three
percent shy of the magic number.
This has put British Columbia into a very awkward situation. Prominent
pro-reform campaigner Julian West (who has sought both Green and NDP nominations
in the past) expressed his frustration this way:
I dont see anything procedure-wise thats putting mixed proportional
back on the stage at this point. Nobody went out and campaigned on the
basis that a [referendum] result like this would be a mandate to do that
[i.e. to set aside STV and move on to a showdown between MMP and FPTP].
There is now no mandate in British Columbia to make any move from the status
quo, given that a mandate for accepting STV requires an unobtainable majority,
but that a mandate to set STV aside and to move on to one of the other
alternatives would presumably requireat the leastthat the STV option
had been rejected by at least half the voters.
But the status quo FPTP system has also effectively been rejected, since
it was presumably supported only by that portion of the 43% of British
Columbians who voted against STV (and who did so for reasons other than
those that motivated Ms. James and Ms. Carr).
An editorial in the Vancouver Sun on July 21, 2005, points out the difficulties
in establishing a mandate for any new electoral system under these circumstances:
[Carole] James has suggested simply mounting a second referendum which
would offer voters more choices. There are practical obstacles to that
route ... , not least what to do if none of the options garnered majority
support. (Stick with the status quo? Hold a run-off referendum?) .... In
any event, civic elections rarely draw the levels of voter participation
you see in provincial elections.
One obvious way to deal with the problem of legitimacy is to avoid the
need for super-majority criteria in any future referendum. In referendum
votes, as in votes taken in the legislature, there is much to be said for
the simple rule that 50% plus one constitutes a mandate.
However, it also seems reasonable to accept Ms. James underlying assertion
at face value. She is saying, in so many words, that she believes that
British Columbia voters would have selected MMP over STV, if those two
options, rather than STV and FPTP, had been on the ballot. We have no way
of knowing whether this is true or not.
Needed: A Preferential National Referendum
At the federal level, we could avoid the BC situation if we were to establish
a Citizens Assembly not for the purpose of pre-selecting a single option
to be placed on the ballot in opposition to FPTP, but rather to design
several options, which would be placed on the ballot in competition to
one another. Voters would then have the option of ranking the competing
I am advocating that Canada should use a preferential referendum whereby
voters would place a 1 on the ballot beside their preferred option, a
2 beside the option that they like second-best, and so on. If no single
option won a majority of the votes, the least-favoured option would be
dropped from the ballot, and the ballots of voters who had chosen this
option as their first preference would be redistributed to the options
that had been their respective second choices. This process would continue
until a single option achieves a clear majority.
Under a preferential referendum, voters would have the option of indicating
their preference for the option of which they most approve, without having
to make FPTP the default option. Advocates of all options could aggressively
campaign in favour of their preferred option without having to become de
facto champions of the status quo, as occurred in British Columbia.
Preferential balloting is the best way of arriving at consensus outcomes,
when no obvious majority exists; this is why it is used by many political
parties, including my own, for selecting their leaders. Moreover, the idea
of using a preferential referendum for selecting a new electoral system
is not new. The process was advocated as long ago as 1997 by the Reform
Partys task force on electoral reform, for which I was the researcher.
More recently, Fair Vote Ontario has taken stock of the strengths and weaknesses
of BCs Citizen Assembly process, and made the following recommendation:
The BC Citizens Assembly was instructed to work with the current number
of seats in the BC legislature and to recommend only one system. We believe
such restrictions should be removed to allow the Ontario Citizens Assembly
to recommend whatever they believe best for Ontario .... If they cannot
reach a general consensus on the single best alternative voting system,
the OCA should be allowed to present two alternatives, with voters using
a preference ballot in the referendum to choose among the alternatives
and the status quo. 12
I support the application of this approach at the federal level, and would
take it further: The Citizens Assembly should be mandated to place several
options before the people of Canada, designing each of its alternatives
to be as complete as possible, as appealing as possible, and as reflective
as possible of the values that Canadians would like to see encapsulated
in their electoral system. Then the decision should be turned over to the
voters, who willas Canadians always dochoose the wisest and most generous
compromise, from among the available options.
1. A notable exception has been Lawrence Solomon. See his various opinion
pieces in the National Post in 2005.
2. The exception to this rule is the Bloc Quebecois. It is very difficult
to imagine an electoral system better suited than FPTP to translating the
Blocs geographically-concentrated base of support into a large number
of seats in the House of Commons. The Blocs record is as follows: 1993:
13.5% of the national vote produced 54 seats; 1997: 14.6% of the national
vote, 44 seats; 2000: 10.7% of the national vote produced 38 seats; 2004
12.4% of the national vote produced 54 seats.
3. Andrew Coyne and Lawrence Solomon, BCs Democratic Revolution,
Post, May 16, 2005.
4. On February 1, 2005, committee members (including me) voted a travel
budget of $289,695 for the European and Antipodean trips. Later, while
the committee was abroad, committee member Françoise Boivin complained
to the media about the large size of the travel budget.
5. I prepared a dissenting report for Conservative MPs on the committee,
which indicated our preference for a Citizens Assembly on the British
6. A Short History of Federal Electoral Reform in Australia. http://www.aec.gov.au/_content/when/history/short_history.htm.
Accessed April 15, 2005.
7. The Australian Senate was, at this time, elected by means of FPTP in
multiple-member constituencies (six senators being elected at a time from
each state. Under this system, known as Block Voting, each voter was
given six votes, and the first six candidates past the post were elected.
This system is still widely used in Canada for municipal elections in small
towns where the town is not divided into wards, and consequently, all voters
can cast ballots for all candidates for town council.
8. John Uhr, Why We Chose Proportional Representation,
Senate of Australia:
Papers on Parliament, no. 34, Chapter 2. http://www.aph.gov.au/Senate/pubs/pops/pop34/c02.htm.
Accessed April 15, 2005.
9. Source for all referendum statistics relating to New Zealand: Helena
Catt (Chief Executive of the New Zealand Electoral Commission), New Zealands
Experience with MMP (AMS). Mimeograph distributed to Canadian MPs from
the House of Commons Standing Committee on Procedure and House Affairs
visiting New Zealand, April 2005, p. 2.
10. Gordon Gibson, the designer of the Citizens Assembly process, stated
I think there were attempts to influence the public hearing process
Adrienne Carr, leader of the Green Party, claimed that 80 percent of the
testimony supported her favoured system [MMP]. If so, that is a great tribute
to her organizing ability, but not a reflection of the views of British
Columbians, in my opinion. Fraser Forum, February 2005, p. 25.
11. Brad Badelt, Vote-reform supporters vow to continue STV fight,
Sun, May 23, 2005.
12. See Fair Vote Ontario, Ontario Citizens Assembly: Improvements on
the BC Model. Mimeograph. January 20, 2005.