At the time this
article was published John Courtney was professor of political science at the
University of Saskatchewan. This is a revised version of his testimony to the
Standing Committee on Privileges and Elections on November 19, 1985.
I would like to
raise six points about the question of representation in light of the most recent
Government White Paper and the legislation, Bill C-74 which you are now
First, although it
may be supported on at least one ground, I am not sorry to see the end of the
present method of distributing seats in the House of Commons the so-called
Amalgam Method which automatically increased by four the number of seats
allocated to Quebec at each redistribution and assigned seats in the other
large provinces in strict proportion to Quebec's population.1 The
Amalgam Method accepted implicitly the notion of at least some growth in the
size of the House of Commons to keep pace with a growing population. How
fortuitous or how deliberate this may have been at the time the legislation was
adopted in 1974 is, I think, an open question. It is a fact that the Amalgam
Method remains truer to the principle of representation by population than any
of the major alternatives considered in the past few years. Of course we must
recognize that would be equally true of any system encouraging growth and a
larger House as opposed to any system deliberately designed to keep the House
at a more or less fixed size.
The objections to
the Amalgam Method are, in my view, powerful, and I am persuaded by them. The
system is complicated and difficult to explain and to understand. There is a
great deal to be said for simple, uncomplicated and readily understandable
rules and processes in politics.
The Amalgam Method
also introduced, through its three categories of provinces, an invidious
comparison amongst the provinces – small, medium and large. In my view, this
sort of policy should be avoided wherever possible in a federal system. The
provinces ought to be treated, as much as is realistically possible, as equals.
doctrine of representation by population is one that has never been greatly
respected in Canada. Historically, this might be best illustrated by the number
of seats awarded tile six provinces to have entered Confederation after 1867.
In every case the number of seats was larger than strict rep-by-pop principle
had it been applied. Manitoba, rather than getting one seat, got four; B.C.,
rather than getting two, got six; and so on.
Once that kind of
bargain is struck (so far as provincial representation in the House is
concerned), it leads tit a very difficult situation. It cannot be significantly
altered in my view without in some way offending the provinces. Significant
alterations are difficult at the best of times, but I think trying to gain
provincial approval would be rather tricky if there were any serious move to
adopt a strict rep-by-pop system. We are into something that is rep-by-pop with
The doctrine of
representation by population has not surfaced in any of the recent
parliamentary debates that I am aware of. The failure of parliamentarians to
consider rep-by-pop as a significant doctrine is, in my view, best exemplified
by the debate at the time of the adoption in the 1960s of the whole new
redistribution system. That happens to be the longest debate on redistribution
in our parliamentary history. It lasted 51 days, over two sessions of the 26th
Parliament and culminated with adoption of Electoral Boundaries Readjustment
Act in 1964.2
virtually no support by any of the Members of Parliament for the cause of
reforming the redistribution process and setting up independent electoral
boundary commissions on the grounds of the one-man one-vote principle. That is
interesting because at that very titre in the United States there was
considerable debate about mal-apportionment resulting from the 1962 Baker vs.
Carr decision and we do not see that reflected in the Canadian context.
Moreover, at the present time I see no ground swell of public opinion, nor do I
see any considerable body of support in Parliament, for the strict adherence to
the principle of representation by population. Canadian political institutions
are notably different from American in this respect and I am not persuaded that
we need to end that difference.
My third point
relates to population projections. Growth and shifts in population are not
necessarily going to produce the kind of results you might expect. Demographers
base their forecasts on a number of assumptions – principally mortality,
fertility, internal migration patterns and immigration as a function of public
policy. One of the reasons that the Amalgam Method ran into trouble was that it
was based on some provincial population projections which proved to be
amazingly wrong. The 1971 projections for the total population a decade later
were accurate to within 200,000 of the actual 24.1 million people in 1981. That
is really very good but the distribution of population among the provinces
proved to be very erroneous. I will give two blatant examples based on the 1971
estimates of how demographic forecasting can go amuck.
In 1971, Quebec
had a population of about 6 million. 'I he demographic forecast that Statistics
Canada put out after the 1971 census for the Quebec population in 1981 showed a
projected population of about 7 million. In fact, Quebec's 1981 population came
in at 6.4 million. It was 600,000 wrong on the basis of a projection ten years
On the other hand
Alberta had a 1971 population of 1.6 million. It was anticipated by Statistics
Canada, based on the projection in 1971, that by 1981 Alberta would have 1.7
million. In fact, 10 years later, it proved to have 2.2 million. It was on the
plus side by 600,000. This becomes important when you look at the public policy
dimensions based on population forecasting.
You might be interested
to know that in 1974, the year that the Amalgam Method was introduced, Mr.
MacEachen presenting the legislation on behalf of the government said that the
anticipated size of the House of Commons after the 1981 census and
redistribution would be 294 members. This would be 12 more than the House of
the post-1971 redistribution. Instead, as we know, the growth in the House of
Commons membership turned out to be over twice that. If the 1981 redistribution
based on the Amalgam Method had taken place the House would have grown by 28 to
a total of 310.
In the past
decade, I have noted at least four different projections by the federal
government and by the Chief Electoral Officer of the size of the House in the
year 2001, all using the Amalgam Method. They range from a low of 354 seats to
a high of 381, a difference of 27 seats or about 8%, and that is just in one
decade, using one formula, but with different population projections.
My advice is to
establish the redistribution principles you want to adhere to but use
population projections as no more than an aid, and a limited one at that, not
as a goal. Even so, having once done that, do not be surprised if by 1991 and
2001 the House is riot of a different size that what you now anticipate it will
be. I think this is the critical point. Ontario is already halfway along its
projected one million person increase in population from 1981 to 1991, and we
are only four years into this period. For its part, Alberta is already at its
1991 projected figure based on projections after the 1981 census.
A fourth point
concerns clauses in Bill C-74 (Redistribution Act) which clearly have
established two categories of provinces i.e. the grandfather and senatorial
clauses which preclude any province from having fewer MPs than it has Senators
arid the limitations in Bill C-74 imposed on the growing provinces limiting
them to half the number of seats to which their growth would otherwise entitle
them. Now, there may be a very good case for protecting provinces from ever failing
below a certain minimum, although the history of Canada in this respect is a
spotty one. You may recall some of the details of how Ontario was protected
earlier in the century through the application of other safeguards against
If my arithmetic
is correct, and if the projected population figures prove to be reasonably
accurate, the size of the House will be somewhere in the range of 298 to the
low 300s in the first decade of the next century. I would commend to you for
serious consideration an amendment to Bill C-74 which would eliminate the
limitation on a province's growth. Rather than having 289, we would go to
something in the order of 295 seats.
Fifth, I seriously
question the need (a) to abandon the rule that limits constituency size to no
more or less than 25% the average size in each province, and (b) to substitute
the notion of "extraordinary circumstances" to justify larger or
In my view, it is
contradictory to introduce an "extraordinary" clause in light of the
fact that the same bill includes the principle that the size of seats will
correspond "as close as reasonably possible" to the electoral quota
(population divided by, number of seats) for the province. While that certainly
would not be the first contradiction in redistribution legislation, I wonder if
this perhaps should not, in some sense, be addressed.
"extraordinary" clause is undoubtedly also going to place a
tremendous burden on electoral commissions. Individuals and groups appearing
before commissions will use whatever weaponry they have available. If it is
possible to make some sort of case on "extraordinary" grounds I
rather expect the commissions are going to be bombarded with exceptions,
particularly on grounds of "community of interest". That kind of
argument can be very easily made by groups and individuals in their formal
presentations and briefs arid testimony before commissions. Commissions,
particularly in the larger provinces, work to a very tight timeframe. In
Ontario they barely got in under the one-year limitation this last time around.
I think we do not want to place any more obstacles than necessary in the way of
a most expeditious hearing.
clause, of course, would not be used only in the isolated and sparsely
populated regions. One can think of metropolitan constituencies in downtown
Toronto, for example, that have a community of interest defined ethnically. It
is entirely conceivable that sort of sociological dimension may very well be
pushed by ethnic communities in urban constituencies.
"extraordinary" lacks precision. I, frankly, do not know what the
limits would be, and I think if you know what the limits are or what you mean
by the term then you should certainly try and spell them out in the
Electoral Boundaries Readjustment Act was adopted a trend has been established
in the direction of meeting the stated goal of proximate population equality;
that is plus or minus 25% at the mean rather than at the extremes. Since 1965,
the commissions have moved away from relying on those upper limits and they
have moved much more in the direction of proximate population size by
that let me refer to the redistributions of 1961 and 1981. In 1961, one out of
every ten seats was at the 20% or 25% range, and one quarter of the seats were
within 1% to 5% of the mean average population. In 1981, only 3% of the seats
were at the outer margins (plus or minus 25%,) and fully 40% of all the seats
in Canada were right at the mean (plus or minus 1% to 5%). Fully 80% of the
seats drawn in the post-1981 redistribution had populations within plus or
minus 10% of their provincial quota.
What is also
striking about this development is that it has taken place in the face of clear
and growing opposition from parliamentarians from northern and sparsely
populated constituencies to the increasingly large geographic areas they were
being expected to represent. So you get a difference between the way the
commissions are moving and the way some parliamentarians would like to move.
With respect to
[lie tolerance limits, my own research suggests a determination on the part of
the commissions to draw boundaries as they – not as the MPs – would have them.
The 1983 redistribution proposals confirm this. Of the five provinces with
sparsely populated arid large geographic constituencies only two, Ontario and
Quebec, made any use of the minimum tolerance limits at all, and sparingly at
that, as only, six seats were so constructed at the margin. In 1983, for the
first time since independent redistribution began, all the seats of one
province were designed within plus or minus 5% of the provincial mean, and that
was my own province, Saskatchewan.
The facts imply
that the various commissions, in keeping with the exhortations of the act, have
shifted the grounds for redistribution much more in the direction of
egalitarianism and away from parliamentary concerns based on territorial
considerations. The impact, as I understand it, of Bill C-74 is to challenge
that trend by encouraging individuals and groups to mount cases on
"extraordinary grounds in special circumstances". In my view, the
plus or minus 25% gives sufficient range to the commissions to construct scats.
Finally, I am
concerned about the time taken to complete redistributions in Canada. I am
impressed how quickly the Australians carry out their redistribution. This is
important because we, in a sense, cribbed the Australian system two decades
ago. We lifted it, for the most part, and transplanted it into the Canadian
context and it has worked very well here. ]'here are a number of parallels that
might be noted, of course, between the two systems but I will not go into
those.3 But it is a matter of months in Australia from the beginning
of the redistribution process to the end.
Since 1983, when
the Australian Parliament amended its legislation, MPs no longer have an
opportunity to approve or in any sense alter or defeat redistribution
proposals. Redistribution responsibilities are taken right out of the hands of
parliamentarians completely and they have been devolved entirely on the
distribution committees at the state level. They promulgate their own decisions
which are then final without any debate or approval in Parliament.
I would suggest
serious consideration be given to adopting the same practice here. In
particular I would eliminate the following stages: tabling of the various
reports with the Speaker, filing objections by members, disposing of the
reports by the committee and then reports back to the commissions. By my
estimation, that could save something in the order of three months in the whole
process in Canada. Time, of course, also means a consequent cost saving as
There is also the
question of fairness. Why should the MPs be more privileged than anyone else in
making their views known at a separate stage? The evidence suggests that
increasingly, MPs are appearing at an earlier stage in the redistribution
process. They are appearing before and making formal presentation. to the
boundary commissions in their own provinces. That is an established and
empirically verifiable trend. Members prefer, at least publicly, to go with
that stage. I would assert that the second stage is somewhat redundant,
especially if the commissioners do little to change their redistribution based
on the parliamentary representations brought forward at the objection stage. In
other words, there is very little impact that parliamentarians actually, have
on the commission's final decisions. I therefore would suggest dropping the
stages that I have referred to and give the commissioners the final say.
1. For more
information on the Amalgam Method see John Courtney, "The Size of Canada's
Parliament: An Assessment of the Implications of a Larger House of
Commons," published as part of volume 38 of the Macdonald Commission
Background Studies by the University of Toronto Press.
2. I have examined
that debate and the principles behind Canadian redistributions in a recent
article entitled "Theories Masquerading as Principles: Canadian Electoral
Redistributions and the Australian Model" in The Canadian House of
Commons: Essays in Honour of Norman Ward, University of Calgary Pres.
3. My testimony
before the Standing Committee on Privileges and Elections on March 19, 1985
dealt with some comparisons between the Australian and Canadian electoral