Canadian Parliamentary Review

Current Issue
Canadian Region CPA
Upcoming Issue
Editorial and Stylistic Guidelines

HomeContact UsFrançais

Electoral Reform and the Future of Parliamentary Institutions
Gary Levy

At the time this article was published Gary Levy was with the Research Branch of the Library of Parliament.

Part I: Electoral Reform

In January 1979 the Task Force on Canadian Unity recommended the size of the House of Commons be increased by about 60 members who would be selected from provincial lists of candidates prepared by the federal parties in advance of a general election. Seats would be distributed between parties on the basis of percentages of popular vote. While this was the first time this particular system has been suggested for Canada the inadequacies of the present electoral system have been frequently exposed.

"In every election since 1896 the incoming government has ridden into power with more seats than its portion of the national vote gave it. Sometimes the discrepancies have been really shocking. In 1930, when the Conservatives won, they pooled 48 percent of the vote and secured 56 percent of the seats. Five years later, after the 1935 election, the shoe was on the other foot. The Liberals won only 44.8 percent of the vote but got 70.6 percent of the seats. These figures also show how our election system multiplies slight shifts in voting at the polls into big landslides for the winner in terms of seats in parliament. In 1930 for example, the Conservatives increased the number of seats from 91 to 137. But there was no landslide in public opinion. The voting showed that they had won only 3.3 percentage points more of the votes. Five years later the Liberals were swept back into power by something that looked more like an avalanche. They just about doubled the number of their seats, from 91 to 173. Almost a 100 percent increase in seats, but how much of an increase in popular vote? 80 percent? 50 percent? No, it was actually half a percentage point less than in 1930!"(1)

According to the Task Force our present electoral system also produces a distorted image of the country by making provinces appear more unanimous in their support of one federal party or another than they really are. It cited specifically the case of Quebec where the Liberal Party consistently wins an overwhelming proportion of the seats without a correspondingly high percentage of the popular vote and Alberta where since 1972 two out of five voters have favoured candidates other than those of the Progressive Conservative Party but only Conservative members have been elected.

The Task Force suggested that experience in other federations indicates that when party membership in a central parliament becomes concentrated in regional blocks it is an advanced signal of eventual disintegration.

"Because we see developing signs of such a situation in Canada we have come to the conclusion that electoral reform is urgent and of very high priority ... Westerners in particular increasingly resent a disproportionate number of Quebec members in the Liberal Caucus which has very few of their own. if there were more Quebec members in the Progressive Conservative Caucus representing more accurately the popular vote in that province, that caucus would be in a better position to reflect and understand the concerns of Quebecers."(2)

The purpose of the present paper is to examine more closely both our present electoral system and various alternative systems in order to decide if these conclusions of the Task Force are justified.

The "First-Past-the-Post System"

First-past-the-post is a term commonly used to describe the present electoral system in Canada, Great Britain, and other countries where the candidate topping the poll in each constituency, however small his fraction of the total vote, is judged to have won the seat. If only two parties contest each constituency it is theoretically possible for one of them to win 100% of the seats with 50.1% of the popular vote. Where several parties are competing it is not unusual for a party to control the assembly with considerably, less than 50% of the popular vote.

The basic virtue attributed to the first-past-the-post system is that it creates legislative majorities which are essential for cabinet stability in systems of responsible government. It also promotes direct accountability of the individual legislator to the constituency which elected him, a virtue often lacking in systems of proportional representation. The first-past-the-post system is intended. to reflect the assumption. that politics is best conducted through. an adversary system with debate taking, place between the Government on one side. and the Opposition on the other. Of course the rules of procedure, the design of the chamber and the role of the Speaker are also designed to foster this type of adversary politics.

Critics of adversary politics say it may also encourage persistent, irresponsible competition and too much oversimplification. For example where conflict does not exist, adversary politics manufactures it; and where the clash of opinions and interest is many-sided and complex, adversary politics offers little hope of creating that basis of consensus which is indispensable if there is to be effective political authority. Adversary politics:

"seeks to make pervasive the convention that debate can take place on a motion which, by definition, must be accepted or rejected. Yet it requires of little reflection to see that this is an artificial and misleading view of political choice. Sometimes we are indeed faced with straightforward choices between policies which really are incompatible one with another. But more of ten it is absurd to talk of only two alternatives. Instead we face multiple possibilities and can secure consent to a course of action only by combining in some ways several of them."(3)

Other defects in the present system have been outlined by Professor Allan Cairns of the University of British Columbia who has noted:

"The electoral system has not been impartial in its translation of votes into seats. Its benefits have been disproportionately given to the strongest major party and a weak sectional party. The electoral system has made a major contribution to the identification of particular sections-provinces with particular parties. It has undervalued the partisan diversity within each section/province. By so doing it has rendered the parliamentary composition of each party less representative of the sectional interests in the political system than in the party electorate from which that representation is derived. The electoral system favours minor parties with concentrated sectional support, and discourages those with diffuse national support. The electoral system has consistently exaggerated the significance of cleavages demarcated by sectional/provincial boundaries and has thus tended to transform contests between parties into contests between sections/provinces."(4)

The tables on the following pages show the percentage of popular vote and seats won by the party which formed the government in each federal and provincial election since 1945. It shows a consistent tendency of the first-past-the-post system to give the government more seats than it deserves according to its popular vote. Indeed on only one occasion (Prince Edward Island, 1966) did a government receive a percentage of seats less than its share of the popular vote. In provincial elections the electoral system frequently provides a majority of seats to a party with less than 50% of the popular vote, however, this is not true of the twelve federal elections held since 1945. On only five occasions (1945, 1949, 1953, 1968 and 1974) has a government. attained a majority of seats with less than a majority of the popular vote. Six times there has been a minority government and once (1958) a government received both a majority of votes and seats. Yet the same electoral system in the provinces has produced only six minority governments (Ontario 1975, 1977; Newfoundland 1971; Manitoba 1958, 1969 and British Columbia 1952) in the 97 provincial elections since 1945. (5) The table also shows the few occasions when a party has formed a government despite having the second largest popular vote. (Quebec 1966, New Brunswick 1952, 1974, Nova Scotia 1970 and in the federal elections of 1957, 1962 and 1979).


















































































































































































































































































































































































































































c indicates coalition government, m indicates minority government, * indicates government formed by party with second highest popular vote.

Proportional Representation

There are as many different varieties of proportional representation as there are countries using it, however, two main types of proportional representation can be distinguished – the single transferable vote and the list system.

A. List Systems

List systems are intended to insure that seats are distributed among parties in proportion to the total number of votes each party receives. For example if all of Canada were one constituency electing 300 Members of Parliament and if the Liberal party received 40% of the popular vote it would be entitled to 120 members. Alternatively a province or a town could constitute a constituency electing five or ten members. A party polling two-fifths of the vote would be entitled to four of ten seats. This ostensibly simple idea has led to the establishment of some very complicated voting systems. Complications arise from trying to relate not only the number of seats held to the polling strength of the parties but at the same time permitting voters an opportunity to express an opinion on the personal merits of individual candidates.

It is usually impractical to treat a whole country as one constituency, although this is done in Israel. Most countries establish a number of large multimember constituencies. To ensure that each party obtains its proportion in relation to the votes it polls, the number of seats is divided into the total votes cast, thus establishing a quota. For example, if a constituency elects five members and the total number of votes cast is 300,000, the quota will be 60,000. Each party will be entitled to one seat. for every 60,000 votes it polls. A complication arises from the fact that there will generally be remainders of votes left over after the quota has been divided into each party's total. This difficulty can be resolved by allotting any seats still available to the party or parties with the largest remainder. However, this method is not as fair as it appears. Assume that three deputies are to be elected and that the total of votes is 3,000. Therefore, the quota necessary to elect one member will be 1,000 votes. After the election it is found that party A received 1,600 votes and party by 1,400 votes. Under the largest remainder system party A, having the largest remainder. would obtain two seats and party B only one seat. But what if the members of party B decided to present two lists so that the resulting vote would be as follows, party A - 1,600 votes, party B-750 votes and party B2 650 votes. The quota would still be 1,000 votes but party A would only obtain I seat, whereas party B would obtain 2 because each of its two lists would show a remainder larger than A's. To resolve this problem a formula was devised by H.R. Droop aimed at reducing the importance of remainders in the allotment of seats .

(Total votes + 1)

Total seats + 1

This method yielded a smaller quota than. the original rule and enabled more seats to be allotted at the first distribution. A further improvement was devised by Victor d'Hondt of the University of Ghent. Its object is to secure that when all the seats have been allotted the average number of votes required to win one seat shall be as nearly as possible the same for each party. The votes cast for each party are ascertained and those totals are each divided in turn by the numbers 1, 2, 3, and so on as far as may be necessary.







Dividing by 1






Dividing by 2






Dividing by 3







The five highest numbers (five being the number of vacancies to be filled) are then arranged in order of magnitude and the lowest of these numbers, 27,381, is the "common divisor" or "electoral quotient". It forms the basis for the allotment of seats. The number of votes obtained by each of the lists is divided by the quotient giving party A two seats and parties B, C and D one seat each. Party E having less than the quotient gets no seat.

In practice it does not matter very much whether one uses the Droop or d'Hondt methods. They produce different results only in a minority of constituencies and even in those cases they rarely affect more than one seat. If the constituencies are large, returning 10 members, the fact that the allocation of the tenth seat might depend on the formula will have little effect on the overall composition of the Parliament. All party list systems give the parties a percentage of seats closely approximating their percentage of votes. Discrepancies are extremely small compared with those under majority systems. But this is only one aspect of the system. Each seat has to be filled by an individual man or woman and the question arises as to how he or she is to be selected.

The simplest form of list system does not attempt to give the voter any personal choice within the party's list. Thus in a given constituency, each party nominates as many candidates as there are seats to tie filled. The party decides the order in which it wishes the candidates to be elected and the names are printed in that order from the ballot paper. Each party has a separate ballot paper. The voter selects the paper of the party he wishes to support and places it into the ballot box. If the formula allots one seat to that party the candidate whose name appears first on the list is declared elected; if the party is allotted two candidates the second name will also be elected and so on. Th e voter as such, apart from any influence he may exert as a party member, has no voice in the selection of the person who is to represent him. He may prefer the last candidate on the list whose chance is virtually nil but there is nothing he can do about it. Most countries that have tried this system including Germany 1919-1933 and France 1945-1946 have abandoned it in favour of a system which gives the individual voter more influence in the choice of his representative. (6)

Most countries using list systems have modified them to give voters a choice between candidates on the same list or even between candidates on different lists. Such lists are printed side by side on the ballot paper and the parties do not necessarily present as many candidates as there are seats to be filled. There may even be "lists" of a single independent. The voter chooses either the entire list or one of the candidates indicating that he supports this party but prefers that candidate to the rest of his colleagues. After the number of seats each party is to have has been determined by the d'Hondt method the candidates who will fill those seats are selected as follows: all the ballots which favoured the list en bloc form a pool from which the candidates on that list draw, in succession, as many votes as are necessary to make their individual total equal to the electoral quotient, the process continuing until the pool is exhausted. Suppose that the electoral quotient is 3,750 and that list 1 contains 3 candidates, A, B and C. The votes have been cast as follows:

List en bloc


Candidate A

500 preferential votes

Candidate B

500 preferential votes

Candidate C

3,000 preferential votes




Since the party has polled twice the quotient it receives 2 seats. Candidate A being the first in order on the list has the first claim on one seat. The electoral quotient is 3,750 and A's total of 500 is raised to this number by the addition of 3,250 votes taken from those cast for the entire list. This secures his election and there remains 750 list votes which are attributed to candidate B, this candidate being the second in order on the list. B, however, has had only 500 votes and his total therefore amounts to 1,250. But candidate C has obtained 3,000 votes all recorded for himself personally. As this exceeds B's total of 1,250, C will become the second member elected from this list.

It can be seen that the candidate placed by his party first on the list has a great advantage and he is nearly always the first to be elected. It is, however, not unusual For the second on the list to be displaced by the third, as in the above example. Should the voters refuse to give any list votes at all, the order of election of the candidates would be determined solely by the voters' preferences without regard to the wishes of the party organisation except insofar as the approval of the latter is necessary for a candidate's appearance on the list in any position. This system is used in Belgium.

Switzerland goes much further, giving the elector the choice of a number of candidates, not necessarily confined to one party list. For elections to the lower house of the Swiss parliament, each canton (or in three cases half a canton) is one constituency, returning a number of Deputies proportional to its population. The four cantons so small as to be entitled only to one member each elect him by the first-past-the-post method; the rest elect several Deputies by a proportional system.

Any 15 electors in a constituency can nominate a list of candidates, not exceeding in number the seats to be filled; they may also indicate that they wish their list to be considered as allied with one or more other lists. Each elector has as many votes as there are seats to be filled, and may distribute them as he pleases among all the candidates nominated, not necessarily confining himself within any one list; he may also cumulate two votes on one candidate. The first operation is to count the votes cast for the candidates of each list and allot seats in proportion to these totals. The seats are then filled by the candidates of those lists in the order of the number of votes they have received. Only in the event of a tie is the order in which the candidates' names appear on the ballot paper taken into account. Thus, control over which persons are elected passes very largely out of the hands of the party organisers into those of the voters.

Luxembourg uses a similar system, but with the important difference that the elector may also cast a vote for a party's list as it stands. A number of other countries also provide for a list vote in addition to a greater or lesser degree of choice within a list. For instance, the Italian voter marks the symbol of the party he prefers and may in addition record personal votes for up to three candidates on that list if there are 15 or fewer seats to be filled, for four if there are more than 15. The candidates to fill the seats are selected in the order of their personal votes, a voter who has marked no individual candidates being deemed to have supported the candidates in the order in which the party has placed them.

Both Sweden and Austria allow the voter to cross out names on a list, but this has effect only if done by more than half of a party's supporters, which in practice never happens. The Swedes get over this absence of effective personal choice by submitting a number of lists whose votes are pooled for the calculation of a party's seats; those Seats are shared out according to which lists get the most support.

It would take a much longer study to outline all the nuances of the party list system in various countries, however, a number of general points can be made. In the first place the list system is found predominantly in non-English speaking countries of Western Europe. It is concerned primarily with insuring fair representation for parties. List systems have never been seriously considered in Canada or indeed in other English-speaking countries. They are often associated with weak or unstable government but proof of this is far from conclusive. Indeed the system is sometimes unfairly dismissed for being responsible for the ills of certain political systems but given no credit for the maturity of others.

B. The Single Transferable Vote

In countries following the British parliamentary system the only kind of proportional representation to receive serious consideration is the single transferable vote (STV). The single transferable vote gives each elector one vote but it is made effective by being passed on as may be necessary from the candidate who is the first choice to the one who is marked second, third and so on. Like the list system the single transferable vote is used in constituencies returning at least three members. Each successful candidate is elected not by a majority but by a quota, which in most systems is determined by dividing the total number of valid votes by one more than the number of seats to be filled. The single transferable vote differs from all list systems, even the most flexible ones in two ways. Firstly, it allows the voter full freedom to express his preference for individual candidates, without regard to those candidates party affiliation. Secondly, it ensures that no vote shall assist in the return of any candidate or any party unless the voter has expressly indicated that he wishes it to do so.

The advocates of single transferable vote are less concerned with the fortunes of parties than with giving greater freedom to the individual voter. This system is designed to make every vote as effective as possible and if voters are guided largely by party considerations the party will obtain proportional representation but this will only be as a consequence of the voters' choice. The primary objective of the vote is to enable each citizen to take part as freely and as fully in the selection of his representative, in the belief that this Is the essence of true democracy.

Studies have shown that the single-transferable vote does produce legislatures in which the strength of parties closely reflects the popular vote. It is also appropriate in pluralistic societies where racial or religious minorities demand representation. Under the first-past-the-post system the only certain way to ensure minority representation is to establish separate electoral roles and separate elections as is done for the Maoris of New Zealand. With the single-transferable vote, however, the minorities can attain representation without any special provisions and without even appearing as a contestant in the election.

The main disadvantage of the single-transferable vote is its complexity even for a sophisticated electorate. The result of the Australian election of December 13th, 1975 was not known until January 15th. Would the Canadian electorate, accustomed to almost instantaneous results stand for such delays? Another difficulty concerns the method used to fill vacancies. If one member of the multi-member constituency resigns or dies it is obviously impossible to hold another election using the single-transferable vote and the question arises as to who will appoint the replacement?

Any discussion of the relative merits of single-member constituency electoral systems versus the various systems of proportional representation is limited by an inability to agree on the values which the electoral system should serve.

"Advocates of proportional representation base their arguments on democratic fundamentalism. The y simply argue that each vote should have equal weight, and that the distortion of the voter's preference by single-member constituency systems is no more to be justified than the use of false scales by a butcher. This idealistic argument is countered by the opponents of proportional representation with the assertion that executive stability is more basic consideration, and that it is well served by the propensity of Canadian-type systems to create artificial legislative majorities."(7)

A major change in the electoral system would mean a change in the way public affairs are conducted. Political values would have to be modified and those who aspire to positions of political leadership would have to learn to use the new system. We are unlikely, therefore, to see a complete change in the traditional first-past-the-post electoral system just as the Italians or Swedes are unlikely to abandon their proportional representation system for ours. On the other hand there are times when it is necessary to look critically at how our traditional institutions operate and if necessary change them.

The Task Force on Canadian Unity has argued that we have come to a point where some of our institutions and practices (such as the electoral system) are working to foster sectionalism and disunity in this country by accentuating local differences and exaggerating the role of specific regions. This also tends to transform party debate into a struggle between areas. Their solution would be to add an additional sixty seats to be chosen in relation to the popular vote accorded parties in each province. This would enable many highly qualified men and women to sit in Parliament without the necessity of fighting an electoral campaign but it would also give them a decided advantage over their colleagues who would have to divide their time between constituency duties and parliamentary work. The reaction of most members is that "it would be grossly unfair to MPs elected by constituencies to create a caste who would not have the same responsibilities sitting in Parliament. " (8) Also, would sixty seats have a great impact on representation in the House? For example, Ontario being the largest province, would receive the most extra seats but there are already members of three parties among Ontario representatives. A smaller province such as Nova Scotia might only be entitled to five or six seats and after allotting them proportionally to the Government and the Opposition very few, perhaps none, would remain for minor parties. Even in Quebec how important, for national unity, would be an extra three or four seats for the Conservative party?

It is rather ironic that the Task Force on Canadian Unity and the independantiste Government of Quebec (9) are the main proponents of some kind of proportional representation. They both recognise that no electoral system will satisfy everyone but they also appear to assume that the system can be modified without clear agreement on the type Of society and institutions it is intended to serve. In the absence of any such consensus, either in Quebec or in Canada as a whole, one wonders if it would be wise to tinker with the present system which despite its failings, does have much practical credibility.


Part II: Reform of the Upper House

Since Confederation there have been many proposals to reform the Senate although extreme suggestions such as abolition or a fully elected Senate have never received much support. Most proposals have addressed themselves to the problem of how to provide more direct and formal expression for the interests of the provinces. In 1969 the Trudeau government proposed the Senate be partly selected by the federal government and partly by provincial governments. The term of membership in the Senate would also be fixed so as to permit changes in provincial administrations to be reflected in Senate membership. (10) Three years later a Special Joint Committee on. the Constitution called for a redistribution of Senate seats giving greater. representation to the western provinces. It also called for one-half of the Senators from each province appointed solely, on the recommendation of the federal government while the other half would be appointed by the federal government from a panel of nominees submitted by the appropriate provincial or territorial governments.(11

In June 1978 the federal government issued a policy statement "A Time for Action" followed shortly afterwards by a constitutional amendment bill. The bill proposed the creation of a House of the Federation consisting of 118 members. One-half of the members for each province would be chosen by the House of Commons following each federal election and the other half by the provincial legislatures following provincial elections. All members would be chosen in such a way as to reflect the political preference of voters following the various elections. This proposal would create an Upper House where the parties would be represented on a proportional basis. Furthermore the proposed House of the Federation would be partially renewable after each federal and provincial election. (12)

The powers of the proposed House would be substantially less than those of the Senate. It would have only a suspensive veto with which it could delay a bill for at least sixty days and not more than L20 days, after which it would be presented for the assent of the Governor General. In addition, if the House of Commons resolved by a two-thirds majority that a measure was urgent it could be referred directly to the Governor General for his assent without further reference to the Upper House providing that seven days have elapsed following its presentation to the Upper House.

The most recent proposal for Senate reform was by a Task Force on Canadian Unity which presented its report in February 1979. It called for the creation of an entirely new Second Chamber to be called the Council of the Federation composed of delegations representing provincial governments and acting under their instructions. The Council would be composed of no more than sixty voting members, to be distributed amongst provinces roughly in accordance with their respective population up to a maximum of 1/5 of the Council but with weighting to favour provinces having less than 25% of the total population. In addition Cabinet Ministers of the federal government would be nonvoting members and would have the right to present and defend central government proposals before the Council and its committees. The Task Force said the Council should not have the power to initiate legislation except in the case of bills proposing constitutional amendments and its decisions should not be regarded as expressions of confidence or non confidence, since the Government should remain responsible to the House of Commons alone. The scope of the powers of the would be as follows:

- Legislation and treaties within exclusive federal jurisdiction would not require the approval of the Council;

- Proposed federal legislation deemed to belong to the category of powers described as concurrent with provincial paramountcy should be subject to a suspensive veto of a longer duration by the Council,

- Proposed federal legislation deemed to belong to the category of powers described as concurrent with provincial paramountcy should be subject to a suspensive veto of a longer duration by the Council, except in the case of measures implemented by lateral agreements between the federal government and one or more provincial governments;

- The ratification of treaties, or parts of treaties, which deal with matters within provincial jurisdiction should require the approval of a majority of the provinces in the Council,, on the understanding that: legislative measures implementing such treaties are to remain within provincial jurisdiction;

- Federal initiatives in areas of provincial jurisdiction that are based on the federal spending power, whether they are to be cost-shared or financed fully from federal funds should require a two-thirds majority in the Council;

- If a province chooses not to participate in a program for which wide provincial consent has been demonstrated, the central government should be required to pay the government of that province a sum equal to the amount it would have cost the central government to implement the program in the province;

- a proclamation of a state of emergency, in either peacetime or wartime circumstances, should require, in addition to confirmation by the House of Commons, confirmation by the Council by at least a two-thirds majority. (13)

Both the Constitutional Amendment Bill and the Task Force proposals are intended to give provincial governments greater input into the central government machinery. On the other hand both proposals would mean an end to the traditional role of the Senate as a Chamber of "sober second thought" and as an investigatory body able to study issues in a more leisurely and less partisan atmosphere than prevails in the House of Commons. The proponents of these proposals argue that the traditional roles of the Senate could be transferred to a strengthened committee system of the House of Commons.


1. Paul Fox, "The Pros and Cons of P.R. for Canada" in Politics Canada, 4th edition (Toronto: McGraw-Hill, 1977), p. 308.

2. Canada, Task Force on Canadian Unity Report (Ottawa: January 1979), p. 105.

3. Neville Johnson, "Adversary Politics and Electoral Reform: Need We Be Afraid?" in S.E. Finer ed., Adversary Politics and Electoral Reform (London: Anthony Wigram, 1975), p. 62.

4. Alan Cairns, "The Electoral System and the Party System in Canada", Canadian Journal of Political Science, Vol. 1, (March 1968) p. 62.

5. There were also five coalition governments in Manitoba and British Columbia in the late 1940s and early 1950s.

6. It could be argued that in practice this system differs very little from the existing Canadian method since many electors vote primarily for party nominee. They may or may not admire the candidate the party has chosen for that particular constituency.

7. Allan Cairns, "The electoral system and the party system in Canada, 1921-1965, Canadian Journal of Political Science, (March 1968), p. 55.

8. Walter Baker, M.P. quoted in the Globe and Mail, March 20, 1979.

9. See Quebec, Ministre d'État à la Réforme électorale, One Citizen One Vote (Québec: éditeur officiel), 1979.

10. See The Constitution and the People of Canada (Ottawa: Information Canada, 1969), pp. 28-34, 76-78.

11. Special Joint Committee of the Senate and the House on the Constitution of Canada, Final Report, 4th Session, 28th Parliament, 1972, pp. 33-37.

12. See Philip Laundy, "A New Direction for Canada", The Parliamentarian, vol. 59 (October, 1978), pp. 211-215.

13. See the Task Force on Canadian Unity, A Future Together, (Ottawa: Department of Supply and Services, 1979), pp. 128-12.

Canadian Parliamentary Review Cover
Vol 2 no 3

Last Updated: 2020-03-03