Canadian Parliamentary Review

Current Issue
Canadian Region CPA
Upcoming Issue
Editorial and Stylistic Guidelines

HomeContact UsFrançais

The Evaluation of Electoral Systems
Marie Lavoie; Vincent Lemieux

At the time this article was written Vincent Lemieux was professor in the Political Science Department at Laval University. Marie Lavoie is a former legislative intern with the Quebec National Assembly.

The question of electoral reform is once again on the political agenda in Quebec. Since 1966 when the Union nationale won the election with less popular vote than the Liberals, the single district majority system (also called first past the post) has been the subject of much criticism.

At the beginning of the last decade, Robert Bourassa, then Premier announced establishment of a Committee of the National Assembly to study the matter. The Committee came out in favour of the status quo but in 1972 a group with a mandate to present alternatives put forth three methods of electoral reform.

The subject lie more or less dormant until the Parti québécois came to power in 1976. A cabinet minister, Robert Burns, was given specific responsibility for electoral and parliamentary reform. Three years later the government presented a Green Paper (One Citizen-One Vote). The fall of 1981 Premier Lévesque said the government was considering introducing a bill on electoral reform. No action was taken until June 1983 when the National Assembly approved creation of a Committee on Electoral Representation consisting of Professors Guy Bourassa, Marc-André Lessard and chaired by the Chief Electoral Officer, Pierre-F. Côté.

Its mandate is to evaluate various proposals for electoral reform. In October 1983 it began province-wide hearings designed to hear all interested groups and report, by February 1984, on the kind of electoral reform which seems to have the greatest support in the province.

Criteria for Evaluating an Electoral System

The parliamentary system, despite its long tradition, is a fragile thing. A change in any of its components, such as the electoral system, may transform the whole in an unforeseen manner. This should not exclude a search for better ways to adapt it to a changing society. Parliament must, to a certain degree, remain open to change and deal with it. Such flexibility ought to be built into the normal process of adjustment to political evolution. To evaluate the introduction of new measures or new features into the parliamentary system, clearly stated criteria are needed. We wanted to identify criteria that can be used to direct any such reform by taking into account various aspects of representation and government.

An electoral system is a mechanism used in a variety of more or less homogeneous and stable societies. It is not enough to blindly adopt a given electoral method. Its success in other countries is no guarantee of success in Quebec. The adoption of any system must take into account both the society from which it is borrowed and the one into which it is to be transplanted. A number of criteria should be defined for use in evaluating what system would be best suited to our type of society and our style of politics.

Two groups are chiefly affected by the electoral system: the government and the electors. Any analysis of the advantages and disadvantages of an electoral system should be guided by its effect on both those who govern and those who are governed. Such analyses have usually been restricted to the formal aspects of representation and government. The formal aspect of representation is equity, which we define as the relationship between the percentage of votes obtained and the percentage of seats won in the legislature. The formal a3pect of government is stability, which corresponds to the capacity of the government party to govern without fear of being defeated.

These two criteria equity and stability are usually the basis of evaluating electoral systems. According to them an ideal electoral system allows for a measure of appropriateness between the percentage of seats and the percentage of votes. It also generates a stability that allows the government to distance itself from everyday matters to concern itself with broader policy matters. But representation cannot be reduced to simple mathematical equity. It must also consider the quality of the representative's role. For its part, government needs a measure of accountability as well as of stability. An electoral system must therefore create a situation that fosters quality as well as equity and accountability along with stability.

The challenge is a difficult one because any attempt to meet one of these objectives brings a risk of adverse effects on the others. There may be no such thing as an ideal system but in a given context some electoral systems meet the criteria of both representation and government better than others. What is needed then is a proper combination of the various factors that make up the electoral system as a whole.


To fully meet the equity criterion an electoral system should allow every citizen's vote to be equal to the vote of every other citizen and represent every citizen equally in Parliament. In practice, Parliament cannot be expected to mirror society exactly because there can never be complete correspondence in terms of sex, age and various social conditions between representatives and their constituents. This is where the quality of representation comes into play.

Equity also means that political parties should obtain a percentage of seats that is approximately equal to the percentage of votes cast for them by the electorate as a whole. This concept of equity is based on the theory that in a pluralist society it is incorrect to think the winning candidate in an electoral district can represent all the voters. Contrary to Rousseau's view, losers do not see themselves represented by the winner when minorities are not prepared to rally to the "General Will" as expressed by the majority. The parties, at least the major ones, must be represented in each riding and in the society as a whole, more or less in proportion, to the percentage of votes received.

The quality of representation does not necessarily mean that parliamentarians must be like their constituents in every respect in order to represent them properly. Depending on the electoral system, the member of Parliament tends to become either the servant of his constituents or the defender of a doctrine, freed from the constraints of a district. The specific nature of this role is largely determined by factors such as the size of the electoral district and the number of seats representing the district. The manner in which candidates are chosen, partly by the party and partly by the population, also affects the quality of the role assigned to the representative. Such factors affect voter representation, and that is why they must be taken into consideration in any evaluation.


While a degree of stability is required for government, it cannot detract from the requirements of accountability, i.e. the need to be answerable for action taken. These two criteria may appear to be opposed to one another, for a party which achieves stability through a decisive victory or through the nature of the electoral system acquires the latitude to ignore parliamentary accountability. What is needed is a balance which, while rewarding the victorious party with the freedom needed to govern, nevertheless subjects it to a measure of control.

Although some controls must be exercised outside Parliament or improved by means of a greater degree of access to information, the electoral system may facilitate control by the opposition as well as by backbenchers of the government party or parties. It is up to the opposition to force the government to account for its administration and to defend the efficiency of some of its programs, but in order to be able to do so, it must have the appropriate means. In a two-party system in which partisan lines are clearly and rigidly drawn and in which the parties are highly structured and disciplined, we cannot expect to find Austrian style co-operation among parties. With virtually no coalition tradition, our parties are much more prone to confrontation than co-operation. Thus daily question period and the Friday morning question and debate, quickly become partisan in tone and the government rarely gives in to opposition pressure.

At the policy level there have in recent years been some special measures to attempt to strengthen the role of the MNA and once again to reassert the importance of the legislature. There was, for example, the revision of the Elections Act, the re-organization of the parliamentary timetable, the introduction of radio and television broadcasting of the debates and procedural changes such as the method of putting a question with debate. In addition, there have been recommendations of the Vaugeois report1 and the many suggestions made by Claude Forget2 concerning the strengthening of parliamentary control over government administration. An electoral system should encourage such reform. In this connection, two features of the system are relevant: the local and regional basis upon which members are elected and whether or not the electoral system gives them exclusive representation of the electoral district.

In the current context we would do well to ask ourselves seriously whether accountable government might not be better than stable government. It is no longer possible to govern with the same methods that were used in the sixties and seventies. Accountability is more than a way of doing things it is a necessity.

As the Lambert report3 made clear, Parliament is, to all intents and purposes, the Alpha and Omega of accountability. We ought not, however, to expect a revolution because some electoral systems allow for greater accountability than others. Obviously no mere system can by itself immediately improve politics; such methods are no panacea. What they can do if they are well-suited to a specific society, is prevent certain types of abuse.

Application of the four criteria to three electoral systems

The Ouebec Government's Green Paper, One Citizen, One Vote published in April 1979 outlined three possible methods of proportional representation. One of these systems is the compensatory mixed (50/50) system which is used in the German Federal Republic. It is so named because one half of the members are elected by a simple majority in single constituencies while the other half are chosen from party lists in proportion to the overall vote cast for each party. It is compensatory in the sense that those chosen from lists are done so in a way as to "correct" imbalance in the results of straight voting.

The non-compensatory mixed (2 3 1,2) system provides for a list of members to be elected in proportion to votes cast but this calculation is done independently of results of the straight majority elections. The Green Paper put forth a third possibility: a system of weighted regional proportional representation (WRPR). In this system there would be only one kind of election. The province would be divided into regions and each would elect a number of members in proportion to votes cast within that region.

Each of these electoral systems has been rated numerically according to certain criteria on a scale from one (that which best meets the criterion) to three (that which meets it least). We believe that as far as representation is concerned WRPR and the compensatory mixed system are preferable to the non-compensatory mixed system. For government, in our opinion, WRPR also comes in ahead of both others with the compensatory mixed system second and the non-compensatory mixed system third.

The compensatory mixed system provides the highest degree of mathematical justice. Its compensatory features considerably reduce the distortions of representation inherent in the majority ballot, Although it is highly equitable, it creates two categories of legislators and could lead to confusion between their roles. In addition, there would be twice as many members in the National Assembly.

The non-compensatory mixed system, in addition to creating two categories of legislators, introduces them in a very unequal manner as a result of the lack of a compensation feature in the formula and the number of MPs assigned under each formula. The possible confusion of roles is therefore even greater here than it is for the compensatory mixed system. Since the effects of the majority ballot are only partly corrected, it is to be expected that representation distortions would remain, although somewhat reduced by the second series of members elected by proportional representation.

Even though it may not be as equitable as the 50/50 compensatory mixed system, WRPR does rather well in terms of equity, distortion between the percentage of seats and the percentage of votes is fairly low. The advantage of this system over the two others is that it requires one category of representatives in the National Assembly, and this could well mean a better quality of representation.

Any system that maintains a high degree of equity risks a lower degree of stability. This is precisely what happens with the compensatory mixed system. Although it scores highest in equity, it is very poor in terms of stability because it does not reward the winning party enough. Moreover it discourages minor parties because of its minimum level of 5% needed to elect members. If this minimum were lowered by a few points, however, there would be better opportunities to form a minority government. which would result in a measure of accountability and an increase in parliamentary control.

As with the majority ballot system, the non-compensatory mixed system is not very equitable in terms of the seats/votes ratio. It therefore has a high level of stability because of the importance given to the victorious party and the absence of fragmentation of the partisan system. However these features make real accountability difficult to achieve.

Weighted Regional Proportional Representation does not require two types of electoral systems and breaks the electoral map down into regional electoral districts. The reduced size of the regional districts combined with the small number of representatives elected for each of them, prevents excessive fragmentation of the partisan system and ensures enough of a margin to the victorious party to allow for a reasonable degree of stability. The regional nature of this system and the small number of extra seats that go to the winning party could be factors that ensure greater government accountability.

Notes1. See Denis Vaugeois, "L'Assemblée nationale en devenir, pour un meilleur équilibre de nos institutions," Quebec, National Assembly, 1982.

2. Former Minister of Social Affairs in the Bourassa Government and Member of the Opposition until 1982.

3. See the Report of the Royal Commission on Financial Management and Accountability, Ottawa, Minister of Supply and Services, 1979.

Canadian Parliamentary Review Cover
Vol 6 no 4

Last Updated: 2020-09-14