Contact UsSearch
Click for Search Instructions
Home > Get Involved

Submission JOBLIN-0399 (Online)

Submission ByMr Christopher Joblin
AddressWhite Rock, British Columbia, Canada
Organization
Date20040514
CategoryElectoral system no change
Abstract
'Trading One Evil For Another: Electoral Systems and Justice':  The submission reviews the variety of considerations involved in electoral change and argues that 'the devil we know is better than the devil we don’t'. [7 pages]

Submission Content
[Too see the submission in its original format with endnotes and works cited, see the linked document below]

Trading One Evil For Another: Electoral Systems and Justice

In today’s society, citizens of democracies take it for granted that from time to time they will go to the polls to elect representatives to act on their behalf when the government convenes to conduct the business of the country.  While this act may be a given in democracies, what is not so certain in the manner in which these representatives will be elected.  Schemes of representation include single member plurality (SMP), a hybrid proportional representation system, a party list system, to name a few.  The method by which representatives are picked greatly affects the composition of the government.  In the most recent provincial election, the BC Liberals received slightly less than 60% of the popular vote and yet they were able to win 77 of 79 seats in the legislative assembly.  Some would say that this is a “just” way to elect representatives because it gives equal respect to very populated areas such as Vancouver and to less populated areas such as Northern British Columbia.  Conversely, others would contend that it is “unjust” that it is possible for a small majority of the population would be able to effectively edge out any opposition in the legislative assembly.  The issue of justice is central to the creation of electoral system; unfortunately, political theorists have devoted surprisingly little time to answering this question.  It is important to preserve the idea of majority rule, but it is simultaneously necessary to control the majority so that it does not develop a tyrannical element.  Deference to minorities is a hallmark of justice and for an electoral system to be “just” it must embody this principle.

Before examining the various types of electoral systems, it is necessary to first define justice.  Political theorists when speaking of distributive justice often use the term justice, but I am using justice in the context of political equality.  An acceptable theory of justice must include some notion of majority rule while still seeking to protect the rights of minority groups.   John Rawls deals with the idea of majority rule and justice in electoral systems in §54 of A Theory Of Justice.  Rawls declares that majority rule is not bad in and of itself, but that it must observe his aforementioned principles of justice in order to be just itself.  However, he also states that

There is nothing to the view, then, that what the majority wills is right.  In fact, none of the traditional conceptions of justice have held this doctrine, maintaining always that the outcome of the voting is subject to the political principles.  Although in given circumstances it is justified that the majority (suitably defined and circumscribed) has the constitutional right to make law, this does not imply that the laws enacted are just.  The dispute of substance about majority rule concerns how it is best defined and whether constitutional constraints are effective and reasonable devices for strengthening the overall balance of justice.

Rawls’ discussion with regard to majority rule highlights an important aspect of the issue of electoral systems, justice and majority rule, namely that there are many factors at work that can contribute to how just the government’s laws are.  Given the lack of adequate space to deal with all these issues it will be necessary to, for the time being, look away from the fact that any electoral system can produce an unjust outcome and focus on the system that, given a strong constitution to protect minorities, will be more likely to produce a just outcome.

When speaking of electoral systems and justice, the key aspect in a limited examination of the subject will hinge upon considerations of the ability for a group to be marginalized by a majority.  A just electoral system will strive to ensure that all individuals in the country have a voice that can impact the course of the elections.  The most basic notion in Rawls’ theory of justice is that majority rule is not necessarily just simply because the majority agrees with the policy.   Democracy does not mean 50%+1 has the legitimate authority to do whatever it chooses.  Any consideration of democracy must include a notion of respect for minority rights.

There is a variety of electoral systems in use throughout the world with varying degrees of success. Our concern is not with the effectiveness of each system, but rather how just they are, in a purely theoretical sense.  Thus, in the initial examination of each electoral system it is of no concern how stable the government would be that is rendered by the electoral process.

There are essentially two different types of electoral systems, with many sub-types of elections.  All electoral systems can be described either as a plurality election system or as a proportional representation election system.  Under the plurality electoral systems would fall single-member plurality systems and at-large systems.   There are many more valid types of proportional representation; among them are a party-list system, additional member system, and the single transferable vote system.

The first system that we will be dealing with is single-member plurality (SMP).  This system is also sometimes called “first past the post” because the person who receives the most votes win (a majority is not needed to win).  SMP is a very simple type of election system whereby the country is divided into single-member constituencies and where voters select a single candidate who needs to achieve a plurality of votes.   This simplicity is in many ways its biggest fault; it attempts to oversimplify the complex task of electing officials.  By playing to the most basic notions of what it means to be democratic (majority rule) SMP fails spectacularly.  Often it produces governments that deviate significantly from the political orientation of the society at-large.   This system is the system that Canadians are most familiar with; Canada, the United Kingdom and the United States of America are examples of countries that employ this system.

The second type of plurality voting is the at-large system.  This method of voting is very rarely used on a large scale, though it is sometimes employed at the local level.   In this method, a voter can vote for multiple people on the same ballot.  For example, six people may be running for three positions; whichever three people get the most number of votes is elected to the position.  This system of electing representatives is very rarely used, though it will typically produce results similar to a SMP election.   As such, it is subject to the same criticisms that are levelled against the SMP system as listed above. 

Majoritarian systems are not really a type of plurality system, but they are also not a proportional system.  There are two types of Majoritarian systems, a second ballot system and an alternative vote system.  The second ballot system is often used at lower levels of government, but not usually at a national level.   The central feature of this system is, as its name implies, the use of two ballots; David Farrell states, “the principal objective is to increase the likelihood that the candidate elected will have an overall majority of support in the constituency, i.e. more than 50 per cent of the votes cast.”   On the first ballot if an individual receives more than 50 percent of the votes cast then they win and there is no need for a second ballot.  However, if a majority of voters do not cast ballots for a given individual then it is necessary to move onto the second ballot.  There are two different versions of the second ballot system, in the first variation all candidates who reached a given threshold (often set at about 12%-15% of votes cast) are put on the second ballot.  This variation does not ensure that there will be a Majoritarian result.  In the second variation of this system, only the top two candidates are put onto the second ballot.  This second version guarantees that there will be a majority of voters who elect the representative.

The alternative vote system is very close to the second ballot system in principle but it has a slightly different application.   All candidates are listed on a ballot, voters are then asked to rank all candidates in the order of their preference.  All the ballots are then counted using all the first ordinal choices of the voters.  If a given candidate receives at least 50 per cent of the votes cast then the winner is declared and the counting process is over.  However, if no candidate receives 50 per cent of the votes cast then the candidate who received the least number of first ordinal votes is dropped from the list and all the second ordinal choices are counted on those ballots.  This system continues until a candidate receives at least 50 per cent of the votes cast.

All Proportional Representation (PR) systems share a number of common characteristics.   First, they all have multimember districts: multiple members seeking multiple seats in given electoral districts.  Second, candidates do not need to get a plurality of votes to be elected.  Third, all proportional representation systems endeavour for proportionality in distributing contested seats.   Proportional representation systems have many variations, though they all attempt to reach the same goal, a more even share of electoral seats to the political parties of a given country.   The aforementioned criteria allow us to categorize these numerous variations of the proportional representation so that we can distinguish these systems from other types of electoral systems that attempt to fulfill different goals.   It is possible to accomplish the goal of proportional representation through a variety of forms that still adhere to the principles of PR.

The most common method of PR used in Western democracies is the party-list system.   Voters typically cast their ballots for a particular party, rather than a certain candidate.  For example, in a ten-person district if party A gets 50% of the vote then 5 of their members are elected to the assembly.  There are two types of party-list systems; similarly if parties B and C get 30% and 20% of the vote respectively, then party B would receive 3 seats and party C would receive 2 seats.  In order to keep fringe parties out of the legislature there is often a threshold which must be breached in order for a party to receive any seats, this threshold is often set at somewhere between 3%-5% of the popular vote.  There are two variations of the party list system.  The first is called a closed list system whereby the party ranks candidates from one to X (X being the number of seats in the assembly).  The second system is called an open list system.  In this system voters can change the order of candidates on the party list system.  This gives voters greater control over who is elected to the assembly, “votes for individual party candidates are totalled, and the ones with the most votes are moved toward the top of the list, where they are more likely to be elected.”   In this second system votes count for the party, but they also serve to rearrange the list of candidates.

The additional member system is a hybrid system that attempts to combine SMP system with a party-list system.  In these systems, voters are often given two ballots, on the first ballot the voter selects who they want to represent their constituency, on this ballot whoever receives the most votes wins, similar to SMP.  On the second ballot they select the party of their choice.  This ballot is used to guarantee that all parties get their proportional share of the legislative seats.  Half the seats are reserved for the district winners, while the second half is used to “top off” the legislature and ensure that the parties receive a number of seats proportional to the number of votes they received on the second ballot.  Usually in these systems, as in the party list system, there is a minimum reserve that must be met.   This system is advantageous in that it ensures that a given constituency will have a certain representative to speak for them, while still being “sufficient in preventing disproportional results at the nation level.” 

The final, and most complicated, type of proportional representation is the single transferable vote (STV).  While not used very often in national elections, political parties often use this system of voting when conventions are used to select the leader of the party.   Essentially, the candidates are listed on the ballot in randomized order and the voter is asked to rank the candidates from one to X.  If on the first count a candidate breaches the threshold then they are elected to the assembly and “any surplus ballots beyond that threshold are redistributed to the next available preferred candidate indicated on the ballot.   However, if no one breaches the threshold then the person who got the least number of first ordinal rankings is dropped from the ballot, and the second choice is counted on the ballots where the dropped candidate is listed as the first choice.  This goes on until all the seats are filled.  Similar to the party list system, “officials in the single transferable vote system are elected in multimember districts, and parties put up a number of candidates for these seats.” 

Each of these electoral systems has positive aspects as well as negative consequences.  It is not possible to look at any of these systems and declare that it is the most just system of electing representatives, as there are many factors that can influence this claim which are outside the scope of electoral systems.  No electoral system will work the same in each country because there are so many dependent variables present so as to skew the outcome in one direction or another.   In order to examine the relative justice of each system it will be necessary to look at a variety of circumstances so as to judge when a system will produce just results, and when a system will produce unjust results.

In Canada, we are most familiar with the SMP system, even though it is not a system that is utilized globally to a great degree.  The biggest impediments to justice with regard to the SMP system is that a party can have a small majority of support spread evenly among the constituencies and can win all or many of the seats.   This is most clearly seen in the last British Columbia provincial election where the Liberal government won 77 of 79 seats despite only winning a small majority of the popular vote.  If we consider justice to be simply a majority voting in a particular way is just no matter what the outcome then this would be an acceptable electoral system.  But, given out earlier conception of justice we must conclude that the SMP can produce very unjust results.

Edmund Burke’s support for the SMP system stemmed from his desire to have a legislature that was an “image of the feelings of the nation’.   However, it is undeniable that “the ‘image’ is usually like that seen in a distorting mirror: it is true that every feature of the reflection corresponds to something in the original, but one feature may be exaggerated out of all proportion, while another… becomes scarcely perceptible.”   The fact that one portion of society, the voters who voted for the loser, are not represented in parliament in anyway is a disservice to democracy and an example of the injustice in the SMP system.   However, according to our earlier definition of justice, it is not necessarily unjust to not have all of society represented; it is the fact that it is possible for the government to ignore the wishes of a significant minority, because those citizens will not form a significant voting bloc so as to ensure that the government could not be re-elected.   Moreover, the high disproportionality of the SMP system indicates that small parties are under-represented. 

It is of no concern that SMP does not uphold the ‘one member, one vote’ principle because our concern is not with the justice of the system per se, but rather the potential outcomes of the system.  As a result of not having to be accountable to all members of society SMP fails the justice test, but this is because there is a potential for unjust outcomes, not because the system does seems to be unjust in its composition.

The greatest virtue of the SMP system is its stability and its ability to produce governments that can really change things in government.   SMP systems usually have the effect of producing majority governments, these are the type of governments can produce the greatest amount of change in a country.  The positive outcome of this, that being the ability to produce a government which is not impotent due to the fact that it is a coalition government, is a positive, just aspect of the SMP system.  The outcome can produce more just results than a Proportional Representation system.

The main critique of the party-list system, which does not also apply to other types of PR, is that it removes the constituent-representative relationship from representation.   This means that areas with little population density could have no advocates in the assembly for them.  This could effectively cause sparsely populated areas to be written off by government, as they would have little chance of impacting elections.  For example, in Canada it would be strategically important to have many candidates from areas that have a high population density, such as Ontario, Quebec and British Columbia.  It would likewise be unimportant for political parties to focus on less populated areas, such as the Maritimes and the Prairies (Manitoba, Saskatchewan and Alberta) as these voters would not have much of an influence on the distribution of seats in a party list system due to the fact that the geographical distribution of the population would allow for the minority of voters in these areas to be marginalized.

Moreover, this system has the potential to create some very unpopular governments.  It is possible to imagine a scenario whereby if there are two strong, popular parties that dislike each other a great deal, and there is a third party which is not offensive to either group, but has little support, the third party could be elected because of they would be the first choice of the people who supported them, and the second choice for everyone else as they are regarded as they are regarded as a lesser evil by each of the strong, popular parties.  Little is done to advance the cause of justice by putting in a party that is preferable to very few people.  The same criticisms apply to this system, that could be applied to the party list system, namely that there is no relationship between the constituent and the representative.   This system effectively renders areas that are sparsely populated to be ignored by the system, as they are politically irrelevant when dealing with elections.

Kent Weaver asserts that any sort of proportional representation system would be a resounding failure if implemented in Canada.   It is a fact that in any system that utilizes proportional representation to any degree will ensure that medium sized parties get an adequate, though not altogether large allowance of seats.  Due to the sheer geographical size of the state, there are regional specific parties, such as the Bloc Quebecois that would have a disproportionate amount of power in a proportional representation system.   Weaver asserts that these parties will tend to receive a moderate share of seats, and as such, will become prime candidates to join a coalition government.  This would become a very problematical because specific regions would still be asserting greater influence in governing policy.  For this reason, the system would again become unrepresentative of the population, though in a way that is different, although not any better or worse, from the SMP system.  Again, we see that a proportional representation can produce unjust results, in a manner similar to a SMP system.  While the example made here relates to Canada federally, it is possible to imagine similar scenarios coming to fruition in British Columbia.

There are a number of other critiques that are made of the proportional representation system, the largest dealing with the claim that SMP systems encourage moderate parties, while PR promotes extremism.   The reasoning behind this view is two part.  The first part deals with the issue of moderation in Single-Member Plurality systems, it is contended that

In a two-party system the parties will tend to concentrate their efforts on recruiting uncommitted voters from the middle of the political spectrum, hoping to gain enough supporters to build an electoral majority.  In order to appeal to such voters, both parties will tend to converge on the political center, offering moderate policies to woo centrist voters.  The tendency toward moderate, centrist politics is seen as a characteristic that enhances stability and quality of democratic regimes.

The second part of the equation has to deal with the extremism of the Proportional Representation system.  Ferdinand Hermens contends that because Proportional Representation allows the representation of small parties, extremist political groups are able to gain a political foothold in the legislature that they can build upon.   It is through this foothold in the legislature that they are able to use the forum to increase their appeal and legitimacy.  If fringe parties are allowed to grow from within the system it is possible that they could produce unjust results as they pursue policies which are damaging to certain groups, extremist fringe parties often have politicise which will be detrimental to certain groups in society and as such must be not be allowed to make great gains in the democratic arena, lest justice be thrown out the window.

The discussion of election systems and their relative pros and cons is very confusing for good reason.  When such complex systems are being examined it is difficult to investigate the different systems are produce one system which is better than the rest.   Each system is good and bad in its own right and could produce just results in certain circumstances; conversely, it is also possible for each system to produce unjust results.   Furthermore, it is impossible to examine electoral systems in a vacuum and it could be beneficial to look at the broader picture.  For example, in Canada’s Parliament if there was both an equal, effective and elected Senate as well as an elected House of Commons the geographical problems that are created with elections to the House of Commons by way of a SMP system would be alleviated.  In this sense, it seems logical that the best we can hope for is the implementation of two systems that are equally just and unjust, but in opposite ways, to counterbalance each other.

Further, each electoral system can produce unjust results certain circumstances and as such must be tailored to the country in which it needs to be used.   Countries such as Canada that have specific regions that have a very uneven population distribution among a small number of regions face the greatest difficulties when analyzing electoral systems.  Often by changing systems, we are only substituting one type of injustice for another.   This is because if something like a Triple-E is implemented it would ensure representation for all the provinces.  However it would also be unjust to give more power to these regions because then bloc voting could occur that could marginalize the majority of the population in Ontario and Quebec, and thus once again unjust results would be produced by a different electoral system.  A country that has many distinct regions, or state-like entities (such as the United States), or a country that has equal population density among a few regions would have an easier time designing an electoral system that would work.  Each electoral system has the potential to be just; the difficult issue at hand is under what circumstances they will be just.  Often there are too many unknown variables present to be able to properly identify what the outcome of a particular electoral system will be until it is in use in a country.

British Columbia suffers from much of the same problems as Canada, some areas of the province have a significant population, but are small relative to the population of the lower mainland.  Consequently it would be imagined that these regions would receive less attention in a PR electoral system as they do not have the same power they once did.  When these regions once had a MLA to a rather small population (compared to the residents to MLA ratio in many lower mainland ridings) they will now have to fight with the whole of British Columbia for political attention.  In PR systems, regions with concentrated populations receive disproportionate political attention.

As we have seen, there is a great deal of controversy about what would occur if a given electoral system were implemented in any country.  It is for this reason that we cannot, as earlier stated, examine these systems in a vacuum absent other factors that could influence the just or unjust outcomes of the system.  Given that there is this controversy, and often conflicting views not only about what is good and bad, but also what would occur or not occur, it is necessary to look at what other factors encourage the just or unjust results that can be a product of electoral systems.  It would be ignorant to only look at half the equation.  The only certainty is that no electoral system is just or unjust in and of itself.  Thus, while it may seem like a good idea to change our electoral system we must be reminded of the idiom “a bird in hand is worth two in the bush.”  Most certainly, the devil we know is better than the devil we don’t.

[See the linked document below for endnotes and works cited]

Related Links
DetailsWord DocumentJoblin0399 - Trading One Evil for Another
DetailsAcrobat PDFJoblin0399 - Trading One Evil for Another
© 2003 Citizens' Assembly on Electoral ReformSite powered by levelCMSSite Map | Privacy Policy