Special Topics

Alternative Voting Systems
May 14, 2011

Recently, we had a federal election here in Canada, and many people noticed that when all was said and done, the distribution of seats in the House of Commons didn’t very closely match the distribution of votes that each party received. There is a good reason for this, and it has to do with the voting system we use here in Canada, a system known as “First Past the Post” or “Single Member Plurality”. Given the recent election here in Canada, as well as the fact that people in the UK recently engaged in a referendum about whether they should change their voting system (they opted not to), we thought our readers might be interested in a brief overview of how our system works, what some of the alternatives are, and the history of alternative voting systems in Canada.

Single Member Plurality

Here in Canada, we use a system based on the Westminster parliamentary system found in the United Kingdom. The Single Member Plurality system (also known as the “winner take all” system) has the advantage of being very easy to understand. Basically, Canada is divided up into 308 geographical areas (known as “ridings”), and voters in those areas cast a ballot for the candidate that they feel would best represent them. The candidate in each riding who garners the most votes is declared the winner and becomes a Member of Parliament. There are a number of advantages to this system, not the least of which is that it is very simple. People cast one vote, and that vote is directly accounted to a single candidate. Moreover, it allows people to vote for a person who is explicitly responsible for the area that they live in, giving a sense of direct representation.

There are, however, some downsides to this system. Because a candidate only has to get more votes than every other candidate to win, the winner doesn’t necessarily have to get the support of the majority of the voters in their riding. Here is an example from the 2011 election:

ConservativeJohn Baird25,22644.7
Green PartyMark MacKenzie2,2794.0
NDP-New Democratic PartyMarlene Rivier11,12819.7
LiberalAnita Vandenbeld17,79031.5

Notice that Conservative candidate John Baird received the support of less than half of the voters in the riding, but because he received a plurality of votes (more votes than anyone else) he was declared the winner and will represent all of the voters in the House of Commons.

While this might seem somewhat unequal on a local level, the difference is even more pronounced if you look at the results across the entire country. In our political system, the party that wins the largest number of seats is generally the party that forms the government, especially if that party wins a majority of the seats, as the Conservative Party did in 2011. But, just as a candidate can win a riding without the support of the majority of voters, so too can a party win a majority of the seats in the House of Commons without the support of the majority of Canadians. In the 2011 federal election, the popular vote each party received is as follows:

Green Party: 3.9%
Conservative Party: 39.6%
Liberal Party: 18.9%
NDP: 30.6%
Bloc Quebecois: 6.0%

But the actual percentage of seats that each party won was very different:

Green Party: 0.3%
Conservative Party: 54.2%
Liberal Party: 11.0%
NDP: 33.1%
Bloc Quebecois: 1.3%

In fact, if the seats in the House of Commons were distributed proportionally to the votes that each party received, the House would look very different:

Image by Bryan Beca. Click for full size

Alternative Voting Systems

So, what are some alternatives to the “Single Member Plurality” system, and what are their advantages and disadvantages? Well, there are a number of popular alternative systems in use around the world. Some of them are directly proportional, and some of them are just “more proportional” than the system we currently use. I’ll talk about two systems in particular, as these systems have a history of being considered in Canada and the United Kingdom, and are in use around the world.

Single Transferable Vote

The single transferable vote system is currently in use in Ireland, and a similar system, known as BC-STV was proposed and voted on in two referendums in British Columbia. Ultimately, the people of BC decided not to adopt this method of voting, but it is still a very popular choice.

An STV system has a number of differences from a Single Member Plurality system. The biggest difference is probably that rather than many small riding represented by a single Member of Parliament, an STV system has fewer, larger ridings that are represented by more than one Member of Parliament. As I will explain later, this arrangement allows for more proportionality in the results.

In an STV system, voters cast a single vote, but that vote can be transferred to other candidates if their preferred candidate doesn’t win. Voters transfer their vote by ranking the candidates according to their preference. Voters can choose to rank as few or as many of the candidates as they like, but they cannot give any candidates the same rank. As well, parties can run as many candidates in a riding as there are seats in the House of Commons for that riding. For example, let’s say Riding A is represented by 5 seats in the House of Commons. The Liberal Party, the Conservative Party and the NDP (for example) could each run up to 5 candidates in Riding A. The voters of Riding A would then see a ballot that would list all 15 candidates, and they could then mark a number next to some or all of those names in order to indicate their preference. So a voter might mark a 1 next to Candidate A, a 3 next to Candidate B, a 2 next to Candidate C, and so on.

Now, this is where things get a little more complicated, but hopefully I haven’t lost you yet! Once the votes are cast and voters have recorded their preferences, the first choice votes are tallied up. Let’s say, for example, that there are 4 candidates in Riding A and 40% of people put Candidate A as their first choice, 30% put Candidate B as their first choice, 20% put Candidate C as their first choice, and 10% put Candidate D as their first choice. Let’s also say that Riding A has two seats in the House of Commons, so two of the candidates can win. In an STV system, a formula is used to calculate a “quota”, or minimum number of votes required to win a seat, and no one wins until they have reached that number. So if candidate A reached the quota after the first round, he or she would be elected, but candidates B, C, and D would still need to try to get more votes in order to win the other seat. So where to those new votes come from? Well, they come from the surplus votes of Candidate A! You see, if the quota to be elected in Riding A was 1,000 votes, but Candidate A received 1,500 votes, then he or she would have 500 surplus votes. Those 500 votes are then redistributed to the remaining candidates based on who Candidate A’s supporters indicated was their second choice. This process continues until all of the seats in Riding A are filled. Confused? Don’t feel bad, it’s a much more confusing system, and that is one if its main drawbacks. Here is a video that makes things a little more clear:

There are arguably a number of advantages to the STV system over the current Single Member Plurality system. The first is that it does a much better job of not “wasting votes”. Because more votes are used to help elect candidates, voters don’t necessarily become disillusioned by the fact that their vote may not have had an influence on who their representatives are. As well, because there are multiple people elected in each riding, the system is more proportional than a Single Member Plurality system, although it is not perfectly proportional. The downside is the obvious complexity involved in such a system. It can be very difficult for voters to understand, and because people are not necessarily sure how their vote will be counted, they may opt out from participating. There is also some regional basis for representation, but the larger ridings with multiple Members of Parliament means that the person representing voters may not necessarily be from the same geographic area as the voters.

Mixed Member Proportional Representation

The second type of voting system I’m going to talk about is called “Mixed Member Proportional”, or MMP. An MMP system tries to take the best aspects of a Single Member Plurality system, while adding proportionality through what is called a “Party List”. Ontario held a referendum on electoral reform in 2007, and the people voted on whether or not to adopt an MMP system in that province. Ultimately they opted to stick with the current system.

In an MMP system, every voter casts two votes, one for a local candidate and one for a party. Candidates are elected in their ridings by receiving more votes than any other candidate (a plurality), just as in the Single Member Plurality, or First Past the Post system. As we saw earlier however, this often results in an electoral outcome that isn’t particularly representative of the actual proportion of the votes that each party receives. So how does the MMP system solve this problem? Well there is a second way that candidates can be elected to the house, and that is by having their name on the “Party List”. Each party creates a list of candidates that they would like to see elected. At the end of the election, the party votes are tallied up, and those results are used to determine how many seats each party should have. So, for example, let’s say that in an MMP system the Liberals received 40% of the party votes, the Conservatives received 35% of the party votes, and the NDP received 25% of the party votes. That means that each of those parties should receive that proportion of seats in the house. So if the House of Commons has 100 seats, the Liberals would receive 40, the conservatives 35, and so on. Let’s say however, that the Liberals only won seats in 30 of the ridings. In that case, the party list is used to “top up” their seats. So if the Liberal Party had 20 candidates listed on their Party List, then the first 10 candidates would be elected to the House of Commons in order to bring their total up to 40 seats.

The system is a little bit more complicated than that, but that is the basic idea. The total number of seats often has to be increased a little bit each election in order to make sure that a mostly proportional result can occur, and most systems also have minimum “quotas”, or percentage of the vote that a party has to achieve in order to be eligible to receive party list seats. The quota is important because it keeps many smaller fringe parties from winning one or two seats just because they were able to gain small amount of support from a broad geographical area. The MMP system is currently being used in a number of countries, including Germany, New Zealand, and the UK in the Scottish Parliament.

Here is a video that gives a clear explanation of the version of MMP that Ontarians voted on:

There are, of course, many different types of voting systems, but the systems outlined here are some of the most popular. They all have their advantages and disadvantages, and it is up to the voters in every country to decide which system works best for them.