Friday, June 10, 2016

Electoral Reform - the 5 major proposals

There are 5 major proposals for electoral reform in Canada, as I see it, and I will detail them, below.

First Past The Post
The default and status quo option is to keep out current system. For those not in the know, FPTP works as follows:

The country is divided into smaller chunks known as "ridings" that are roughly equal in population. Each riding elects one member. The election method is that every voter gets one ballot, and marks an X beside the candidate they most want to win. When the ballots are counted, whatever candidate has the most votes, wins the riding, and becomes the MP.

This is perhaps the simplest election method, which is why it is so widespread in history. The problem comes when you have more than two candidates running as someone can win with under 50% of the vote. Imagine if you will a riding where 65% of voters want someone named Jean to be their MP. There are three candidates, Jean Smith, Jean Beliveau, and Bob Menendez. Smith gets 32% of the vote, while Beliveau gets 33%. Menendez however gets 35% of the vote. Despite the fact that 65% of the riding want a Jean as MP, they get a Bob instead.

The Tories and to a degree the Bloc support this system. Both pretty much for the same reason, there are far more federalist or progressive parties out there than there are either conservative or sovereigntist parties. FPTP is the best system to maximise CPC and BQ seats.

This system is used in the USA and UK.

Ranked Ballot
Also known as Single-Member STV, the Alternative Vote, Instant Runoff Voting, and the Preferential Ballot. The general thinking is the Liberals want this method.

In this method, you divide the country into ridings like you do in FPTP, but the voting and counting is different. Rather than simply marking an X beside the candidate you want to win, you rank the candidates you want to win 1, 2, 3, and so on. When counting the ballots, you only find a winner when someone has 50%+1 of the vote. To get there, you remove the least popular candidate and redistribute his votes based on additional preferences.

This is usually one of the go-to systems for those looking for something more fair. The problem comes with the fact that you end up with a result where you do not pick a winner based on who everyone wants, but based on who everyone does not want. People will rank candidates lower if they want to stop them from being elected, meaning that moderate candidates who are inoffensive have a higher chance of winning, even if few really want them elected.

Such a system would generally benefit the Liberals, especially early on. After a few elections with this system it's quite possible we would eventually see Tories and New Democrats preferencing one another to stop the Liberals from winning. Regardless, if any party is to see a boost from a changed voting system, this is the system expected to best boost the Liberals.

This system is used for house elections in Australia, and has been used in various provinces at various times (Manitoba from 1927-1953 for example)

Single Transferable Vote
This is the Multi-Member variant of STV, and is traditionally what is meant when people say STV. This system is also known as "BC-STV"

In this method, you divide the country into chunks that we will still call "ridings". Each riding need not be equal in population. When it comes time to vote, every voter still gets one ballot like in the Ranked Ballot, and they rank them 1, 2, 3, and so on, however, when it comes time to count, things are done differently. Each riding elects multiple members. The method to get there is a bit complex so I'll skip it for now, but be aware the results are 'a bit proportional' with some caveats. Generally each riding in a STV system will elect between 3 and 9 members.

These systems tend to balance the desire for local representation - something that many voters in countries moving away from FPTP are afraid of losing - with the desire for proportional representation - since nearly every riding in this system will have MPs from multiple parties. The major problem with this system is that it does neither of these as well as other systems, and the fact that you can secretly gerrymander the system by adjusting the number of MPs elected per riding to best advantage your own party.

From what I can calculate, this is actually the best system for the NDP, because it allows for Liberal and Green votes to transfer. On the math, it's likely that an STV would cause the largest number of NDP MPs to be elected based on any given popular vote share. Despite this, the NDP does not currently back this as their prefered system.

This system is used in Ireland, and for Senate elections in Australia.

Parallel Proportional Representation
This is sort of a "hold over" system I've included to represent various adjustments that can be done to PR to make it more possible to achieve majority governments. This entry is intended to represent all such adjustments, of which, Parallel is just one.

In this method, you do not need any ridings for the proportional seats, though, many countries will use various electoral districts (such as states or provinces) to help keep things at least minimally local. Other places do still create ridings, very large ridings (with 10 or more members) to achieve this purpose. This is a mixed-member system, meaning that you still have single-member ridings electing MPs, but you also add the MPs elected proportionally to that total to help create a more proportional result. The proportional MPs are elected based on the proportion of votes each party gets.

This is the weakest version of real proportional representation that can be adopted and is often looked at by countries where there is a desire to fix the problems with FPTP, but where there is a strong aversion to creating an endless string of minority governments. The major problem with this system is that it is not fully proportional, and as such, is unlikely to be the "first choice" of a voting system by many voters. The main 'benefit' of this system is that it is a compromise, which is also its greatest drawback. It's difficult to look at this system positively for that reason.

No one party stands to gain the most from this system, but all parties do stand to gain in one particular way; no longer will there be swaths of the country without MPs from major parties. If such a system had been in place historically, we'd have seen Quebec NDP MPs for decades, and the Reform Party would have had a small number of MPs from the Atlantic in 1993.

This system is used in Japan and Mexico.

Proportional Representation
When used in the context of current electoral discussions here in Canada, PR generally refers to one system, FUMMPPR, or Fill Up Mixed Multi Member Proportional.

Like the above PR system, you still use ridings, and add proportional members to that. Unlike the above, the proportional MPs are not assigned based on vote totals alone, but based on vote totals minus the number of seats already won in the ridings. While this sounds complex (and can be complex) the simple way to explain it is that if a party takes 40% of the vote, it's final end total will see it win 40% of the seats in Parliament. Everything else in the system is just math to ensure that this will ring true at the end of the day.

The reason to move to such a system is simple, to ensure that parliament looks similar to the vote totals as cast by the voters in the last election. The biggest drawback is that this will almost always mean that no one party can obtain a majority; which has time and time again been shown to be the biggest concern of Canadians when it comes to electoral reform.

The Green Party likely has the most to gain from this system, and the party supports it. Additionally, the NDP is currently pushing for this. Recently, after the Liberals changed the makeup of the committee that will look at electoral reform, there has been some thought among some that the Liberal Party may be willing to look at this system as well.

This system is used in Germany and New Zealand.

Our next election will all but certainly be held using one of these systems. Parallel systems, as noted, are adopted to allow for majorities, and, in fact, Japan has had a healthy string of majorities since it's adoption of a parallel system in the 90's. There are other 'variants' that one can add to PR to make it more friendly to majorities. Italy for example simply gives the winning party 55% of the seats, even if that party does not take 50%+1 of the nation-wide vote. Greece awards number of free seats to the winning party, meaning that about 40% of the vote will result in a majority. It's quite possible that such a variant may be looked at here as well.

In my next post I will look at what may happen if we move to the systems outlined above.

No comments:

Post a Comment