Electoral Reform Committee on Sept. 22nd, 2016

[Yvan Dutil]

Ranking seven persons or seven anything in order is known in psychology as a human limitation. I think Raymond Côté will speak more about that in a few minutes. It's just a limitation of the human brain.

I saw this in the French election, an experimental election they did. They asked people to use other voting systems, and in one of those you ranked your candidates. People were pretty reasonable for the first seven, but then the system crashed and it became random. At that point, it was more than what the human brain can do. It was just too hard.

I have a post-doctoral person in my university who is living in India, and they have sometimes tens of candidates on the voting tickets. They are that large, and it's just impossible to do. Some people say we could just have a cut-off and order the first five. Depending on the voting system, some might allow that and some might not.

As for the voting systems using some kind of ranking, there are about 20 of those. There is one, STV, which is used in Ireland, but I don't have much expertise on that. One that is very good is the Borda count, in which you give points depending on the order. I think it's done the way journalists might find the best hockey team. It's five points for the first one, four points for the second. It's very simple arithmetic. You find the best candidate defined as the most central one, the guy or woman who is most average. It's very nice if you want to elect a president or you want someone who represents the population.

There is a philosophy here. Do you want that for an MP, a member of Parliament who represents a riding? Do you want the best average guy, or do you want a Parliament that is very diverse? If you go for that, you willl end up with the most central parties. In Canada, in most cases, it would be the Liberal Party or the Conservative Party. In some regions the NDP might be considered more central. You end up with a concentration.

I did a simulation in Quebec. I've never done it in Canada. In Quebec I ended up with Liberal, PQ, and three or four members of the ADQ. It usually wipes out those who are somewhat less central and it converges towards two parties. That's what we are supposed to fight against now.

I would use this in one case, though: the case of partial elections. At that point, you might want to reset the system to get a replacement guy. We're replacing the actual MP. We'll use one of those sophisticated methods to find the best average guy for the rest of the term. It might give a chance to another party that otherwise would not be detected by the system.

[Elizabeth May, Green, Saanich—Gulf Islands, BC]

By “partial election”, do you mean a by-election?

[Yvan Dutil]

Yes. That's my fault.

[The Chair, Liberal, Francis Scarpaleggia]

Thank you.

Mr. DeCourcey is next.

[Matt DeCourcey, Liberal, Fredericton, NB]

Thank you, Mr. Chair.

My thanks to the witnesses. On behalf of my fellow Acadians and Brayons from New Brunswick, I thank you all here in Quebec City for your welcome. I also thank our team for giving us a little time to find our way around this beautiful city. We made good use of the time.

Mr. Rémillard, in your system, could it happen from time to time that the party that won the election and formed the government would be given extra seats?

[Jean Rémillard]

No. In the plurality system that we have in Canada, the party that wins the election usually has a higher percentage of members than their percentage of votes, except in cases when a government is elected with a minority of members. That requires a coalition government.

In Canada, contrary to what people think, we are very open to coalition governments. In fact, 40% of the Canadian governments have been coalitions. Since 1963, they have been in power about 20% of the time. One government that worked very well was Pearson's. It was not a coalition government but a government supported by the NDP. It was elected twice and it worked very well. It has been called one of Canada's best governments.

In Canada, when a party has a majority of the elected members, it forms the government. Personally, I favour parliamentary functions to be assigned according to the percentage of votes won rather than according to the percentage of members elected. At the moment, what counts is the number of members elected. In some cases, a party wins a majority of seats with a minority of votes while the official opposition has more votes in percentage terms than the governing party.

We have to turn that around.

Those kinds of situations have often happened in Quebec. With the system I am proposing, we could have turned things around, that is, we could have given the government to the party with the highest percentage of votes, rather than to the party that officially won according to its number of seats. However, there is one case in the history of Quebec where that logic was completely impossible. It was when the Parti québécois was elected and the Liberal Party, I believe, came second. The Parti québécois got so many MNAs that, even applying high ratios—like 1.2—the PQ could not be moved.

I don't think that a situation like that has happened in Canada, probably because we are playing with bigger numbers. In that case, the danger of such a problem happening is less in Canada. In general, the government elected with a good majority stays in place, just like in Germany.

[Matt DeCourcey, Liberal, Fredericton, NB]

Thank you very much for those clarifications.

Mr. Dutil, I would like to leave you with the last word.

You mentioned that a psychological barrier for voters is seeing their vote manipulated, one vote counting less than others, that is.

You and Mr. Rémillard mentioned the challenges involved in a proportional system. What kind of psychological barriers do you see in the various electoral systems we have discussed with all the witnesses who have appeared before the committee?

[Yvan Dutil]

Everything that changes people's way of voting a little is not going to go over well.

In 2007, I did an experiment with some students at the Université Laval. We simply asked them to put candidates in order and to provide an evaluation. We used three voting methods: approval voting, evaluation voting, where candidates are given points, a grade, and preferential voting where they were asked for an order of preference. Ten per cent of the students were not able to do the exercise; they made mistakes, they got things wrong, they reversed the candidates, they didn't enter the points. We were able to observe that, because the results were inconsistent from one voting method to the other.

So, the further we move away from the current voting method, the more problems there will be. In practice, at least in the first election, it will be very difficult. But that does not mean that we can't do it.

Another factor is that we are used to having majority governments. People do not like minority governments and they do not last long. So that means coalition governments.

There is one more important factor to note that is not so much psychological as operational. The Governor General has to have the moral authority to be able to say that he prefers a coalition government to a minority government, if that proposal is made to him. If an election is called for frivolous reasons, for the Governor General to say no, he would have to be elected by Parliament, not appointed by the Prime Minister's Office. It is like that everywhere in Europe where they have proportional systems. At a certain point, a judge needs to decide. If one party is not capable of forming the government, the Governor General has to be in a position to ask the party with the second highest number of members to take power. That does not exist in our current system, especially since our Queen does not operate politically. So that is something we would need. Of course, that kind of thing could shock people. At the moment, the Governor General is a figurehead, but he could become important. That also could turn out to be a little problematic.

[The Chair, Liberal, Francis Scarpaleggia]

Thank you.

I have a question to bring this part of the meeting to a close.

We have talked a lot about the stability of parliaments elected by proportional representation. Up to now, we have been told that proportional systems create compromises and consensus and give rise to a measure of stability, because people have to get along. I assume that, with too many parties, it could be difficult to reach compromises but that alliances could form.

Depending on the context, do you feel it is possible, in a proportional system, that we could find ourselves looking at two major parties, sometimes diametrically opposed, with the result that the possibility of an election is always hovering over us?

Ever since I was elected to Parliament, I have gone through a number of similar situations. The government was in a minority, but that did not mean that parties got along. It meant that they were waiting for the right time to trigger an election.

Do you think that, even with a proportional system, we could end up with two ideologically opposed parties, resulting in alternance and waves of change? Is that possible, or does it not happen anymore once proportional representation has been adopted?

[Yvan Dutil]

A situation like that is becoming much less likely. Fragmentation is not impossible. However, everywhere in the world, we are seeing the number of parties increase. So a danger arises in the first or second elections because, if there are not enough parties, coalitions have to be formed in an intelligent way. Let me explain.

At the moment, we cannot imagine a lot of coalition combinations that can possibly form a majority government. To form coalitions, there have to be enough parties. So, with no fragmentation—there may be some, but it may only happen after at least three elections, the equivalent of 12 or 15 years—we still may have those two blocks bumping into each other for several years. That could end up falling apart.

I know that there were still two major blocks in Spain when I was there in 2000.

[The Chair, Liberal, Francis Scarpaleggia]

Okay.

Mr. Rémillard, do you want to comment?

[Jean Rémillard]

I want to point out that, in Germany, the situation is very strange. Yet the country has a proportional system. The CDU-CSU alliance has formed the government since 1949, except in three elections. Those two parties form coalitions with other different ones.

[The Chair, Liberal, Francis Scarpaleggia]

The same people are always in power.

[Jean Rémillard]

The same party is always at the centre of power.

[The Chair, Liberal, Francis Scarpaleggia]

Yes. So that can result in a kind of monopoly.

[Jean Rémillard]

It's worse than in Canada. At least here we have some alternance.

[The Chair, Liberal, Francis Scarpaleggia]

Right.

[Jean Rémillard]

In Quebec, it is even better in a way, because the alternance is at the head of the system, meaning that the government party can be something other than Liberal or, back in the day, something other than the Union nationale, and so on. It also affects the party that forms the official opposition. There is a certain progression and, eventually, you really get to form the government.

[The Chair, Liberal, Francis Scarpaleggia]

Thank you. That was very interesting. As I mentioned at the start, we have moved in a new direction. We have learned a lot from your points of view; they seemed very original to me. Thank you very much.

We are now going to take a short coffee break. When we come back, we are going to hear from two other witnesses. After that, we will have an open mic session as part of the meeting.

Thank you.

[The Chair, Liberal, Francis Scarpaleggia]

Colleagues, let's start our work again.

I would like to welcome two other witnesses. First, we welcome Raymond Côté, whom we know well, since he sat in the House of Commons with Mr. Boulerice, Ms. May and myself. We also welcome Professor Jean-Pierre Derriennic, associate professor in the Department of Political Science at the Université Laval.

I imagine that the witnesses know how a session unfolds. Each witness makes a 10-minute presentation. The presentation is followed by a period for questions in which each member has an equal amount of time to interact with the witnesses.

We will start with Mr. Côté.

You have 10 minutes.

[Raymond Côté, As an Individual]

Thank you for allowing me to speak, Mr. Chair.

I must confess that I am a little frantic at the idea of being on the other side of the fence. It’s my turn to be grilled, after having perhaps terrorized some witnesses who appeared before four House of Commons standing committees on which I sat for four and a half years. I am ready to answer your questions.

My thanks to all the members of Parliament who are here to participate in this committee. This is a fundamental topic that I have been passionate about for a long time. It was one of the many reasons that led me to become active in politics.

I had the great privilege of being a member of Parliament for four and a half years. I was defeated in 2015. It was a great privilege for me to participate in four federal elections, in 2006, 2008, 2011 and 2012. I mention that because, in 2006 and 2008, my chances of being elected were very slim. That was a huge sadness for my mom, who also wondered why I was running for the NDP.

That said, I always campaigned in the streets, and I knocked on an enormous number of doors. By so doing, I also had the great privilege of observing the voters’ behaviour. From then on, I got clues that led me to understand the extent to which our voting system influences their behaviour.

In my first campaign in 2006, I remember that we were trudging around in snow. I’m sure you will remember the never-ending, 56-day campaign; the one in 2015 was not the first. Voting day was January 23. In Quebec City, it snowed practically every two or three days; it was horrible. At the time, Jack Layton, my leader, was not too well known. I actually had to give voters an idea of who he was by describing him as the man with a moustache who was smiling all the time. People then recognized him and told me that they liked him very much. That allowed people to see me, not as a candidate, but as a representative of that party leader.

So, in 2008, in most Quebec homes, Jack was already a member of the family, so to speak. That allowed me to make a very important, very interesting observation. People already were extremely fond of Jack Layton. However, most people’s reaction was to tell me that they wanted to vote for me but they didn’t believe that I had any chance of being elected at all.

That behaviour is widespread. From square one, it is caused by our first past the post voting system.

For most people— by which I mean my mom and that group of voters—what is important is that their vote should have some use, by voting for the winner. During an election campaign, the real challenge is to consider who will win the election and how one’s vote can go to that winner, even if the candidate in question does not necessarily reflect the interests, the needs and the objectives of each voter. It’s a flaw in our current voting system, it is very imperfect and we cannot keep it. The status quo forces us to continue to live with results that do not reflect the will of the people.

This is also the case for a simple plurality voting system with two or more rounds. France operates with a two-round system, which has led to absolutely extraordinary distortions. Though my memory is very short, I might mention the surreal results of the presidential elections in 2002. Jean-Marie Le Pen, the Front national candidate, slipped into the second round. He was then humiliatingly crushed by obtaining only 17.7% of the votes, while Jacques Chirac obtained more than 80% of the votes. That showed that a huge number of French voters were absolutely not inclined to support Jean-Marie Le Pen. As a candidate, he even repulsed them.

That election was the catalyst that lead two French researchers to come up with the voting system that I am going to present to you. It convinced them that it was really necessary to give people another option.

I also went through the briefing note on voting systems. The preferential system has more or less the same problems as the simple plurality system. Arranging people in a hierarchy, ranking them, can lead to distortions that result in the candidate who, by all objective measures, should be the winner not necessarily ending up as the winner. That is because, by playing with the second and third rankings given to their candidate, voters can basically arrange for an unexpected candidate to come out on top.

We have been using the first past the post system for 150 years now. We could very well keep doing so. Up to now in Canada, to my knowledge, there have been no riots after the results of an election. The system seems to be accepted, even acceptable, for a good many people. However, unfortunately, it is a system that does not necessarily reflect the diversity of the voting and the opinions expressed. I repeat that it also gives results that can even be contrary to the interests of the majority of the voters, either within a constituency or nationally.

Today, I am introducing the majority judgment system. It was developed by two French researchers, Michel Balinski and Rida Laraki. The system breaks the hierarchical approach and the distortions that can be seen in the various voting methods around the world. My brief—which the previous witnesses mentioned—contains thoughts on election results. This has been studied for many years. Borderline cases have been considered by mathematicians, by great thinkers, who have shown the extent to which the result cannot reflect the will of the people. The majority judgment voting system really brings everyone's opinion together.

In my brief, I explain that Kenneth Arrow's theorem gives three very important criteria that a voting system must meet. It must always declare a winner for all voters, which our current system does not do. Then, it must prevent the addition or removal of a minor candidate from influencing the final result. Third, it must ensure that all votes are treated equally. I would go even further and say that it must ensure that the vote of each and every voter is considered in the result.

Let me explain how the majority judgment voting system works. Instead of creating a hierarchy or choosing a winner, voters express a judgment on each candidate. The judgment can range from “excellent”, which is really good for the ego, to “reject”, which can be very hard for a politician trying to survive personally.

I am exposing you to danger by proposing this voting system. You can get even later on when you ask me questions.

The two French researchers I mentioned conducted very interesting experiments during the presidential elections in 2007 and 2012. One of the things the experiments revealed is what I call the hidden thinking that a country's entire electorate does not see. We will consider the experiment in Orsay, a suburb of Paris; it was conducted as people left three polling stations in the municipality. Voters were invited to try the majority judgment system as they saw fit. The results were obtained both nationally and for the municipality of Orsay. Those results were very different. I kept five candidates in the Orsay experiment, but we must remember that, in 2007, there were 12 candidates.

In the first round of elections in France, Nicolas Sarkozy came first; Ségolène Royal, second; François Bayrou, third; and Jean-Marie Le Pen, fourth.

For the purposes of the exercise, I also chose a candidate from the Green Party—not necessarily as a tribute to Elizabeth May—simply because the difference in the percentage of votes for her was truly remarkable: she came in eighth. In France, she was still seventh in the town of Orsay. If you look at the results of the experiment with 1,752 voters, you will see that the hierarchy changes. With the majority judgment voting system, François Bayrou comes first; Ségolène Royal, second; Nicolas Sarkozy, third; and the first major change, Dominique Boynet, the Green Party candidate, is fourth. As for Jean-Marie Le Pen, the leader of the Front national, ranked twelfth, so last. In his case, over 70% of the 1,752 voters graded him as “reject”. This made it possible to clearly show—which I find particularly important—the opinion of all the voters.

I did not want to tell you about the mechanics of this voting system. I especially wanted you to think about the fact that people can see their will truly and fully reflected in an election through a radically different voting system, which may well prompt a significant change in the behaviour of voters and politicians.

Mr. Chair, I see that my time is up. Thank you.

Anyway, I look forward to being grilled by members of the committee.

[The Chair, Liberal, Francis Scarpaleggia]

Thank you very much for your presentation, which was very eloquent.

We will now turn to Professor Derriennic for 10 minutes, please.

[Jean-Pierre Derriennic, Associate professor, Department of political science, Université Laval, As an Individual]

Thank you, Mr. Chair. Thank you for inviting me.

I submitted a brief summarizing the findings of a small activist book that I wrote and that was published at the start of this year. It is called A Better Electoral System for Canada. I will summarize the conclusions of my brief.

In my submission, I drew your attention to the flaws of the voting system we are using before telling you about the two reforms that should be avoided and the two reforms that would significantly enhance our political life. They can work harmoniously together.

The flaws of the single member plurality voting system are well known. In my book, I mention some very striking figures. Last year, the House of Commons was elected by 48% of people who voted; 52% of people voted for candidates who were defeated. The Liberal Party won 39.5% of the votes, one-third of which went to Liberal candidates who were defeated. So the Liberal majority in the House of Commons was elected by 26% of the people who voted, and if we factor in abstentions, 18% of Canadians of voting age. If we did the math for previous elections, we would sometimes find even more distressing figures. I think that leaves significant room for improvement to reduce the dissatisfaction of many of our fellow citizens who think that our electoral mechanisms are very flawed.

The first thing to avoid is proportional representation in districts where large numbers of MPs are elected, which allows many parties to have elected MPs in the House and poses a serious risk of political indecision and instability. All the awful speeches given by opponents of proportional representation are partially justified in cases where proportional representation allows the House to have a large number of MPs. However, it is possible to achieve proportional representation without this shortcoming.

I think the second thing to avoid is mixed electoral systems, combining single member districts with some kind of proportional representation. This would have little benefit compared to a simpler, more reasonable proportional system and it would be difficult or impossible to implement before the next election. The thing to have is moderate proportional representation in districts of three to five seats. I stole the term moderate proportional representation from Vincent Lemieux who was one of my mentors on these issues.

So the system to adopt is moderate proportional representation with a preferential vote. The two formulas are not at all conflicting. They could be easily combined and so would their benefits. Moderate proportional representation would not be very difficult to implement because it could be done by joining together the existing districts without changing their limits, without changing the number of MPs in the House of Commons and without changing the number of members by province. It would be less difficult and faster than establishing the new single member districts that would be needed for a mixed system.

Moderate proportional representation allows for a more equitable distribution of elected representatives between the major parties and significantly reduces the number of votes that elect no one. It would also have the advantage of ending the situation whereby it is impossible for major parties to have elected representatives in one province or region of the country. This is the case today, such as the Conservative Party in the Atlantic provinces. We have known for a long time that this regional concentration of elected officials is one of the most negative effects of the voting system on Canada. This is a very old and justified idea.

Thanks to the small number of elected MPs in each constituency, these benefits would be achieved without the risk of government instability caused by the proliferation of parties able to have elected representatives. Moderate proportional representation would ensure local representation, which is so important to so many people in Canada, as well or even better than the current system or a mixed system.

In a mixed system, some of the members do not have a home base whereas in constituencies with three to five seats, all members are elected in a territorial division. They share it with other members. Some may be allies, others may be opponents. The number of MPs relative to the population would remain the same. Members would not be further away or more difficult to reach. Sometimes, the physical distance would be somewhat greater in cities. If there were a single constituency in Quebec City, it would be no problem to meet with a member about the situation today.

This would not be very problematic in densely populated rural areas either. In Gaspé, there would need to be a constituency that goes from Montmagny to Gaspé. There perhaps distance might become an issue, but there would be four MPs who have their offices in the same place.

In the northern territories, we could keep three single member districts, where distance becomes a very unique problem. This would not change the overall logic of the system.

For constituents who want to approach an MP, I think multi-member districts with three to five members would be a tremendous advantage for them. First, most of them would have the choice of going to either a member of the majority government or a member of the opposition. Today, they do not have that choice at all.

Today, in the case of more than half of those who voted, when they meet their MP, they meet a member they voted against, whereas in the new system, they would almost certainly be able to choose between their three, four or five members. I am confident that the voters of Canada would love to have that opportunity.

To the moderate proportional representation, we have to add the ranked ballots. Preferential voting is the most effective way to ensure that every vote counts. They are never fully equal. I do not think that's possible. We can have a voting system in which every vote counts. In terms of being exactly equal, I don't think there's a system that produces that result anywhere in the world today.

Preferential voting should be a sort of ethical obligation, not in the voting system, because it allows voters to vote sincerely without fear of wasting their vote or of being forced to vote strategically. Strategic voting is not immoral. What is immoral is maintaining a voting system that requires a large number of voters to choose between voting strategically or casting a vote that is completely useless at election time. I think that's what's immoral, not strategic voting, which is a logical thing for voters to do.

Preferential voting places voters in a much better intellectual and moral position when deciding how to vote. It also has the advantage of allowing small parties to know what their real popular support is and to play a bigger role in public debate without the risk of proliferating parties in the House. It encourages major parties to address the concerns of those who voted for smaller parties, which they know had elected representatives through their second or third preference. Small parties do not necessarily have more elected members, but they play a much bigger role in public debate through ranked ballots. This might encourage the major political parties to produce less simplistic propaganda, because they always know that they may need their opponents' second or third preferences in a given constituency.

With moderate proportional representation, parties that get less than 17%, 20% or 25% depending on the size of constituencies or votes, cannot have elected representatives. So it makes sense to have ranked ballots between parties. For the supporters of those parties, the need for strategic voting does not completely disappear with proportional representation. We must add the ranked ballots. We can also have a slightly more complex system in which voters indicate their preferences not only from the parties, but also from the candidates. This complicates the counting of votes a little. I don't think it complicates the casting of votes for voters.

There is an electoral system like this in Ireland, which has its merits or flaws. For Irish voters, the system is not difficult to apply.

It is simply that we will have to wait for the results until the next evening because of how long it takes to calculate the results. This is something that can be extremely justified.

Do I still have a minute? I would like to conclude.

[The Chair, Liberal, Francis Scarpaleggia]

Yes, you still have one minute.

[Jean-Pierre Derriennic]

If the reform you are studying leads to a moderate proportional representation system with ranked ballots, there will be a number of benefits for the citizens of Canada.

The members of the House of Commons elected by 48% of the citizens who voted will be elected by almost all voters. If preferential voting is applied and if everyone completely lists their order of preference, members will be elected by all the people who voted and listed their order of preference. The majority of MPs will really represent a majority of voters, instead of a little over a quarter as is the case today. Voters will be able to vote honestly, without having to make assumptions about strategic voting decisions. Political parties will be encouraged to pay more attention to the concerns of all citizens, and the public debate will become less simplistic and less confrontational.

Thank you.