Electoral Reform Committee on Sept. 22nd, 2016

[Jean-Pierre Derriennic]

Closed lists are exactly the same as the current situation. Right now, we don't choose the candidate in a constituency. We either vote for the person because we are happy with their party, or we vote for the person and ignore the party they are representing—I think very few people do that—or we vote for the party, regardless of the candidate. Instead of having one candidate imposed by the party, we would have five.

[Alain Rayes, Conservative, Richmond—Arthabaska, QC]

By the way, the candidate is not imposed by the party but by the party members in the constituency—

[Jean-Pierre Derriennic]

People on the list will be appointed by the party members in the constituency.

[Alain Rayes, Conservative, Richmond—Arthabaska, QC]

Let me go back to the same exercise we did for Quebec City. In my region, in central Quebec, there are three MPs. In this region, people vote primarily for the Conservatives. When my constituents are going to vote, they have a choice. Some people vote for the platform, others vote for the party and others vote for the leader—in our system, we seem to give a lot of power to the leader. However, I have to say that other citizens vote for the individual. I think people win an election because of their reputation, their experience, in particular. Suppose we bring together candidates from the various parties on their respective lists, and a person wants to vote for the Bloc. It is possible that the candidate of their choice, who lives in their riding, finishes in sixth or seventh place, and therefore is not elected.

I want to get this straight. I'm not judging the model you are proposing. I feel that grouping the individuals devalues the vote, compared to the current system. People who voted Conservative in my constituency did not do so for the leader or the platform, but for Alain Rayes. They made an informed choice. It would not be the case once the candidates were grouped into a list. Am I wrong?

[Jean-Pierre Derriennic]

So you would prefer ranked ballots to be applied as they are in Ireland, and allow voters to pass judgment on both the person and the party, and possibly—

[Alain Rayes, Conservative, Richmond—Arthabaska, QC]

Ultimately, yes, if I have to choose between the two options in your system. However, I am not sure I would opt for this system.

[Jean-Pierre Derriennic]

My preference would also be for the Irish system. People who say that the Irish system is dangerous for the cohesion of the party may obtain a concession. It is possible to have ranked ballots between the parties, without—

[Alain Rayes, Conservative, Richmond—Arthabaska, QC]

Okay, it is very clear, Mr. Derriennic.

I want to make sure I understand. In your system, there would be the same number of members as right now. There would not be more or fewer members. This system would allow MPs to rack up a higher percentage of votes. People who are in favour of proportional representation claim that, in the case of the Conservatives... We always give the example of the Conservatives and the Liberals, but I often like to say that, in Alberta, the NDP has a majority with 39% of the vote. Distortion is everywhere.

Would your system allow us to be closer to a percentage of seats rather than a percentage of votes?

[Jean-Pierre Derriennic]

Of course. This would not lead to perfect proportionality. No electoral system provides a perfectly proportional result. It may be more or less proportional. The determining factor is the number of elected representatives in constituencies.

In the Atlantic provinces, where 30% of votes were for the Conservatives, there would clearly have been more Conservative MPs elected if we had used this system. I would have to do the math. It would not be exactly proportional, but it would be much less unbalanced than it is today.

[Alain Rayes, Conservative, Richmond—Arthabaska, QC]

Do you want to connect your model to a regionalized preferential system?

[Jean-Pierre Derriennic]

No.

To me, Vincent Lemieux's term “moderate proportional representation” is excellent. If you want to develop a system proportionate with the size of Canada, it is sufficient to obtain one-third of 1% of the vote to be elected. It is also possible to implement a proportional system based on the size of the provinces. In Ontario, it is sufficient to garner 0.8% of the vote to be elected. In Quebec, it takes a little over 1% of the vote and about 10% in New Brunswick. Proportionality depends on the size of the constituencies.

Vincent Lemieux's argument is that, by creating small enough constituencies, we avoid the proliferation of political parties. This is a fundamental political issue.

[The Chair, Liberal, Francis Scarpaleggia]

Thank you.

Mr. Boulerice, the floor is yours.

[Alexandre Boulerice, NDP, Rosemont—La Petite-Patrie, QC]

Thank you, Mr. Chair. That is a fascinating discussion.

Mr. Côté, we were talking about voter behaviour earlier.

When I was a candidate, in 2008 and 2011, people often told me that they agreed with me, that they liked me a lot and that I was a very good guy, but that they would not vote for me because they believed I had no chance of being elected. I always answered that they were absolutely right because, if they did not vote for me, I had no chance of being elected.

At some point, it's simply the build-up of changes in intellectual or psychological paradigms that makes things happen. You told us about the principles underlying your preferred, majority judgment model, but you did not talk about the mechanisms. I would like you tell us about that.

When the term “excellent” or “gross” is selected, how is that quantified in terms of the numbers of members? I guess there's a point system.

Do you win four points with “excellent” and lose 10 with “completely reject”?

How does it work?

[Raymond Côté]

Thank you, Mr. Boulerice. That's a very good question.

Basically, the system is very simple and requires a relatively simple mathematical operation.

For each grade, we compile the votes, the percentages. Whether we start with the worst grade, “reject”, or the best, “excellent”, we count the votes to reach a threshold of at least 50%. That's how we determine the majority grade. That's how the system got its name.

There are a number of aspects I have not mentioned, including the major advantages of this electoral system. Mr. Derriennic talked about the preferential voting system, which I personally loathe. Just like the first past the post system, that system is easy to manipulate. However, the majority judgment system is practically immune to external manipulation.

To illustrate the point simplistically, say that someone with a lot of money can buy their election by renting buses and surrounding themselves with many volunteers with cars, in order to give voters rides. In a close election, this would ensure the candidate has 500 or 1000 more votes in order to beat an opponent.

In the majority judgment system, since there is a threshold, the sole interest in filling up the buses is to engage more voters. The system completely prevents candidates from choosing voters to tip the balance in their favour. The people the candidate would transport and who would vote for them would be in the 50% of voters. That may be a bit difficult to understand, but this system is extremely transparent and prevents manipulation.

There are other aspects we can talk about, but I'll let you ask other questions.

[Alexandre Boulerice, NDP, Rosemont—La Petite-Patrie, QC]

Thank you, Mr. Côté.

Mr. Derriennic, I started studying the Irish electoral system as well. Along the lines of your proposal, the Irish have three to five members per multi-member constituency. The Irish, instead of choosing the party only through closed lists, can choose their candidate without having to make 25 choices. They can choose their preferred candidate from a given party. This would answer Mr. Rayes' question. A voter who wanted to vote for Mr. Rayes would ensure that he is at the top of their list. That selection would mean that the voter was voting specifically for that individual.

I admit that the two systems are an improvement over the archaic system we currently have in Canada. Someone even said that this system was designed for England in the 16th century, while we are in Canada in the 21st century. Jason Kenney said that.

Why three to five members per district? Yesterday, Professor Stephenson suggested three to seven members. Earlier, Mr. Dutil said that it takes at least six to achieve better proportionality. The debate is about the number of members that we want for our region.

[The Chair, Liberal, Francis Scarpaleggia]

Please provide a quick answer. We are running out of time.

[Jean-Pierre Derriennic]

I tend to think there should be three or four. To avoid having too many parties in the House of Commons and compensating for small political parties, the transferable vote helps them find out where they stand and helps them look after their voters. Having three or four members per district would be a huge improvement over the current situation. It is the way to minimize the danger of instability that is always condemned in the case of proportional representation. However, I would not be upset if there were six or seven.

[The Chair, Liberal, Francis Scarpaleggia]

Thank you.

Mr. Ste-Marie, you have the floor.

[Gabriel Ste-Marie, Bloc, Joliette, QC]

Thank you, Mr. Chair.

Good afternoon, gentlemen.

My questions are for Mr. Derriennic.

To begin, as you just repeated for Mr. Boulerice, we need to avoid ridings with several members that would be too large to prevent the proliferation of small parties. If I understand correctly, you are of the opinion that Parliament shouldn't have too many parties. Your preference would be for three to five members per regional riding. To that, you are adding the preferential system.

Have you done a modelling exercise for some or all of Canada to see what the system you're recommending would look like? Modelling is possible because there are polls about the second choice candidates. Since the committee has already discussed the preferential system at previous meetings, I think we will need to see the outcome to properly assess the quality of the model.

Take the Maritimes, for example, where the Liberals won all the constituencies. Theoretically, in a proportional system without preferential votes, the Liberals could still win all the constituencies, but it would become impossible in a preferential system. So it's possible that the first choice of the electorate would not be respected if we had to rank all the parties, right?

[Jean-Pierre Derriennic]

I think the opposite would be true for the Maritimes.

If we held a proportional election without preferential votes, the Liberal Party would lose seats in the Maritimes. I don't know which ones or how many. If preferential voting was added, since the Liberal Party is the centrist party, I think it would lose fewer. But I haven't done any modelling, and I doubt that modelling would be entirely useful because people don't always vote the same way.

We can do a modelling exercise. It gives an indication, but it is very uncertain because people would understand very quickly how a preferential system works. The political parties that use it to designate their leaders wouldn't have any difficulty making their voters understand how it works.

[Gabriel Ste-Marie, Bloc, Joliette, QC]

I have nothing against your idea about regional constituencies with several candidates, but I am against the preferential aspect because, you said, it benefits the centrist parties. In the current system, it gives the Liberal Party a direct advantage because it will probably be the second choice of New Democrat and Conservative voters.

Yesterday, I think it was, one witness said that he did not feel that the elected candidate was the second choice for everyone, while the first choice wasn't. I think that's a problem. Can you briefly respond to that?

Afterwards, I'll have a question for Mr. Côté.

[Jean-Pierre Derriennic]

The example of the 2002 presidential election that Mr. Côté used is very interesting to see the merits of a preferential vote. Le Pen got in just ahead of Jospin, the socialist candidate, because 25% of voters voted for other leftist parties.

If there had been a preferential vote instead of a two-round vote—two-round voting is partial preferential voting in which voters can give only two preferences, for which they must vote twice—from the early counting of the results, Jospin would have beat out Le Pen. Perhaps Jospin would have won the election, or Chirac would have won the election, but with a normal figure opposite Jospin.

The kind of drama that the French pretended to play in 2002 is entirely the result of a flawed electoral system. The French didn't know what a preferential system was.

[Gabriel Ste-Marie, Bloc, Joliette, QC]

I'm going to interrupt you to ask Mr. Côté a question.

Having said that, thank you. It's clear. I remember the situation well. I was studying in Toulouse at the time when this shockwave happened.

Mr. Côté, the concern I have about the majority judgment voting process is the matter of strategy. Suppose candidates A and B are neck and neck, and they are both good candidates. If I support candidate A, I will be inclined to choose “reject” for candidate B. If good faith was at play, I would be in favour of this principle.

Convince me very quickly that we can trust this strategy.

[Raymond Côté]

Certainly, Mr. Ste-Marie. Thank you.

I'll take this opportunity to make a comment. Mr. Derriennic spoke about territorial proportionality, namely, many candidates in the same territory. A vote by majority judgment would be quite suitable in this case.

My short brief includes a bibliography. In the summary of the Orsay experiment, French researchers give details on the transfer of votes. For example, if someone wants to fill the ballot boxes and convinces friends to choose “reject”, those votes would be lost in the bulk of the 50%. So there would be no impact on the final result. That's the beauty of this system. To really manipulate the results in a majority judgment system, you need to draw thousands of voters. That's so significant because it's practically impossible.

[The Chair, Liberal, Francis Scarpaleggia]

Thank you.