Electoral Reform Committee on Sept. 22nd, 2016

[Jean Rémillard]

Personally, I would not stop at just one specific method.

There is a series of criteria that could be ranked by importance. For example, one of the criteria could be the regional representation of a party. If a party has no representation, it would be a priority to give it an MP.

The parties can also make choices. In any given party, there are candidates who make a very good showing but usually do not win. That is another avenue to consider.

There are other criteria. I covered them in my full brief. I also studied the Quebec elections. There is a hierarchy of three or four criteria that could be used as a way to designate additional MPs that would not be arbitrary, that would not depend solely on the party, as do votes based on proportional lists, and so on. The party often calls the shots in this regard.

[The Chair, Liberal, Francis Scarpaleggia]

Thank you very much.

Mr. Ste-Marie, you have the floor.

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

Thank you, Mr. Chair.

Hello, gentlemen. Thank you for coming today to present your positions.

I would like to thank the team travelling with us that has done an excellent job since the start of the week. I would also like to thank my colleagues, especially Mr. Deltell, who joined us today. Thank you also to those in attendance.

My questions are fairly technical and I will start with you, Mr. Rémillard.

Mr. Boulerice asked you a question and your answer has left me even more confused.

Let's say we keep the current system with 338 MPs. You said we could add an average of six seats per election by way of compensation, based on the criteria of your rationalized majority system.

[Jean Rémillard]

Those are my findings so far.

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

Those seats will not be among the 338 other seats. So that would mean 338 + 6 seats. There are certainly other ways of establishing criteria to designate them, but they would not necessarily be attached to a riding.

[Jean Rémillard]

Not necessarily.

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

Okay. Thank you.

[Jean Rémillard]

I would like to add something in that regard. Six seats is an average. In some cases, there could be up to 20 or 23 seats.

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

So, in one election, we could have 338 + 2 MPs, and 338 + 20 MPs in another.

[Jean Rémillard]

That is based on the simulation I did using the 1963 election results for the Parliament of Canada.

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

Parliament is currently under renovation. Under that system, we would have to have a room with fold-up or fold-away seats.

[The Chair, Liberal, Francis Scarpaleggia]

You would need moveable walls.

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

Mr. Dutil, I find that the system you are proposing is very interesting. As my colleague Alain Rayes said, specifically, in terms of the basic politics, the fact that a candidate who is the second or third choice in a constituency might end up being elected could cause riots. You talked about iron bars, but—

[Yvan Dutil]

The Brits sent me their codes. Over there, they have had the fifth candidate declared the winner. That might indeed cause riots.

Switzerland has a system like that. It is an irritant, but people understand it. For them, the system is already proportional by region. In addition, the coefficients are either 1.0 or 1.1. The adjustments are small and it is much less of a shock. It can also be done in a hierarchy. We can take a province and do it in a region with three or four members. It ends up proportionally, but with fewer shocks. The multiplication coefficients are less scary.

In England, they had no electoral threshold. I think there were 12 parties. Their proportional system was adjustable between 0 and I, meaning pure plurality and pure proportionality. Halfway along, the top three were in play, which was already less of a shock. There is also a question—

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

Let me interrupt you, because my time is slipping by. Thank you very much.

The fact that they use it in Switzerland does not convince me that it is acceptable. Switzerland is a curious country made up of a number of nations. For their system to work and for things to happen, all parties need to be unanimous. That means things move slowly.

I would like to ask each of you to do a brief critique of the compensatory mixed-member proportional system, as it has been debated in Quebec. Why did it not convince you as a system?

[Yvan Dutil]

The problem is that if you increase the number of members by 30% or 40%, it may well be unpopular. Previously, I proposed 10 more members in Quebec and that caused a problem. In addition, you are going to have to increase the size of constituencies by 30% or 40%, which may be very unpopular too.

The other limiting factor is that it cannot be done before the next election. You need at least two years to draw up a new electoral map. And there is always drama because some constituencies in Canada are the size of a country.

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

Thank you.

Mr. Rémillard, what do you think?

[Jean Rémillard]

The mixed voting system in Germany is definitely seen as a pure proportional system. A proportional system comes with problems. For example, if there is no quorum, no internal limit, there will certainly be a large number of parties. Quite often, you will end up with an elected parliament that is absolutely impossible to control. That is what allowed the rise of Hitler, by the way. It can go that far. But the Germans understood something at the time.

With a pure proportional system, there is a real danger that there will be confusion in the parliament. At times, there will have to be a “referee”, someone who really holds the power. So, at the time, the Germans decided to give the president of the republic the power to make decrees. It was as a result of that confused situation that Hitler began his rise. That is not a good way to go.

The other way is to cut from underneath. I think we can do better than that. There is no reason to cut off the Green Party. In the German system, the Green Party would have no seats, unless they got three members elected.

[The Chair, Liberal, Francis Scarpaleggia]

Thank you very much.

Speaking of the Green Party, we are now going to ask Ms. May to take the floor.

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

Thank you very much, Mr. Chair.

My thanks to the witnesses for joining us today.

I am really delighted to be here. This is not a good thing to say politically, but Quebec City is my favourite city in Canada. I am really happy to be here because it is magnificent.

In my opinion, today’s testimony contains the most innovative proposals that we have heard since we have been working on this task of finding a new voting system that will benefit Canadians and Quebeckers.

I have difficulty understanding all the proposals you presented because French is not my mother tongue. However, I do understand that your two presentations contain some absolutely unique features. Presenting unique proposals is not a bad thing, but I would like to understand them better.

Mr. Rémillard, thank you for thinking of the Green Party. In my opinion, it is important for each vote in our electoral system to be equal. I think that Canadian voters expect each vote to be equal.

In your view, is it possible to explain to voters that a system may change the weight of each vote, but that each vote remains equal, such as in Zurich, Switzerland, specifically? Has any research been done to determine whether or not voters in Zurich actually think that their vote will be equal in a system where votes have different weights?

[Yvan Dutil]

I don’t think there is real research. I have a comment on my blog, as I mentioned previously, from another blogger, a Swiss scientist. He told me that there was a problem with the interactions because their proportional system is global. So there is interaction between the local and the global. That is an irritant, but for others, as Apple or Microsoft would say,

it's not a bug, it's a feature.

[Some hon. members]

Oh, oh!

[Yvan Dutil]

The algorithm is a little difficult to understand but people can certainly be assured that it provides the right result. The electoral officer issues an Excel file with the correction coefficients and people can check that the result is correct. The proportionality is easy to calculate; everyone can do it with Excel at home, and it gives the right result. So people have a degree of confidence.

In the canton of Geneva, it was proposed but not accepted. The argument from the politicians was that, in that canton, there was no electoral threshold and if they were to proceed, they would need an electoral threshold. They had never had one and they did not want one. An electoral threshold would mean that some parties would fall below the threshold.

This has been studied in a number of places. Everyone has the same concern. It's not super-complicated but it comes as a bit of a shock. Votes are not equal at the moment. So you are making them equal again. If you understand that, everything is fine. The algorithm is not very long—20 to 30 lines—but it is a bit mysterious, which is where the psychological barrier lies. When you talk about compensating with a second category of members, it is more intuitive. You add members in order to achieve balance. The advantage is that you can keep the geography much more precisely. In the case of Canada, that is more of an advantage.

It has been debated in the United States, but not very publicly. In England and in the Faroe Islands, it was examined a little. Often, it is in countries that already have a regional proportional system. They have a problem because there are never fewer than three members in an electoral district. That creates distortions because the small parties in three… The psychological distance is less because those places already have a proportional system. The others see no proportionality because the electoral thresholds are too high.

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

Are the proportionality algorithms decided before or after the elections? That is not clear. Is it before the elections? I feel that it is only afterwards. It's difficult to calculate algorithms afterwards and to explain the decisions to the voters. For me, that is a bit of an obstacle.

[Yvan Dutil]

The algorithm in question could be in the public domain a long time in advance. People can do simulations. It took decades before it was done successfully. The algorithm is used to create balance. Originally, it was used to make nice statistical tables that provided results at 100%. With whole numbers, it is more complicated. It took decades to establish the algorithm that provided exactly the right number of members.

It has been in existence for about a decade. There are three or four versions. I am in contact with all the experts in the world. They have told me that, if ever I need help, they would be happy to provide it.

The real difficulty is in making sure that people trust it. Of course—