Thank you very much, Mr. Chair, and thank you to all of our witnesses. Both presentations were excellent, but I'm going to confine my questions to Professors Behnke and Pukelsheim to follow up on the model you've proposed.
First of all, let me say that this is a very impressive paper. You should consider publishing it in an academic journal. I think it would meet the standards of most political science journals.
I want to ask about the compromise that I think you have been trying to incorporate as you apply Germany's model to certain practical considerations that exist in Canada. Of course, in Canada we have this very inflexible requirement that the number of seats in each province be fixed. There's no room for adjustment to that without an amendment to the Constitution.
As I understand it, you've said that on the one hand we have the problem that in Germany, and also in New Zealand, is dealt with through overhang seats, and on the other hand we have a goal of achieving as much proportionality as possible so that the total number of seats for each party in each province accurately reflects the proportion of the vote that was cast for that party on the party list vote.
You have gone in the direction of saying that when we face this conflict of proportionality versus every constituency seat being allocated to the person who wins it, we have veered or chosen in favour of the seats, not in favour of proportionality.
I assume that the way you've compromised is that you've made the number of list seats as high as possible. It can go only as high as 50%, I think, under any version of the MMP model, and you've gone right to the largest number in order to deal with that. Is that why the constituencies are only 50% of the total number of seats per province?