My response to Stéphane Dion
Thank you for mentioning my simulation in your speech to the Green Party Saturday.
Your speech was persuasive, and obviously well-received.
I was very interested in your comment that, under the recommendations of the Law Commission of Canada, “The number of candidates per party included on these lists would vary immensely, from just one in Prince Edward Island to about a dozen for some regions in Quebec and Ontario. This would effectively divide Canada up into several political microclimates. I would find that worrisome.”
So would I. Luckily, that was NOT what the Law Commission recommended. That was only their “demonstration model,” an example of how the calculations worked. For this purpose they had to use large regions because they used the 2000 election results. In Quebec in 2000 the PC Party got only 5.6% of the vote for three list seats, and the Canadian Alliance got only 6.2% for four list seats. Dividing Quebec into five regions of about 15 seats each would have largely excluded those parties. The same for the NDP in Alberta, if Alberta was split in two.
But the Law Commission said “adding an element of proportionality to Canada’s electoral system, as inspired by the systems currently used in Scotland and Wales, would be the most appropriate model for adoption.” Scotland has 16-MP regions. Wales has 12-MP regions. The Law Commission’s actual recommendation left the size of the regions open.
And they recommended “5. Within the context of a mixed member proportional system, Parliament should adopt a flexible list system that provides voters with the option of either endorsing the party “slate” or “ticket,” or of indicating a preference for a candidate within the list.” Those sample very large regions would have meant “bed-sheet ballots.”
That is why Prof. Henry Milner’s presentation of the Law Commission model used five regions in Quebec, seven in Ontario, two in Alberta, and two in BC.
He presented this at an electoral reform conference Feb. 21, 2009, in Toronto sponsored by Fair Vote Canada. As you will see, all across Canada the regions average 14 or 15 MPs. Of course, in the four Atlantic provinces they average, of necessity, 8 MPs. That is not an immense variation, nor a micro-climate.
My main concern with your model is that it will not pass a referendum in the 55% of Canada where communities have a single MP.
In Quebec it will be more acceptable because Quebeckers are accustomed to your 17 administrative regions. Ontario is not so lucky. We have 14 health regions, 31 public school boards, and no consensus on regions.
Between the GTA and Ottawa are nine single-MP-community ridings who will not like your model. Northern Ontario has nine ridings; seven of them will not like it. (Thunder Bay might like a northwestern three-seater.) Sarnia, Chatham, St. Thomas, Muskoka, Orillia, Midland, Collingwood, the Bruce Peninsula, Stratford, Oxford, Cambridge, Guelph, Orangeville, and many other such communities will reject it. If you insist on five-MP districts, even Charest’s mixed compensatory model with five MNAs (three local, two compensatory) would be more acceptable in single-MP communities.
My other concern is the high threshold of about 15%.
No, I do not insist on pure proportionality. On the 2008 votes, Milner’s model gave the Greens 18 MPs, not the 21 which province-wide proportionality would have given them. The NDP was short two, the Liberals got a bonus of two, and the Bloc got a bonus of three. That’s okay. In the UK, the Jenkins Commission recommended regions averaging eight MPs. That’s a high threshold, but not as high as your 15%. But you would not propose a region of eight MPs at large, because it is not local enough in most regions outside Montreal.
So I still think only a mixed compensatory model will suit Canada.
A moderate model is better than nothing. The NDP likes a 5% threshold, which the Greens can live with. I can see the Liberal Party and the NDP agreeing on a moderate model in 2015. For example, with Alberta’s new 34 ridings, metropolitan Calgary and Edmonton could each be 11-MP regions (7 local, 4 compensatory), and the rest of Alberta 12 MPs (eight local, four compensatory).
But I am sure you know that the Turkish 10% threshold, which they adopted to keep out the Kurdish party, is regarded in Europe as undemocratic. You do not want to be in such company, do you?
(One point Dion made, I agree with. He says “People like Elizabeth May on Vancouver Island and Ralph Goodale in Regina would have probably garnered more votes for the Greens and the Liberals.” In my previous post I said “In Dion’s model, due to his small districts, I calculate the results as 128 Conservatives, 115 NDP, 47 Liberals, 18 Bloc, 0 Green. . . . Dion's "moderate" model might be good for the Greens after all, if they got more votes.” So I agree Elizabeth May and Ralph Goodale would very likely have won, making 128 Conservatives, 113 NDP, 48 Liberals, 18 Bloc, and 1 Green.)