The multi-member district seats would not be selected based on the share of the votes the party received in those areas, but to compensate for the non-proportionality of the results therein. (Explained in the addtional member article linked above)
For example: Each region would have 9 ridings, each electing one member. Thereafter, the region would have 5 more members selected to represent the whole region. If Party A won 8 of 9 ridings with only 55% of the vote (i.e. they barely won each of eight, and barely lost one) and Party B had received only one seat on 45% of the vote, then Party B would receive all 5 of the regional members to compensate for having won only one traditional seat on 45% of the votes.
In order to presuppose the structure of the ridings, we'll have to overlay the existing 55 ridings into each of the regions suggested by the commission. That means 14 single member ridings will have to be transposed onto 3 of the regions and 13 will have to go into a final region. We'll place the Moncton-area districts into the 13 seat region, because those districts are the most overpopulated in the province.
The Commission recommended the following regions: North, Central, Southwest, Southeast. I will assign our current electoral districts to these regions as follows:
NorthA lot of these regions are less than ideal but this is the unfortunate result that can happen when you have to carve the province up into four equal-sized regions. Presumably when boundaries for ridings would have been redrawn, some would have been combined and others split in a fashion that would try to avoid this.
Grand Falls-Drummond-St. André, Restigouche-La-Vallée, Edmundston-Saint-Basile, Madawaska-Les-Lacs, Campbellton-Restigouche Centre, Dalhousie-Restigouche East, Nigadoo-Chaleur, Bathurst, Nepisiguit, Caraquet, Lamèque-Shippagan-Miscou, Centre-Péninsule-Saint-Sauveur, Tracadie-Sheila, Miramichi Bay-Neguac
Central
Victoria-Tobique, Carleton, Woodstock, York North, Fredericton-Nashwaaksis, Fredericton-Fort Nashwaak, Fredericton-Lincoln, Fredericton-Silverwood, New Maryland-Sunbury West, Oromocto, Grand Lake-Gagetown, Southwest Miramichi, Miramichi Centre, Miramichi-Bay du Vin
Southwest
York, Charlotte-Campobello, Charlotte-The Isles, Fundy-River Valley, Quispamsis, Rothesay, Saint John East, Saint John Harbour, Saint John Portland, Saint John Lancaster, Saint John-Fundy, Hampton-Kings, Kings East, Albert
Southeast
Petitcodiac, Dieppe Centre-Lewisville, Moncton East, Moncton West, Moncton North, Moncton Crescent, Riverview, Memramcook-Lakeville-Dieppe, Tantramar, Shediac-Cap-Pelé, Kent South, Kent, Rogersville-Kouchibouguac
So now let's try to re-run the 2006 election under NB MMP. In each region, we'll take the actual results from 2006 in terms of seats, then scale that down to 9 ridings, then apply the NB MMP method to each set of regional seats.
2006 actual results | Projected NB MMP seats | |||||
Seats | % vote | Riding | Regional | Total | ||
PC | 26 | 47.5 | 16 | 13 | 29 | |
North | 5 | 45.0 | 3 | 4 | 7 | |
Central | 5 | 46.5 | 3 | 4 | 7 | |
Southwest | 7 | 46.6 | 4 | 3 | 7 | |
Southeast | 9 | 55.9 | 6 | 2 | 8 | |
Lib | 29 | 47.1 | 20 | 7 | 27 | |
North | 9 | 48.7 | 6 | 1 | 7 | |
Central | 9 | 48.3 | 6 | 1 | 7 | |
Southwest | 7 | 47.0 | 5 | 2 | 7 | |
Southeast | 4 | 39.5 | 3 | 3 | 6 | |
NDP | 0 | 5.1 | 0 | 0 | 0 | |
North | 0 | 6.0 | 0 | 0 | 0 | |
Central | 0 | 5.2 | 0 | 0 | 0 | |
Southwest | 0 | 5.6 | 0 | 0 | 0 | |
Southeast | 0 | 4.7 | 0 | 0 | 0 |
I am not sure if this is the result that PR proponets would have been looking for. The right-of-centre party, which won a bare plurality of votes (47.5%) would win a majority government, while left-of-centre parties who received 52.2% of the votes would be in the minority. The NDP, despite winning 5.1% of the vote would remain unrepresented in the legislature. The Liberals, despite winning a plurarilty of the vote in 3 of 4 regions, will have split the seats evenly in each of those regions, which seems somewhat unfair.
As I said in a comment to the last post on this, "the problem with the electoral reform debate in Canada (is that e)veryone seems to see it as black and white, when in fact there are countless models to choose from."
Using the model that was proposed by the Commission on Legislative Democracy would not have improved, in my view, the results of the last election. Regions with an odd number of members would likely have worked better, or perhaps the top up could have been done on a province-wide scale which would have allowed the NDP to get seats. Or we could have moved wholly to multi-member ridings (which our province used from pre-confederation to 1974 anyway) using the single transferable vote. Or we could have gone to countless other varieties of options.
As I said, that is the problem with the PR debate. When it is the status quo vs. the nearly infinite number of alternatives, change is quite popular. But when it is the status quo vs. one particular PR model, then these tend to fail (as votes in BC, Ontario and PEI have shown).
I suspect that most members of the NDP support PR in principle. I doubt many would favour the NB MMP model if they knew it would have meant they would still have been shut out of the legislature, and that a conservative government would have been re-elected with a larger majority in the last election.
This is why I've always prefered a gradual electoral reform. Canada was born out of evolution as opposed to revolution and it has served us well. We should do the same with any changes to our electoral system. A move from FPTP to a the preferential vote would be a relatively simple change that wouldn't turn our system on its head, and wouldn't confuse voters, but would answer many of the legitimate grievances of electoral reform proponents. Thereafter, if our objectives haven't been met, we could move further down the road.
Doesn't this seem more reasonable than totally replacing a system that, though imperfect, has served this country for well over a hundred years?