Redistribution of Federal Electoral Districts 2012

[Linus Van Pelt]

Come on team, we've been through this before. The exact census numbers will be released in Feburary 2012, so of course we don't know the exact seat distribution till then.

[Linus Van Pelt]

No, they aren't.

Here is the bill as currently tabled:
http://parl.gc.ca/HousePublications/Publication.aspx?Language=E&Mode=1&DocId=5202547&File=24#1

There is no mention of specific numbers for any province. The key operative part of the law is here (aside from a bunch of minor amendments to make the procedural parts of the Elections Act consistent with it):

You must be logged in to read this quote.

[Linus Van Pelt]

This helps Tories, potentially Grits.

http://www.theglobeandmail.com/news/politics/john-ibbitson/adding-seats-to-house-of-commons-a-political-windfall-for-tories/article2271940/

This article is just hopelessly confused. The seats created in the new bill are not concentrated in fast-growing areas of the provinces in question. There are two separate issues here. Even under the current formula, seats are reallocated after every census, and so seats are added to high-growth areas. This is not a result of the new bill. The inter-provincial inequities amended by the new bill are province-wide inequities. After each redistricting under the current formula, Windsor is just as underrepresented as Brampton. It's true that Brampton becomes more underrepresented as the decade goes on relative to Windsor, but this would be fixed after the census with or without the bill. It's the underrepresentation of both Brampton and Windsor relative to Winnipeg and Halifax that gets (well, really only partially) alleviated by the new bill.

Thus the partisan effect of the new bill doesn't depend on strength in the 905 and Lower Mainland, but on strength in the three fast-growing provinces in general. Now, in fact it does help the Tories, but for reasons other than those articulated by Ibibitson. Certainly it does not help the Liberals under conditions in which the majority of their caucus in east of the Ottawa River, granting that those conditions might not last as DL says.

[Linus Van Pelt]

So, I realize now I was mistaken about the timeline and the use of estimates; my apologies. I think RogueBeaver was right about the provincial numbers.

Some time between this conversation of ours and now, Elections Canada has updated its web site to include this timeline that, as you can see, somewhat confusingly has at the top the sentence "The process is set to begin in February 2012 when Statistics Canada releases the population numbers from the 2011 Census", but then the first item in the timeline is the allocation of seats by province on December 16, 2011 (just a day after this thread ended above) based on "estimates" provided to Elections Canada by the Chief Statistician. There is also now this other website giving the official provincial seat counts, and they are exactly as RogueBeaver was saying.

I was confused about why they would use the estimates rather than the census numbers, but Elections Canada links to this Statcan website that claims that for overall population numbers, though not for smaller groups, they think the estimates are more accurate than the census itself due to correction for undercoverage based on "postcensal coverage studies of a representative sample of individuals". So (at least as far as I can tell) they use the December estimates to allocate by province even if the census numbers in February don't actually add up to the same sum. This is different from how it is done in the U.S.

[Linus Van Pelt]

No sub-provincial regional groupings or individual ridings are legally mandated; it has been at the discretion of the commissions. Hopefully it will stay that way.

[Linus Van Pelt]

I find this quite awkward in both of the areas that have been changed the most (west-central downtown and the Don Valley ridings). The northern half of the current Trinity-Spadina is much more oriented towards the inner city than to northern and eastern St. Paul; going from College and Ossington up to Forest Hill and Yonge and Eglinton just feels very unnatural.

With Don Valley Central and South, meanwhile, you now have two ridings that combine a very affluent non-immigrant west side with inner suburban South Asian areas in the east. The existing Don Valley West has this of course, but at least it has a certain logic from using the river valley itself as the boundary.

Since Trinity-Spadina is the most overpopulated riding and is the merger of two historic ridings, I would think a natural starting point might be to recreate Spadina and Trinity, with Spadina including the rapidly growing condo areas on the fringe of the downtown core and Trinity combining the southern half of Davenport with roughly the western third of Trinity-Spadina, which together form a natural constituency of very urban but still low-rise row housing.

How are you doing these, by the way?

[Linus Van Pelt]

Populations down to the block level: http://geodepot.statcan.gc.ca/GeoSearch2011-GeoRecherche2011/GeoSearch2011-GeoRecherche2011.jsp?lang=E&otherLang=F

Thanks! Not exactly the most user-friendly system , but more power to you for making all those maps with it.

You must be logged in to read this quote.

Well, there's always a bit of a question as to what's "the" new riding and what's an existing riding shifted half-way over, but I see your reasoning; definitely the north-central city has to be shaken up a bit somehow. If the Scarborough ridings are underpopulated though another possibility would be to put the east end of Willowdale with Agincourt; these areas are pretty similar to each other, and I don't really see why Scarborough should be sacrosanct when the other old municipalities are being ignored.

BTW, they're not going to divide up Davenport, because the Portuguese community neatly fits into that riding. I think that's why the riding exists with those boundaries.

The boundaries commissions haven't explicitly used ethnic criteria like that (aside from language and Aboriginal %, both of which are more constitutionally important), and I think it would be pretty controversial if they did. (No VRA here! ). But we'll see how it goes.

[Linus Van Pelt]

Playing around with the geosearch tool, I guess there still isn't really enough population for three west-side districts south of the tracks, so that Trinity plan won't work. But I still think it makes more sense to shift some territory in T-S's northwest quadrant over to Davenport, which would then have to lose its north end, than to bring St. Paul's so far south. Those formerly non-residential condo areas in the southeast of the riding where the growth is would also fit well with Toronto Centre, but that would bring the cutoff of that riding south of Rosedale into the high-rise area, which maybe wouldn't be so great.

[Linus Van Pelt]

New Brunswick proposal is out.

http://www.redecoupage-federal-redistribution.ca/content.asp?section=nb&dir=now/proposals&document=index&lang=e

Basically the current map with some changes, particularly the urban ridings shrinking due to population growth. Fredericton loses some rural areas, and the Moncton riding, perhaps more controversially, loses Dieppe (to Beauséjour), which drops the % Francophone significantly.

They're still giving a rural exception variance to Miramichi (-28.6% population) which seems a bit dubious since by national standards it's small and non-northern.

[Linus Van Pelt]

It seems bizarre that provincially Toronto used the old ward names, despite not being in existence for 100 years. Speaking of which, why is St. Paul's called that? There was never a municipal ward named that.

I'm not exactly sure about this, but I believe that it may be named for St. Paul's Methodist (later United) on Avenue Road.

[Linus Van Pelt]

Naming aside, I generally like this Montreal proposal better than the current map; the whole west-central region, especially the line along a little residential street through NDG, was really awkward on the old map.

[Linus Van Pelt]

Actually, upon closer inspection in the Mapviewer, there is a weird practice on the new map of putting boundaries one street over from a main street, thus having both sides of the main street in one riding. For example the boundary of Papineau runs along Birnham instead of Acadie, marooning one block of Parc Ex in John Peters Humphrey. Odd.

[Linus Van Pelt]

Does Cotler have an issue with more working class areas being added?

While you're right that the changes are small, John-Peters-Humphrey is a bit less working-class and probably a bit more winnable for the Conservatives than the old Mount Royal. It loses the main section of Cote-des-Neiges, a low-income area with many English-speaking (Black Caribbean and South Asian) immigrants that stayed very Liberal in 2011, while gaining some generic suburbia by the airport that voted NDP and a few pretty well-off polls in that sort of transition zone between NDG and Westmount where all three federalist parties are competitive.

[Linus Van Pelt]

Its very hard to say what will happen in Mount Royal next time (whatever it ends up being called). I think that the Tories hit their absolute ceiling there last time with 36% ...

No. The Sephardic and Orthodox populations are growing rapidly here due to immigration and large families; this is one of the reasons for the recent increased Conservative strength.

[Linus Van Pelt]

I will be interested to see the redistributed results for Saskatoon West and Regina-Lewvan; eyeballing things on The506's web site, neither of them looks unwinnable for the Tories in a good year. For this reason I'm not really convinced that this is a favourable division of Regina for the NDP; the poorest part of the city in North-Central is still marooned in the rurban Regina-Qu'Appelle, which went CPC by 15 points, whereas if this part of the city were in the new urban riding it would likely be a safe NDP seat.

It will also be interesting to see whether the NDP can make a serious play for Wascana in the event Goodale stays around. The strategic considerations here are kind of complicated, since it's no longer winnable for a Conservative in a two-way race, but it probably still is in a three-way race.

[Linus Van Pelt]

I like the way they've done Toronto; the boundaries are mostly pretty natural. Both St. Paul's and Toronto Centre look winnable for the NDP to me.

That Haliburton-Uxbridge subrural thing looks very odd.

It's basically a rural riding; suburbia in Durham is still mostly out along the lake.

You must be logged in to read this quote.

This makes sense I think; it's the rural parts of the "City" of Hamilton - the old A-D-F-W was a weird rurban riding.

Scarborough-Rouge River is clearly no more.

If you look at the 2011 poll map for Pickering-Scarborough East, though, the Scarborough part off the lake is mostly somewhat similar areas that voted for the Liberal incumbent McTeague. Barring a Liberal revival I think Sitsebaiesan should be able to pick up these areas.

I totally agree Earl about keeping the overpopulated ridings. Essex is particularly weird - at least with Kingston or Guelph you can kind of see the logic of keeping one city as a riding.

[Linus Van Pelt]

Nice find!

So, assuming the site is correct, the SK urban results are as follows - quite swingy, really. Clearly there are proportionally more people in that eastern suburban section of Saskatoon C-U than we thought.

Regina--Lewvan: NDP 45.9 CON 43.14
Regina--Qu'Appelle: CON 53.31 NDP 38.37
Wascana: LIB 41.79 CON 35.67 NDP 20.09
Saskatoon Centre-University: NDP 44.55 CON 43.92
Saskatoon--Grasswood: CON 49.38 NDP 40.52
Saskatoon West: CON 48.30 NDP 43.12

Also of miscellaneous interest-
Esquimalt--Colwood: NDP 40.61 CON 36.97
South Cowichan--Juan de Fuca: NDP 43.13 CON 42.91
Vancouver Granville: CON 32.85 NDP 29.19 LIB 27.86 (so actually quite winnable for the NDP, considering the strategic votes in the section taken from Quadra).
Toronto Centre: NDP 37.58 LIB 37.1
St. Paul's: NDP 36.43 LIB 33.54 (!)
Don Valley East: LIB 39.14 CON 35.81.
Brampton--Gore: NDP 39.07 LIB 31.29 CON 27.27
Scarborough East: LIB 34.76 CON 31.62 NDP 30.95
Scarborough North: NDP 35.42 CON 33.38 LIB 28.80
John-Peters-Humphrey: LIB 40.61 CON 32.76 NDP 20.62


Of course overall many CON gains in the Anglo suburbs.

(edit: silly mistake in original post).

[Linus Van Pelt]

Some other ones of note -

Winnipeg North: NDP 37.06 CON 30.55 LIB 29.44 - lol
Welland--Fort Erie: CON 43.09 NDP 38.71 - tough, but not impossible for Allen under a small national swing to the NDP
South Okanagan--West Kootenay: CON 44.72 NDP 39.76 - similarly
Markham: CON 38.35 LIB 36.02
Markham--Unionville: CON 46.38 LIB 33.77 - McCallum's in tough with this combo
Avalon: CON 38.83 LIB 35.21 NDP 24.71

[Linus Van Pelt]

Has no one yet produced the transposition list of proposed new ridings showing the number of voters transposed from each old riding? So you can see which riding is successor to which, and which are the "new" ridiings?

Eligible voters, I'm not sure, but the transposition of 2011 votes is at the link Earl gave above.

[Linus Van Pelt]

Do these boundary commissions not employ someone who knows how to use a GIS program? I would think that any one of us who know our way around the province could sit down with Alcon or Averroes Nix and get this done in a couple of days.

[Linus Van Pelt]

I just had coffee with Rosario Marchese (my MPP) he's speaking today at the Toronto hearings. He's looking for a) the commission to take into account future population growth (like the BC commission) and b) following that thought, Trinity-Spadina should be basically split in half, two new ridings, due to the fact that there is expected to be 50-60k more people in the riding (which is 149k already).

Anyone agree with a) and/or b)

I think both are good moves the commission should make; but that being said it would mean, possibly three ridings in the proposed two ridings of TS and TC.

"Taking into account population growth" doesn't actually reduce malapportionment; it just shifts it from the end of the ten-year cycle to the beginning. If there's going to be malapportionment somewhere, I'd prefer that at least it was based on concrete census data rather than more uncertain projections. Also, although this reasoning doesn't depend on partisan considerations, this is also really really not a game the NDP wants to get in, because most high-growth areas are still Conservative suburbs.

Now, I once suggested on the forum that a split of Trinity-Spadina on east-west lines might make sense, but without very underpopulated ridings it pushes Toronto Centre far enough that it has to go up into Rosedale. I actually don't think this is as bad an idea as the rest of the internet does, since really any version of Toronto Centre is going to consist of a grab-bag of very unlike areas (is Church and Wellesley more like Regent Park than like Rosedale? Not actually all that clear), but I can see the motivation against it.

[Linus Van Pelt]

Sounds like everyone just wants to keep their own riding as safe as possible, with no co-ordinated strategy.

[Linus Van Pelt]

That is not at all a community of interest. It might make sense to call it such if all the government had the power to do was to tax income and redistribute it directly in the form of cash grants. But in reality most government spending is on programs whose effect on people of different ages is very different, and a lot of regulation affects different urban forms differently. The waterfront is mostly childless young professionals; Rosedale is inhabited by wealthy middle-aged parents and seniors. They are affected quite differently by policies on education, pensions, health care, etc. The difference between the areas is even more pronounced with regards to zoning and transportation - TTC funding, island airport, the OMB, etc. Try building a 30-story condo at the corner of Spadina & Bremner, and then try building one at Glen Road and South Drive, and you'll get a different reaction.

[Linus Van Pelt]

The "disposition of objections" in the report (link) seems to suggest that Benj's guess is exactly right; the MP's used community of interest between the two sides of the port as their reason for objecting, and then the commission took them quite literally in the interests of population equality.

It's not a totally silly concern; port authorities are federally managed and are often large landowners, which sometimes leads to controversial relations with municipal planners.

[Linus Van Pelt]

Good names, and good map, overall. Just a few suggestions:

- Calling 16 "Lawrence Park" seems a little strange, since that's normally used for the area just east of Yonge. (I realize the school's west of Yonge). Maybe "Avenue Road"

- I would call 20 just "Spadina". Trinity-Bellwoods park, which is the source of the Trinity name, is actually just west of the ward.

- It's a bit weird to use "Don Valley South" for 26, since the river and valley run down to the lake. Maybe "Leaside-Thorncliffe".

- Don Mills is the neighborhood around Don Mills Road and Lawrence, which is not in Ward 33. "Fairview" might work.