Sunday, January 29, 2012

Could Canada have 105 proportional MPs and scrap the Senate?

Could Canada have a proportional parliament while keeping all our local ridings, by enlarging the House of Commons? From time to time, someone suggests scrapping the Senate and adding a new tier of 105 MPs in place of the 105 Senators, with no net additional Parliamentarians.

First, scrapping the Senate needs unanimous consent of the provinces, while a normal proportional representation system needs no constitutional amendment.

That’s because voters still elect MPs from each province proportionate to its population. That’s the basis of the model recommended by the Law Commission of Canada: the number of MPs from each province stays the same, but about 65% are elected from local ridings while about 35% are elected regionally as “top-up” MPs.

That’s the Mixed Member Proportional system (MMP), combining direct election by electoral district and proportional representation. We still elect local MPs. Voters unrepresented by the local results top them up by electing regional MPs. The total MPs match the vote share.

However, let’s assume fast consensus on abolishing the Senate.

(Otherwise, onward to the Law Commission Report. For example, NDP policy in the 2011 platform and the March 3 House Motion is “We will propose electoral reform to ensure Parliament reflects the political preferences of Canadians. To this end we will propose a new, more democratic voting system that preserves the connection between MPs and their constituents, while ensuring parties are represented in Parliament in better proportion to how Canadians voted. Your vote will always count.” “That the House appoint a Special Committee for Democratic Improvement, whose mandate is to engage with Canadians, and make recommendations to the House, on how best to achieve a House of Commons that more accurately reflects the votes of Canadians by combining direct election by electoral district and proportional representation, and that the Committee shall report its recommendations to this House no later than one year from the passage of this motion.Here’s what that would have looked like.)

If fast Senate abolition is possible, would 105 “top-up” MPs really be enough? With 338 MPs to be elected in 2015, those 105 would be only 23.7% of 443. That’s “MMP-lite.” No MMP model now operating has fewer than 33% “top-up” MPs (Wales).

Speculating on the 2015 election results is not easy. But we can see how it would have worked to add 105 “top-up” MPs to the 308 elected in 2011: 25.4% of 413.

This projection assumes voters voted as they did in 2011. In fact, if voters knew every vote would count, more would have voted -- typically at least 6% more. And some would have voted differently, perhaps 18% of them by one study. No more strategic voting. We would likely have had different candidates -- more women, and more diversity of all kinds. Who knows who might have won real democratic elections?

Open regional list MMP

With the Law Commission model, the regional MPPs are the party's regional candidates who get the highest vote on the regional ballot. The voter casts one vote for local MPP, and one for their party and (if they wish) for their favourite of their party's regional candidates. An exciting prospect: voters have new power to elect who they like.

So we’re talking about MMP with open lists from middle-sized regions. All MPs have faced the voters. All MPs are “locally anchored” and accountable.

In Ontario’s 2007 referendum, 63.1% voted against that MMP model, with closed province-wide lists. About 31% were simply against proportional representation. Many more were voters who wanted all MPPs to be personally elected, not on closed lists. Many more were voters who wanted all MPPs to be anchored in their own region, not on province-wide lists. Another 7.5% were voters outside Toronto who disliked province-wide lists even more than Toronto voters did.

How proportional is MMP-lite?

I’ve compared my spreadsheet’s projection of MMP-lite with a fully proportional result, and assumed each region and province gets 34% more MPs.

When you look at regions, you see many differences between MMP-lite and a normal MMP model. Not bothering with one-seat differences, here are the bigger ones:

Peel-Halton with 12 local MPs and 4 regional: Con. 12 (should be 8), Lib. 2 (should be 5), NDP 2 (should be 3).

Montérégie with 13 local MPs and 4 regional: NDP 13 (should be 8), Bloc 2 (should be 5), Con.1 (should be 2), Lib. 1 (should be 2).

York-Durham with 12 local MPs and 4 regional: Con. 11 (should be 8), Lib. 3 (should be 4), NDP 2 (should be 3), Green 0 (should be 1).

Quebec City and Eastern Quebec with 18 local MPs and 6 regional: NDP 13 (should be 9), Con. 5 (should be 7), Bloc 4 (should be 5), Lib. 2, Green 0 (should be 1).

Central Ontario (Belleville to Owen Sound) with 12 local MPs and 4 regional: Con. 12 (should be 9), Lib. 2 (should be 3), NDP 2 (should be 3), Green 0 (should be 1).

Laurentides--Lanaudière--West & North Quebec with 13 local MPs and 5 regional: NDP 13 (should be 9), Bloc 3 (should be 5), Con. 1 (should be 2), Lib. 1 (should be 2).

Luckily, on the 2011 votes, these tend to cancel each other out.

When you look at each province, again you see many significant differences.

Ontario: Con. 75 (should be 64), NDP 33 (should be 37), Lib. 30 (should be 36), Green 4 (should be 5).

Quebec: NDP 59 (should be 44), Bloc 17 (should be 24), Con. 12 (should be 17), Lib. 12 (should be 14), Green 1 (should be 2).

Saskatchewan with 14 local MPs and 5 provincial: Con. 13 (should be 11), NDP 5 (should be 6), Lib. 1 (should be 2).

New Brunswick with 10 local MPs and 3 provincial: Con. 8 (should be 6), NDP 3 (should be 4), Lib. 2 (should be 3).

Luckily these again tend to cancel each other out.

So Canada-wide with 413 MPs we get: Conservatives 178 (should be 168), NDP 137 (should be 128), Liberals 70 (should be 78), Bloc 17 (should be 24), Greens 11 (should be 15). Luckily, not bad at all. An NDP-Liberal-Green coalition government would have had a majority either way: 218 (should be 221).

But one can easily imagine an election where MMP-lite would let voters for one party elect a false majority with less than 50% of the votes.

Conclusion: MMP-lite can get close to proportional results, and is certainly a vast improvement over our undemocratic winner-take-all system. However, it sometimes does not make every vote count equally.

Note: This simulation uses no legal threshold. If there was a legal threshold of 5% in each province, Green voters would have elected no MPs from Ontario, Quebec, Manitoba and Nova Scotia, electing only 4 MPs not 10. But of course in a real MMP election Green voters would likely have cast over 5% of the vote in many provinces.

Saturday, January 21, 2012

Where will British Columbia's six new seats be?

The six new seats will be four in the Lower Mainland, one on Vancouver Island, and one in the Interior, using the 2011 census figures.

Growth in Kelowna (now too big for a single riding), Kamloops, etc. requires a new Interior riding. Four present Interior ridings are all well over quotient: Kelowna - Lake Country is 28.6%, Okanagan - Shuswap is 15.6% over, Kamloops - Thompson - Cariboo is 14.8% over, and Okanagan - Coquihalla is 8.7% over.

The North and the Kootenays can, in my opinion, remain unchanged. The present three Northern ridings do not, as a whole, over-represent the North. The present Kootenay--Columbia riding is almost 16% below quotient, but it can remain without causing other areas to be under-represented, and with six new ridings for BC, it would be shocking if the Kootenays had to lose representation.

So here's my prediction of what the Boundaries Commission is likely to do.

Details, referring to quotients from the census (104,763):
The Lower Mainland, including Sunshine Coast and the Squamish area, has 2,637,911 residents, 25.1798 quotients, 25 ridings. Today it has 21, one of which includes Powell River, and 14% of Chiliwack—Fraser Canyon is in the Interior.

Coast-North Shore (not including Powell River or Whistler):
WEST VANCOUVER--NORTH VANCOUVER--SUNSHINE COAST (including Squamish) 10.4% over quotient.
NORTH VANCOUVER 10.98% above quotient.

Vancouver-Richmond-South Delta gets 8 (now 7).
VANCOUVER EAST 5.09% over quotient, unchanged The other ridings would be about 0.78% over quotient.
VANCOUVER CENTRE loses 31,692 residents to a new VANCOUVER FAIRVIEW.
VANCOUVER KINGSWAY gains 38,236 from Vancouver South and loses 57,597 to VANCOUVER FAIRVIEW.
VANCOUVER SOUTH-RICHMOND NORTH gains 20,133 from Richmond.
RICHMOND STEVENSON (the remaining Richmond riding) might be renamed.
DELTA - RICHMOND EAST largely unchanged, move about 5,000 residents of the Hamilton neighborhood to New Westminster, about 1.57% over quotient.

Burnaby-New West-Maple Ridge gets 6 ridings (now 4.7).
BURNABY NORTH about 3.49% below quotient.
BURNABY SOUTH about 3.49% below quotient.
NEW WESTMINSTER--BURQUITLAM about 3.61% below quotient.
COQUITLAM--PORT MOODY about 3.43% below quotient.
COQUITLAM CENTRE—PORT COQUITLAM about 3.22% below quotient.
MAPLE RIDGE—PITT MEADOWS—PORT COQUITLAM (taking about 7,000 from Port Coquitlam) about 3.58% below quotient.
Surrey/White Rock-North Delta gets 5 (now 4).
The new SURREY FLEETWOOD riding might get 51,800 from Fleetwood, 19,600 from Newton, 19,400 from Cloverdale, and 17,600 from North Surrey.
The four present ridings would then be about 3.4% over quotient:

Langley-Abbotsford-Mission gets 3 (now 2.3).
LANGLEY loses about 28,000 Aldergrove residents, 3.28% under quotient.
ABBOTSFORD WEST—ALDERGROVE about 3.33% under quotient.
ABBOTSFORD EAST—MISSION about 3.63% under quotient.

CHILIWACK--FRASER CASCADE loses the areas outside the Fraser Valley RD, and is 0.86% below quotient.

The North, still 3:
SKEENA - BULKLEY VALLEY add 8,524 in Vanderhoof area, 5.96% below quotient.
CARIBOO - PRINCE GEORGE after transferring those 8,524, 4.24% below quotient.
PRINCE GEORGE - PEACE RIVER unchanged only 0.97% over quotient

Interior, 7, now 6
KAMLOOPS--SHUSWAP—THOMPSON including 65% of Kamloops City, 1.19% over quotient. About 67% of these residents are within the present Kamloops--Thompson–Cariboo. (The growing city of Kamloops is large enough for its own riding, but I cannot find a good way to do that.)
OKANAGAN NORTH--LAKE COUNTRY--NORTH KELOWNA, 4.7% over quotient. About 78% of its residents are in the present Okanagan—Shuswap riding.
KELOWNA has all but 7,000 City residents, 5.3% over quotient.
OKANAGAN SOUTH-CENTRAL (remove the 13,244 residents of Okanagan–Coquihalla who are in Thompson-Nicola), 3.98% under quotient.
The new riding of COQUIHALLA—LILLOOET—CARIBOO is centered on Merritt. It includes about 31,500 Kamloops city residents, along with 3,633 in Logan Lake and area. It includes 11,171 people in Merritt and area. It includes 32,621 residents of Squamish-Lillooet RD, all of the district west to and including Whistler. It includes 4,492 residents of Princeton and area who are, for example, in the same Nicola-Similkameen school district as Merritt. And it includes the 14,421 residents of Cariboo who are now in Kamloops - Thompson – Cariboo. This riding reminds one of Cariboo-Chilcotin which, from 1976-1987, included Squamish-Lillooet Regional District, parts of Thompson- Nicola Regional District, and Cariboo Regional District. About 48% of the residents of COQUIHALLA—LILLOOET—CARIBOO are now in Kamloops - Thompson – Cariboo, while the other 52% are now in three other ridings. It has 97,838 residents, 6.61% under quotient, quite appropriate for a far-flung riding.
BRITISH COLUMBIA SOUTHERN INTERIOR, after losing the Princeton area, 10.79% under quotient.
KOOTENAY—COLUMBIA unchanged, 15.98% under quotient.

Vancouver Island gets a new riding:
VANCOUVER ISLAND NORTH—POWELL RIVER (includes Comox), 1.3% under quotient.
NANAIMO (includes Lantzville), 2.84% over quotient.
COWICHAN—LANGFORD (includes Highlands), 6.6% over quotient.
ESQUIMALT - JUAN DE FUCA, loses Langford and Highlands but gains 8,000 from Saanich, 4% over quotient.
SAANICH—GULF ISLANDS loses 8,000 residents, now 3.8% over quotient.
VICTORIA unchanged, 5.9% over quotient.
Note that every riding is within a 10% deviation except the two remote ridings of Kootenay-Columbia and Southern Interior, and the two North Shore-Coast ridings are again just over 10% as they were in the 2003 report, to avoid a cross-Burrard-Inlet riding.