Thursday, September 30, 2010

Canada: Multipartisanship per province

Here's a quick-and-cheap way to consider the extent to which any given province is truly multipartisan. It works like this: for each province, for the federal government and for the Yukon, I'll look at the most recent election and add together the percentage of popular vote for each of the top two parties. I've been thinking about it, and it really is a simple but ultimately practical test of true multipartisanship.
  • 94.3%: Prince Edward Island: Liberals and PC.
  • 91.3%: Newfoundland and Labrador: PC and Liberals.
  • 88.2%: Saskatchewan: SaskParty and NDP.
  • 88.0%: British Columbia, Liberals and NDP.
  • 85.9%: Manitoba: NDP and PC.
  • 83.3%: New Brunswick, PC and Liberals.
  • 79.2%: Alberta, PC and Liberals.
  • 77.3%: Québec, Liberals and PQ.
  • 75.3%: Yukon, Yukon Party and Liberals.
  • 73.9%: Ontario, Liberals and PC.
  • 72.4%: Nova Scotia, NDP and Liberals.
  • 63.9%: Federal, Conservatives and Liberals.
It didn't surprise me, looking at that list, that PEI was most bipartisan and that Nova Scotia was least. I was, however, surprised by how multipartisan Ontario appears as of the last election (and current polls are showing it even moreso). I also wasn't expecting the Federal election scene to trump the rest by almost ten percent - no surprise, thinking about it, but still quite telling about the current political reality in Canada.

I was also surprised to find that Western provinces by and large are not more multipartisan than the rest of the country, though of course in the West it's right-vs.-left as opposed to right-vs.-centre, confused by the fact that the rightmost party in Saskatchewan and BC uses a different name. Also interesting how the traditionally one-party hegemony in Alberta is not the case, as it shows up quite multipartisan by these standards. At the moment, of course, Alberta is more multipartisan than it has ever been before, with its whole electoral system in flux. If an election were held today it would be incredibly fascinating, whereas it's usually pretty sleepy cakewalks for the PCs.

Tuesday, September 28, 2010

New Brunswick and Disproportionality

Lately I've become interested in 'disproportionality', the extent to which a general election can produce a seat count in the House whose percentages differ from the percentages of the popular vote that produced them. If in an election, 48% of voters voted for Party A, 39% voted for Party B and 13% voted for Party C, and if that wound up with a House of Commons where 52% of the seats were filled with Party A members, 40% of the seats were Party B and the remaining 8% were party C, what would generally be considered relatively proportionate. By and large, there would be little dispute about whether the makeup of the House reflected the voting intentions of the people.

In fact, though, such proportionality is rare, particularly in Canada with its First-Past-the-Post and multiparty systems. Calls for abolition of FPTP (or 'winner-takes-all') election systems are strongest in multiparty systems; in bipartisan political systems, the effect is more minimal.

Canada is an interesting country, in that at the federal level and across each province, a number of parties vie for office. Federally, for example, we currently have five that can be called 'major'. Yet ultimately, federally and in most provinces it ultimately comes down to two parties with a realistic chance of governing and a group of others with, at best, a chance of holding sway in minority governments. The Atlantic provinces are an interesting case in point, ranging from PEI, the most truly bipartisan part of Canada, to Nova Scotia, which is perhaps the only legitimately multipartisan area of Canada at the moment. Somewhere in the middle lies New Brunswick.

For most of New Brunswick's history, it's been genuinely bipartisan: in recent years the NDP has at best been able to get one seat - the leader's - in the Legislative Assembly. Apart from that, and with one major exception (which I'll get to), the trend has been that in a provincial election the Progressive Conservatives and the Liberals will combine for about 90% of the vote. In other words, the vast majority of people in New Brunswick see their elections on a strictly binary basis.

When this is the case, disproportionality changes in importance: when one party squeaks above fifty percent, disproportionality either doesn't really matter (is there a real difference between, say, 60% of the legislature or 70%?) or is of terrible importance (if the elections return a simple majority of votes for Party A but a simple majority of seats for Part B). While looking blandly bipartisan on the surface, New Brunswick actually has had quite an interesting electoral history:
  • In both 1970 and 1974, the Progressive Conservatives got parliamentary majorities despite getting fewer votes than the Liberals, making fully eight years of majority government of questionable legitimacy.
  • In 1987, a landslide 60.4% of the vote (compared to 28.6% for the PCs and 10.6% for the NDP) returned every single seat in the Assembly to the Liberal Party, leading to four years of unopposed government.
  • The very next election in 1991 led to what appeared to be a complete restructuring of the electoral scene, as the Liberals got a second majority, but the upstart Confederation of Regions Party got eight seats, 21.2% of the vote and the status of Official Opposition, and both the PCs and the NDP has representation in a four-part parliament. Yet by the next election, the Liberals and the PCs were back to a combined vote of 82.5%.
One of the major talking points this time around was a return to multipartisanship. Theoretically, a rejuvenated NDP would become a major electoral presence, while the Green Party and the upstart People's Alliance would also be on hand to contest. At the outset of the campaign, it appeared that anything could happen. And the results? Well, that's a bit more difficult to gauge. A combined total of 16.7% of the vote for the three parties and independents looks like a victory for multipartisanship. And in fact it is - especially when the non-Liberal/PC numbers for the last election were 5.3%. Yet that 16.7% translated to no seats whatsoever - and New Brunswick is left again with a bipartisan Assembly, with only the PCs forming the government and only the Liberals in opposition.

I've become interested in the Gallagher Index, a tool to measure disproportionality. It gives a particular number, based on some fancy math resulting from the difference between percent of popular vote and percent of seats in the Legislative Assembly for each party.

I calculated it last night based on the numbers then posted on the CBC's website. It seems they have shifted a little bit, but at the moment I don't feel like recalculating the numbers. In any case:
  • The PCs got 48.92% of the vote and 76.36% of the seats, a difference of 27.44 points.
  • The Liberals got 34.45% of the vote and 23.64% of the seats, a difference of -10.81 points.
  • The NDP got 10.29% of the vote and no seats, a difference of -10.29 points.
  • The Greens got 4.54% of the vote and no seats, a difference of -4.54 points.
  • The People's Alliance got 1.18% of the vote and no seats, a difference of -1.18 points.
  • Independents got 0.62% of the vote and no seats, a difference of -0.62 points.
A little bit of number crunching gets, based on those mostly-accurate CBC numbers, a Gallagher Index of 22.34, which is remarkably high. Over the coming days I want to look at the Gallagher Index a bit more and compare results across provinces and across eras. At the moment, though, I'll finish up by calling attention to the fact that while the PCs did get a grossly overinflated percentage of seats, by and large it's tough to see the legitimacy being questioned: 48.92% (or 48.87% at the moment) is more than 14% above the Liberals and a hair's-breadth from being a simple majority. The lack of minor-party representation is problematic, and the Liberals' underperformance in number of seats cause for concern, but David Alward's mandate is not. He won the election, fair and square, and the parliamentary makeup shows that.

Wednesday, September 22, 2010

Québec: Who Supports the NDP in Québec?

So back to Québec. Leger Marketing / Le Devoir have a new poll out. Le Devoir's polls are cool because they're so comprehensive. They focus in great detail on a single province, and they poll federal and provincial intentions at the same time. They split the province into three, Greater Montréal, Greater Québec City and the rest of the province, and they divide results into Francophone and 'non-Francophone' (a three-way Franco, Anglo, Allo would be nice but the sample size would presumably be too small to be useful anyway). I'd love to see similarly detailed polling results for the rest of the country.

One thing that's interesting about the poll is the extent to which the appearance of near-hegemony federally by the BQ is not at all reflected provincially. Québec would stand to benefit from a more proportional method of representation at least as much as any other part of the country would. While they poll 36% province-wide, fourteen percentage points above their nearest rival, they have made little inroads mong non-Francos, who support them at 6% compared to 46% for the Liberals. They trail the Conservatives in and around Québec City, where Harper's party is polling a remarkable 37%. And yet the party polling first in one of Québec's two metropolises is polling fourth in the other. Averaging them out, the Conservatives are, bizarrely, at second province-wide and are currently the highest-polling federalist party. Incidentally, replace "BQ" with "PQ", "Conservatives" with "ADQ", and "Liberals" with, er, "Liberals", and you have rough provincial numbers too.

So Québec's all over the shop. Parties' support depends on region or on language. Yes, but with one very interesting exception. Let's look at the NDP's support in Québec:
  • Province-wide: 17% (of decideds)
  • Among Francophones: 17%
  • Among Anglophones and Allophones: 14%
  • In Greater Montréal: 18%
  • In Greater Québec City: 15%
  • In the rest of the province: 15%
Remarkable, isn't it? Looking at each of those six stats gives the BQ a range from 6% to 43%, the Liberals from 17% to 46% and the Conservatives from 16% to 37%. And yet province-wide in all of the above stats, the NDP's lowest is 14% and highest is 18%, a difference of four percentage points that itself undoubtedly falls within the margin of error.

It's an interesting statistic, one that further confounds the question of who exactly NDP supporters in Québec are. The answer appears to be 'everyone, in almost equal measure'.

By some standards, this is a bad poll for the NDP. Undoubtedly because of the long-gun registry issue, the NDP drop nationwide is certainly mirrored in Québec. The NDP remains fourth province-wide, which is no cause for celebration. When you're only polling in the high teens, even distribution of support is a terrible thing. It's tough, based on these polls, to see how the NDP can have any breakthroughs at all in Québec.

But the silver lining for them is this: it proves that the NDP are not merely a Montréal flirtation. It shows that their rather Anglo image seems to be dissipating within Québec, where their support among Francos is now higher than their support among Anglos: I saw a pro-Bill 101 poster recently and looked at the list of official sponsors and was pleased to note the "NPD" logo alone amongst federalist parties there, alongside the BQ, the PQ, QS and the ADQ. Their logo included the only maple leaf on the whole page. "Rest of Québec" is too monolithically large to be sure, but they seem to cross the urban/rural divide in Québec, while in the rest of Canada that very question is pulling at the heart of the party.

Most importantly, it means Québecois NDP seats, if they come at all, will not come one at a time in various parts of the province. If the NDP can see their fortunes rise in Québec, they might start to see seats being filled from all over the province. It's not a realistic idea at all at the moment, but one day? Well, stranger things have happened.

Friday, September 17, 2010

Canada: Polling houses and provinces

So this week, Ipsos Reid, Environics and Ekos have all released polls. They don't have the same sample sizes and they don't cover exactly the same dates. The three houses have slightly different methodologies and tend to produce differing results. Yet despite all of this, they have pretty close numbers at the moment nationwide. For example, Ipsos Reid puts the Conservatives at 34% nationwide, Environics has them at 35% and Ekos, which typically gives the lowest Tory numbers, has them at 32.4%. All well within the margin of error. Ipsos Reid and Environics have the exact same numbers for the Liberals, Ekos has them lower by little more than 2%. All in all, pretty good numbers. So good, in fact, that we can conclude that (a) the three houses must be doing reasonably good jobs if their numbers are so similar, and (b) these numbers must be pretty accurate, if three different polling houses have produced them, more or less.

National Voting Intentions, September 2010

The national sample size is going, of course, to be some ten times larger than the sample size of each province. Of course as a result the margin of error is going to be greater at the provincial level. But what strikes you looking at the three houses is just how deviant their provincial numbers are from each other, in spite of their similarities at the national level. It's an interesting paradox that unreliable numbers at the provincial level can combine to produce more reliable numbers at the national level, but it's a fact of statistics that deviation gets 'rounded out' as numbers increase. Still, it's tough to make any real statements at all about what's going on provincially based on these three polls.

For example, let's look at the four-province megaregion pollsters call 'the Atlantic'. The four provinces east of Québec all obviously have different voting trends and traditions, but due to their relatively smaller population bases are still lumped together in polls. Look at each of the three pollsters, though, and you get a remarkably different picture of these provinces.

Atlantic Voting Intentions, September 2010

Ipsos Reid has the Liberals at 48%, domination on a level we rarely see anymore. Ipsos Reid will tell you that just slightly less than half of all Atlantic Canadians are planning to vote Liberal. Ekos, on the other hand, has the Liberals slightly behind the Conservatives - 32.5% to 33.2%, with the NDP doing historically well at 22.4% in a legitimately tripartisan race: numbers that might be plausible in Nova Scotia, but as for the other three provinces? Environics, on the other hand, not only agrees with Ekos by putting the Conservatives first but puts them far ahead, at 35% to the Liberals' 31%. Has Danny Williams's ABC campaign lost its fizzle? Or are these numbers just all over the place?

BC Voting Intentions, September 2010

Going from one coast to the other, we have BC. Now, yes, BC is notoriously volatile. And yes, BC is legitimately three-way (or even four-way) to an extent unheard-of in any other anglophone province. But... Ekos puts the NDP ahead at 30.8% and has the Conservatives and the Liberals neck-and-neck at 27.0% and 26.5% respectively (and Green at 15.0%). Those numbers might be plausible if surprising, but contrast them with the other two: Ipsos has the NDP at a sad third with 21% behind a surging CPC at 41% and the Liberals at 25% (Green at 13%). Environics also has the NDP at third at 23%, well behind the Liberals at 30% and the Conservatives at 35% (Green at 10%). Admire that range: are the NDP polling 31% or 21%? Are the Tories polling a sad 27% or a mighty 41%?

The national numbers produced by these polling houses provide great talking points. But come election day national numbers mean nothing (remember 1993, where 16% of the national vote got the PCs two seats but 14% of the vote got the BQ 54 seats). And just how much can we trust these polling companies to give us useful provincial (in fact regional) data when it's so very hit-and-miss?