Note to my English readers: since I firmly believe in a bilingual Canada and since this post is (mostly) about Quebec, I decided to write it in french. You can use Google translate if you really have to, or better, try to read it anyway. Posts in french will remain the exception.

J'avais écrit ce billet il y a quelques semaines pour étudier un scénario hypothétique dans lequel le Bloc n'existerait plus. C'était naturellement avant la soudaine montée du NPD qui rend ce billet moins fictionnel. Au fait, si je mettais à jour le scénario du Bloc disparu pour tenir compte du changement dans les intentions de votes au Québec (ainsi que dans les second choix), cela ne donnerait plus une course serrée à trois, mais le NPD à 50 sièges. Cela démontre bien que ce genre de scénarios hypothétiques est largement dépendant des intentions de votes courantes.

Aujourd'hui, j'essaie un autre scénario: et si le Parti Libéral n'existait plus? Qui serait le principal gagnant?

En utilisant les dernières projections ainsi que les seconds choix du sondage Ekos (ajustées pour tenir compte du fait qu'au Québec les électeurs ont un choix supplémentaire avec le Bloc), voici à quoi ressemblerait le paysage politique Canadien si le PLC n'existait plus. Un scénario fiction bien sûr, mais peut-être indicateur de la possible horrible performance qui attend ce parti la semaine prochaine (si les sondages plaçant le NPD 2e au pays sont vrais).

sièges	PCC	PLC	NPD	Verts	Bloc
Avec PLC	146	73	49	0	40
Pas de PLC	164	0	129	0	15

Remarquez que les projections (avec PLC) donnaient 9 sièges au NPD au Québec. Une fois disparu, la plupart des électeurs Libéraux se tourneraient ainsi vers le NPD (dans une proportion de 47%). Vu que ce parti bénéficie déjà d'une poussée incroyable au Québec, le résultat serait de pratiquement éliminer le Bloc du paysage politique. On parle souvent de la division du vote "progressiste" entre le PLC, NPD, Verts et le Bloc, ce qui avantage bien sûr le PCC (regardez mon post sur le mode de scrutin Alternatif et vous verrez que le PCC gagne environ 10-15 sièges grâce à cela), mais on n'évoque pas si souvent la division du vote fédéraliste. Bien sûr, la chute du Bloc ne serait pas aussi importante si le NPD ne pointait pas à 28% au Québec (si le Bloc était à 35% et le NPD à 18%, comme il y a 3 semaines, le Bloc ne perdrait aucun siège avec la disparition du PLC). Il n'empêche qu'il est ironique de voir que le Bloc survit grâce à la présence du parti de Pierre Elliott Trudeau!

L'autre grosse différence est bien sûr que les Conservateurs seraient majoritaires, même s'ils ne bénéficieraient que de relativement peu de gains (18) à comparer au NPD (70). Les détails comté-par-comté sont ici. Il est intéressant de savoir que dans ce scénario fictif, le PCC récolterait moins de votes (40% contre 42% au NPD) mais gagnerait davantage de sièges.

I'm a little bit busy today preparing the class I'll have to teach starting May 8th, but I just wanted to let you know of two posts coming tomorrow.

The first one is the simulation "What if the Liberals Party didn't exist?", a scenario not that fictional looking at the possible epic collapse this party is heading for... You'll be surprised by some of the results, especially in Quebec. Post will likely be in french for this reason.

The second one is to investigate whether we should start considering a NDP-led coalition. After all, the latest Ekos numbers are crazy.

In the mean time, if you can't wait to see what the Ekos numbers would mean in terms of seats, let me remind you that I provide an online simulator here. Just enter the percentages and you'll see the seats (and you also have riding-by-ridings projections under the second tab). So have fun and let me know if you used the simulator for a political pool.

In the UK, the Lib-Dem agreed to enter a formal coalition with the Conservatives. One of the conditions was to hold a referendum on a reform of the electoral system. Namely, the switch from the current first-past-the-post (that we also have in Canada) to the alternative vote, also known as instant run-off. In this system, if no candidate gets more than 50% of the votes, we eliminate all but the two candidates and redistribute the votes according to the second choices (of course, it means that your ballot is slightly different. Indeed, you can rank your candidates). Since we have the second choices from every Ekos poll, we can run a simulation.

I already ran such a simulation of how things would be if Canada had this system here. However, with the recent NDP's surge, I think it is worth revisiting. On top of that, as with first choices, second choises can change throughout the campaign. Therefore an update is maybe necessary. On top of that, I'm guessing some new readers of this blog didn't get the chance to see the first post.

So let's talk about the second choices. I'm using the latest Ekos poll, but I then have to adjust and make some calculations for the second choices for the Bloc. Indeed, Ekos only provides us with the second choices nationally. It means for instance that while only 2.7% of Liberals voters identified the Bloc as their second option, we know these 2.7% all come from Quebec. Here are these second choices for Quebec and the ROC (Rest of Canada - term usually used in Canada for every province except Quebec).

Second	Quebec	First Choice
		CPC	LPC	NDP	Green	Bloc
	CPC	-	10.3%	15.4%	10.5%	10.2%
	LPC	15.9%	-	27.1%	19.6%	13.9%
	NDP	21.3%	46.9%	-	22.9%	40.1%
	Green	8.3%	13.1%	16.8%	-	16.0%
	Bloc	10.7%	14.5%	24.4%	23.5%	-
	other	1.7%	1.2%	0.1%	2.8%	1.0%
	none	42.0%	14.0%	16.1%	20.7%	18.8%
Second	ROC	First choice
		CPC	LPC	NDP	Green	Bloc
	CPC	-	12.0%	20.4%	13.8%	-
	LPC	17.8%	-	35.9%	25.6%	-
	NDP	23.9%	54.9%	-	29.9%	-
	Green	9.3%	15.3%	22.2%	-	-
	Bloc	0.0%	0.0%	0.0%	0.0%	-
	other	1.9%	1.4%	0.1%	3.7%	-
	none	47.1%	16.3%	21.4%	27.1%	-

One think that is quite amazing is that the NDP is actually the main second choice of EVERY other party (excluded the "none" category of course and the Green in Quebec)! That answers the questions we had in some comments on this blog last week: yes people can "jump" directly from the Tories to the NDP, without passing by the Liberals case.

What this means is that the NDP has a huge pool of potential voters. We always hear the a lot of NDP voters switch to the Liberals at the end of the campaign (which is probably partially true), but we have here evidence of the other side of this effect. Therefore, with the AV electoral system, the NDP would naturaly be favoured, as this system is all about getting second choices.

So let's compare the seats projections.

seats	CPC	LPC	NDP	Green	Bloc
current	146	73	49	0	40
AV	130	78	67	0	33
share	42%	25%	22%	0%	11%

A couple of things to take from this table. First of all, notice how the share of seats is very close to the share of votes (the % are almost identical to the ones used for the "latest" projections on the right column of this blog). Yes the Tories get a little more than their "fair" share, so does the Bloc thanks to regional concentration of its votes, but overall, the AV system would result to a very proportional system (except for the Green). By the way, the redistribution would be necessary in 193 ridings.

Secondly, you can see the NDP would clearly be the winner of this reform. If you want to see where these seats are coming from, here is the pdf. Please note I haven't scaled the percentages to sum up to 100% (to take into account that some voters have no second choices) as it doesn't change the winner of the riding. Of the 18 new seats for the NDP, 12 are in Quebec.

The Bloc to some extent and the Conservatives mostly would be the losers of such a reform (they would really be under the threat of "coalition" NDP-LPC). Of course, this simulation assumes that people wouldn't vote differently under the new system, which is quite a strong assumption.

So should the NDP push for such a reform? I personnally would prefer the introduction of a german-like system, with the compensation operating at the province-level (for those of you who know the system). But we have to admit that the AV system woud probably be good for the party of Jack Layton. Although, a true proportional system for this party would insure that gains would be less dependent on Quebec only.