Tuesday, June 30, 2009

New Léger Marketing Poll: Bloc and Liberals Tied

Léger Marketing has released a new poll taken between June 25 and June 27 and involving 1,021 interviews - all from Quebec. Here are the results:

Bloc Quebecois - 35%
Liberals - 35%
New Democrats - 15%
Conservatives - 11%
Greens - 4%

This poll puts into question the results of both Nanos and EKOS over the last week, who showed a Tory bump in the province. This poll puts the Conservatives back into fourth place behind the NDP, who gets a very strong result. The NDP are in a bit of a nebulous position, with polls putting them either capable of winning two seats or likely to win none. The Conservatives are also difficult to track, and look either to go back to 2004 numbers or maintain at least a few seats in the province.

The Liberals should be happy with this number, which I believe is likely their high-water mark in any election. The Bloc shouldn't be too worried with 35%, but I would consider that the line below which they should start to worry.

The francophone vote, which determines the vast majority of seats in the province, is as follows:

Bloc Quebecois - 42%
Liberals - 30%
New Democrats - 15%
Conservatives - 11%
Greens - 3%

This sort of result is why the Bloc shouldn't be in panic-mode. They are still comfortably ahead in most of the ridings they currently hold. That the Conservatives aren't doing better in this demographic puts their Quebec City ridings into question, but the NDP should be happy that they're polling as well with francophones as with the province as a whole, as their winnable ridings are quite bilingual.

The poll also asked people which political leader had their confidence. The result:

Gilles Duceppe - 30%
Michael Ignatieff - 21%
Jack Layton - 20%
Stephen Harper - 11%

This puts Duceppe and Ignatieff swinging below their weight, but puts Layton in a good place for growth. Harper seems to be unable to reach beyond his party faithful.

The poll also looked into provincial politics, which will be the subject of an upcoming CentVingtCinq post.

Monday, June 29, 2009

Pollster Leanings - Update

I've updated the pollster leanings chart with CROP. They only conduct polls in Quebec, and it will come as no surprise that they are the least favourable pollster when determining Bloc support. CROP tends to poll the Bloc 3.7 points lower than the average. They are also unfavourable to the Greens (1.8 points lower), but favour the NDP (3.4 points higher), the Conservatives (1.3 points higher), and the Liberals (1.2 points higher).

This is the last pollster to be added to the chart. From now on, I will be updating the chart with more recent calculations as time permits.

This chart tracks how each pollster tends to lean when calculating support levels for the various parties, as compared to the average polling results from other pollsters each month. This does not necessarily equate to a deliberate bias, but instead is more reflective of the polling methods used.

The following chart shows each pollster's average variation from other polling firms. The numbers are the amount of percentage points a particular pollster favours or disfavours that particular party compared to other pollsters over a similar period of time.

Saturday, June 27, 2009

New Poll from Nanos: 4.1% Liberal Lead

Nanos released a new poll today, taken between June 17 and June 21 and involving 1,004 interviews.

As mentioned earlier, projection updates will only be made on Thursday, but in the meantime here are the national results:

Liberals - 36.3%
Conservatives - 32.2%
New Democrats - 16.8%
Bloc Quebecois - 9.8%
Greens - 4.8%

This counter-balances the EKOS poll from Thursday which put the Tories in the lead. Nevertheless, this Nanos poll does show a drop in Liberal support from 37.2% in May, and a gain of Conservative support from 31.8%. The NDP and Bloc also made gains.

Atlantic Canada continues to show massive improvement for the NDP, who poll 31.8% (up from 17.5% last month). They've supplanted the Tories as the second party in the region, which is definitely somewhere Jack Layton wants to be.

In Quebec, the Bloc also shows improvement, back up to the 38% that won them 49 seats in 2008. That is up three points from the previous Nanos poll, which corresponds with the three point drop of the Liberals, who are now at 35.4%. That is still a very strong number for them. This poll also echoes the EKOS poll in that it shows a bump for the Tories, rising from 11.6% to 14.0%. The NDP is back in fourth at 10.8%.

The biggest "story" from this poll is the flip-flop in Ontario, where the Tories have gained almost nine points and now stand on top with 42.4%. The Liberals slipped only slightly from 42.1% to 40.9%, but this has put the NDP in a very difficult position: 11.5%. They need to be at least at 15% if they want to win anything more than a couple seats.

Nanos has seemed to removed the "West" category and has replaced it with "Prairies" and British Columbia. I'm seeking clarification as to whether the Prairies includes Alberta or not. But in British Columbia, Nanos gives the Liberals the lead at 36.4%, a significant 6-point lead over the Tories who are at 30.4%. The NDP remains in third at 24.0%.

UPDATE: Thanks to a quick response from Nik Nanos, I can tell you that the "Prairies" in their poll refers to Manitoba, Saskatchewan, and Alberta. Unfortunately, that means it cannot be used in my projection model, but the new category for British Columbia is certainly helpful.

They also asked whether people have a positive or negative impression of the two main party leaders, and after the calculations are done it comes out that Stephen Harper has a -14.8 net impression score to Michael Ignatieff's +5.2. Harper's best score was in the Prairies at +9.1, but he had had negative scores everywhere else, including a -33.3 in Quebec. Ignatieff only had a negative score in the Prairies (-11.1) and had his best score in Atlantic Canada (+18.0).

This poll would result in the following seat totals:

Liberals - 123
Conservatives - 116
Bloc Quebecois - 48
New Democrats - 21

The close result in Ontario is counterbalanced by the strong Liberal performances in British Columbia and Atlantic Canada, giving them a small minority government.

A good poll for the Liberals, but Nanos has tended to poll better for them in the past. I'm sure several other polls will come out this week, plus polls will be "aged" on July 1, so the July 2 update could have some interesting results.

Friday, June 26, 2009

Pollster Leanings - Update

I've updated the pollster leanings chart with Strategic Counsel. I've also removed the Ontario part of this chart, as it was usually similar to the national leanings. I've kept Quebec because of the presence of the Bloc Quebecois and how that changes things.

Strategic Counsel is the most favourable pollster to the New Democrats and the Greens, nationally, and to the Greens in Quebec.

This chart tracks how each pollster tends to lean when calculating support levels for the various parties, as compared to the average polling results from other pollsters each month. This does not necessarily equate to a deliberate bias, but instead is more reflective of the polling methods used.

As you can see, the list is not complete yet.

The following chart shows each pollster's average variation from other polling firms. The numbers are the amount of percentage points a particular pollster favours or disfavours that particular party compared to other pollsters over a similar period of time.

Thursday, June 25, 2009

Projection Update

With an election impossible until the fall, and with EKOS seemingly having a regular arrangement with the CBC for weekly polls every Thursday, I will be updating the projection only once per week, on Thursdays. Throughout the week I will still report on new polls and post other interesting tidbits, but projection updates will only be done on a weekly basis.

Two polls have been added to the projection. The first is by Angus-Reid, taken between June 17 and June 18 and involving 1,005 interviews. The national result:

Conservatives - 32%
Liberals - 31%
New Democrats - 18%
Bloc Quebecois - 11%
Greens - 7%

This poll translates into 121 Conservative MPs, 101 Liberals, 53 Bloc Quebecois MPs, and 33 NDP seats. One of the questions asked in this poll was who would be best to handle the economy. Stephen Harper was first with 34%, followed by Michael Ignatieff at 20% and Jack Layton at 10%. The Liberals had a very strong result in the Prairies in this poll (39%, just below the Tories at 43%). The Conservatives took a small lead in Ontario (37% to 35%), and the NDP ranked first in Atlantic Canada with 34%, just ahead of the Conservatives at 33%. The Bloc had a strong result in Quebec with 42%.

The second poll is from EKOS Research, taken between June 17 and June 23 and involving 3,505 interviews. The national result:

Conservatives - 34.8%
Liberals - 32.6%
New Democrats - 14.3%
Greens - 9.3%
Bloc Quebecois - 9.0%

This poll translates into seat totals of 126 Conservatives, 112 Liberals, 47 Bloc Quebecois, 22 New Democrats, and one Green.

This marks two polls putting the Tories slightly ahead of the Liberals - a significant change from the last few months. In the EKOS breakdown, the Conservatives led in most demographics - males, females, 25-44 year olds, 45-64 year olds, those over 65, those with a high school education or less, those with a college education, and in the cities of Vancouver, Calgary, and Ottawa. The Liberals led among those younger than 25, those with a university education, and in the city of Toronto. The Bloc led in Montreal.

The Liberals have not managed to recover from their slide at the end of the last EKOS poll. This shows that Ignatieff was wise to avoid an election, but it also might show that he was unwise to threaten one in the first place. In any case, the volatility of the voting public is clear, and it would seem to be necessary to go to an election with solid justification rather than a simple desire to replace the government. And while the NDP had an excellent result with Angus-Reid, this EKOS poll is very dangerous for them - 14% in Ontario, less than 8% in Quebec, and less than 20% in British Columbia.

The short-term projection has changed significantly. The Conservatives have gained ten seats and the NDP three, while the Liberals have dropped 13. The Conservatives have gained 1.2 points nationally, while the Bloc has gained 0.4 and the NDP 0.2. The Liberals have lost 1.0 points and the Greens 0.6.

The long-term projection has changed as well. The Conservatives have gained 0.2 points and the New Democrats and Bloc Quebecois have gained 0.1 points each. The Liberals have lost 0.2 points and the Greens 0.1. The NDP has gained one seat to reach 22 while the Liberals have lost one, and stand at 120. The Tories lost one seat in the Prairies but gained another in Ontario. The Liberals lost a seat in Ontario and Atlantic Canada, but gained one in the Prairies. Finally, the NDP made their seat gain in Atlantic Canada, where recent polling numbers are showing great gains for them.

Regionally, the Liberals have gained 0.3 points in British Columbia and 0.8 points in the Prairies. They've lost 0.4 points in Ontario and 0.7 in Atlantic Canada. The Conservatives have gained 0.4 points in Ontario but have lost 0.4 in the Prairies. The NDP has gained 0.7 points in Atlantic Canada and lost 0.4 points in British Columbia. Other movements are less than 0.3 points.

So! Things remain incredibly close, but for the first time in a very long time the worst news in recent polling is for the Liberals. They have work to do to get themselves back in the lead.

Tuesday, June 23, 2009

Party Contributions and International Affiliations

Though political parties now rely more on government funding than ever before, political contributions by Canadians are still an important part of the finances of any Canadian political party. In addition to providing parties with funding, they also allow us to gauge the enthusiasm the population, and particularly those with party memberships, have with each of the parties.

Elections Canada has all of the financial returns available for public consumption, but from what I can tell only those including the 2007 fiscal year are currently available (except for the 2008 return of the Animal Alliance, those keeners).

Government funding is a relatively recent factor, and it is demonstrated in the returns of some of the parties. The Bloc Quebecois and New Democrats appear to be giving less emphasis on raising funds the old way, as they have seen a drop in political contributions which does not correspond to their electoral performances.

Here are the political contributions, starting with 2007 and running down to 2003.

Conservative Party

The 2003 total for the Conservatives is the combination of the Canadian Alliance and Progressive Conservative totals. It is interesting to see how things took off with the formation of the Conservative Party:

2007 - $16,983,000
2006 - $17,392,894
2005 - $17,915,780
2004 - $12,907,357
2003 - $8,458,472

Liberal Party

The Liberals saw big drop-offs in contributions in 2004 and 2007. The drop-off is probably a result of the previous years (2003 and 2006) seeing leadership races.

2007 - $4,737,220
2006 - $11,261,293
2005 - $9,121,716
2004 - $6,085121
2003 - $14,618,039

New Democratic Party

The NDP has seen a steady decline since 2003 in funding, which I have to chalk up to a lack of interest on the part of the NDP itself. During this period, the NDP was on the rise, so it doesn't compute that there would be a lack of interest on the part of contributors.

2007 - $3,912,029
2006 - $3,954,501
2005 - $5,073,310
2004 - $5,187,142
2003 - $6,176,423

Bloc Québécois

The Bloc has also seen a big drop in funding, which again needs to be chalked up to increased reliance on government funding.

2007 - $593,036
2006 - $689,682
2005 - $965,089
2004 - $1,051,851
2003 - $1,244,612

Green Party

The Greens, however, have increased their contributions by more than three times since 2003.

2007 - $972,022
2006 - $962,927
2005 - $409,357
2004 - $351,031
2003 - $296,773

International Affiliations

Onto another topic, most Canadians probably don't realise that their political parties are affiliated with parties in other countries through international organisations. The affiliation is, of course, rather limited but parties within these organisations get together and discuss issues. For example, Michael Ignatieff will be addressing the Liberal International this summer.

The Conservative Party is a member of the International Democrat Union. Fellow members include the Liberal Party of Australia (of John Howard fame), the Kuomintang of China (formerly led by Chiang Kai-Shek and currently in government in Taiwan), the Union pour un Mouvement Populaire (Nicolas Sarkozy's party in France), the Christian Democratic Union (Angela Merkel's party in Germany), the New Zealand National Party (in government), the Democratic Party of Serbia, the Moderate Party of Sweden, the Conservative Party of the United Kingdom and, of course, the Republican Party of the United States.

Countries led by members of the IDU include Austria, Canada, Republic of China, Croatia, Denmark, Estonia, Finland, France, Germany, Greece, South Korea, New Zealand, and Sweden.

The Liberal Party is a member of the Liberal International. Fellow members include the Mouvement Réformateur and Vlaamse Liberalen en Democraten parties in Belgium, the Yabloko party in Russia, the Liberal People's Party and Centre Party of Sweden, and the Liberal Democrats of the UK. Countries with an LI member forming the whole or part of government include Belgium, Bulgaria, Croatia, Denmark, Estonia, Finland, Honduras, Paraguay, Senegal, Slovenia, Sweden, and Switzerland.

The New Democrats are a member of the Socialist International. Member parties include the Australian Labor Party (currently in government), the National Democratic Party of Egypt, the Socialist Party of France (of Ségolène Royale fame), the Patriotic Union of Kurdistan, the Social Democratic Party of Germany, the Labour Party of Ireland, and the Labour Party of the UK.

Countries with an SI member forming the whole or part of government include Andorra, Angola, Aruba, Australia, Austria, Belgium, Bosnia and Herzegovina, Brazil, Bulgaria, Cape Verde, Chile, Cote d'Ivoire, Costa Rica, Cyprus, Ecuador, Egypt, Estonia, Germany, Ghana, Guatemala, Haiti, Hungary, Iceland, Iraq, Israel, Lebanon, Luxembourg, Mali, Mauritius, Mongolia, Montenegro, Morocco, Mozambique, Namibia, Nepal, Netherlands, Nicaragua, Norway, Pakistan, Panama, Paraguay, Peru, Portugal, Romania, San Marino, Serbia, Slovakia, Slovenia, Spain, South Africa, Switzerland, Tunisia, United Kingdom, Uruguay, and Zimbabwe.

The Green Party is a member of the Global Greens and the Federation of the Green Parties of the Americas. Fellow members of the GG include the Green Party of the United States, the Australian Greens, Les Verts of France, the Green Party of England and Wales, and the Scottish Green Party. The FFGPA seems to have mostly the same members, but limited to North and South America.

The NDP seem to have the best international affiliation what with all the governments formed by members of the Socialist International, but it is difficult to put yourself up against the IDU when its members form the government of three of the G8 nations (France, Germany, and Canada). The SI isn't completely left in the lurch, as it does have the UK and a party that forms part of the governing coalition of Germany.

Monday, June 22, 2009

Pollster Leanings - New Chart

I wanted to clean this up a little, so here is a chart that tracks how each pollster tends to lean when calculating support levels for the various parties, as compared to the average polling results each month. This does not necessarily equate to a deliberate bias, but instead is more reflective of the polling methods used.

As you can see, the list is not complete yet.

The following chart shows each pollster's average variation from other polling firms. The numbers are the amount of percentage points a particular pollster favours or disfavours that particular party compared to other pollsters.

Proportional Representation in the 2008 Election

I thought it would be an interesting exercise to look at what the 2008 election would have been like if Canada had a proportional representation system. I chose a regional PR system, with each province voting MPs according to PR and giving them each the amount of seats they have presently. What do we get?
The Conservatives win a minority with 119 seats, rather than 143. The Liberals form the Official Opposition with 84 seats, rather than 77. The New Democrats form the second opposition party, rather than the third, with 56 seats. The Bloc Quebecois wins 29 seats and the Greens elect 17 MPs.

This would be a far more representative system of government. Rather than the Tories winning 72 of 95 seats in Western and Northern Canada, they would win 51. The Liberals would have 16 MPs from the West and North (rather than eight), the NDP would have 21 (instead of 15), and the Greens would have seven.

Ontario's representation would not change dramatically, but the Greens would elect eight Ontarian MPs. One of the major differences would be in Quebec, where the Bloc would elect only 29 MPs instead of 49. The Tories would elect 16 MPs rather than 10, the Liberals 18 rather than 14, and the NDP would have nine Quebec MPs instead of one. Even the Greens would elect three MPs from Quebec.

While this sort of system would all but guarantee minority governments, it would be far more representative and the parties would learn to work together. With this Parliament, a Liberal-NDP-Green coalition government could even be possible. Coalitions like that have worked in Europe and have worked well. Or, the Conservatives could have governed with the Liberals or on a case-by-case basis with any of the parties.

This sort of system would also encourage fringe parties, though there would probably be some mechanism requiring, say, 5% nationally before any MPs anywhere could be elected.

In any case, it is just an interesting little exercise. Based on the current projection, using this system would result in the following Parliament:

Liberals - 102 seats
Conservatives - 102 seats
New Democrats - 51 seats
Bloc Quebecois - 27 seats
Greens - 26 seats

That would be interesting to see. The Tories would have 45 seats in Western Canada, the Liberals 22, the NDP 19 and the Greens 9. Ontario would have 37 Conservatives, 43 Liberals, 16 New Democrats, and 10 Greens. Quebec would have 27 Bloc MPs, 24 Liberals, 11 Conservatives, 8 New Democrats, and 5 Greens. Atlantic Canada would have 13 Liberals, 9 Conservatives, 8 New Democrats, and two Greens.

At first it would cause some instability to have such weak minorities, but eventually I believe it would be more healthy for our democracy.

Sunday, June 21, 2009

Projection Update

Along with the Ipsos-Reid poll referred to yesterday, CROP has released a Quebec-only poll. It was taken between June 11 and June 18 and involved 1,003 interviews. The results:

Liberals - 35%
Bloc Quebecois - 31%
New Democrats - 17%
Conservatives - 13%

Terrific number for the NDP and a bad one for the Bloc. But we've seen how varied the polling results can be for these two parties in the province. This poll would have resulted in 42 Bloc Quebecois seats, 26 Liberal seats, four Conservative seats, and three New Democrats.

Moving on to the Ipsos-Reid poll yesterday, the seat results would have been:

Liberals - 122
Conservatives - 116
Bloc Quebecois - 50
New Democrats - 19

The Liberal number is down because of the close vote in Ontario, but with 40 seats in Quebec and Atlantic Canada, the Liberals would form government with this poll. The Tories manage to stay close with 46 seats in Ontario. The New Democrats are saved from complete irrelevance thanks to seven seats in Atlantic Canada.

The short-term projection has changed slightly, with the Conservatives and Bloc gaining 0.2 points nationally, while the Liberals, NDP, and Greens lose 0.2 points each. This translates into two seat gains by the Tories and the New Democrats, with four seat losses by the Liberals and one by the Bloc Quebecois.

The long-term projection hasn't moved in terms of seat totals, but nationally the Tories have gained 0.1 points and the New Democrats have lost 0.1 points. Regionally, the only significant movements were by the NDP and Bloc. The NDP lost 0.3 points in British Columbia but gained that amount in Quebec. The Bloc lost 0.3 points in that province.

We're seeing things gel nicely as we move into the summer. One thing is clear: the race is a tight one. The Tories and the Liberals are neck-and-neck, but the edge goes to the Liberals. Would a four-seat plurality be worth an election call by Ignatieff?

We've also seen the NDP start to bounce back from hitting a floor over the past few months. I might be going out on a limb, but I feel that we might see the NDP move back to 25 or so seats during the summer. I don't believe the Liberal and Conservative numbers will move very much, but a wild card could be in Quebec. Will the Bloc start to slide into the low-30s, as shown by the CROP poll? Or will they improve into the high-30s, as shown by the Ipsos-Reid poll? Things to watch.

CROP also had a provincial poll in Quebec, so check out CentVingtCinq later today for a projection update.

Saturday, June 20, 2009

New Poll: Ipsos-Reid

Ipsos-Reid released a new poll today, taken between June 16 and June 18 and involving 1,000 interviews. It has a margin of error of +/- 3.1%. This poll was taken in the midst of the election crisis and the Harper/Ignatieff negotiations. Here are the national results:

Liberals - 35%
Conservatives - 34%
New Democrats - 13%
Bloc Quebecois - 10%
Greens - 8%

This shows a modest Tory improvement, though we should note this result is very close to the last Ipsos-Reid poll released on June 4 which had the Liberals at 36%, the Tories at 33%, and the NDP at 12%. In any case, this poll demonstrates that an election would have been a risky venture for everyone - except the Bloc, which polled at 38% in Quebec, identical to the 2008 electoral result.

The poll was broken down into several categories, including urban/rural. The urban vote is 36% Liberal, 31% Conservative, and 14% New Democratic. I would have expected the NDP result to be higher. The rural vote is 44% Conservative, 29% Liberal, and 11% New Democrat.

Regionally, the Liberals and Conservatives are tied at 35% in British Columbia, and the Tories have a comfortable lead in Alberta and the Prairies. In Ontario, the result was very close, with 40% going to the Liberals and 38% going to the Conservatives. The NDP received a very small 13%. In Quebec, the Bloc remains on top with 38%, the Liberals second with 33%, and the Tories third with 16%. The Liberals lead in Atlantic Canada at 37%, but the New Democrats are a close second with 34%.

I will update the projection with these results tomorrow, as I do not have the time today.

Friday, June 19, 2009

At Issue Survey

Some of the other political blogs in Canada are giving their own thoughts on the end-of-session survey the At Issue panel responded to last night on The National. Here are my thoughts:

Most Underrated Politician: Gilles Duceppe. Both Stephen Harper and Jack Layton have seen their support levels drop significantly, while Michael Ignatieff is only returning the Liberals to their pre-Dion levels. Duceppe has been the leader of the Bloc Quebecois since 1997 and has been in the House of Commons since 1990, yet he remains popular in Quebec. People outside of Quebec don't hear much about Duceppe except a rare phrase or two, usually when there is some sort of issue that involves Quebec or could result in an election. Duceppe is more than just an "oppose, oppose, oppose" sort of politician.

Most Overrated Politician: Michael Ignatieff. While he isn't a bad politician or a bad party leader, he isn't the next Trudeau, as he was touted for awhile. He's managed to return respectability to the Liberal Party, but the Tories have had a bad first-half of the year and it is hard to say whether Ignatieff has earned that 6-10 point bounce or if Harper has given it to him. Ignatieff has managed to be a pretty good Opposition leader, but he still hasn't shown why he should be seen as a good replacement for Harper as Prime Minister.

Most Shamelessly Exploited Issue: Canadians' ignorance of the parliamentary system. The Conservatives did have some justification in strongly criticising the coalition in terms of whether it would have been a good government or whether Stéphane Dion would have been a good prime minister. But the party exploited the ignorance Canadians have of their own parliamentary system, describing the coalition as anti-democratic or an over-throw of the government, when in fact it was nothing of the sort. Canadians understand their Parliament less today than they did before the coalition talks began.

Most Under-Reported Event: The War in Afghanistan. I'm going to go along with Allan Gregg here and say that this war hasn't been given as much coverage as it deserves. It rarely makes the news other than when a Canadian soldier dies. As a corollary, I'll add the Iraq War. I feel completely uninformed concerning the current situation in Iraq. Granted, I don't follow American news closely, but you would imagine that this story would be important enough for Canadian news as well.

Best Political Play: Michael Ignatieff taking Liberal leadership. Without any real difficulty, without dividing the party (at least in public), Michael Ignatieff walked into the Liberal leadership. Think of how difficult it had been to choose a leader last time and how divisive the Chretien/Martin conflict had been in the party during the 1990s and early 2000s. And this wasn't a case of a very successful Minister of Finance taking the leadership after having proven himself, like Paul Martin. Ignatieff had been an MP for only a couple of years but was handed the leadership of the most historically important political party in the country.

Worst Political Play: Conservative Slide in Quebec. The attacks on the Bloc Quebecois during the coalition days and the successful attack by the Bloc on the Conservatives for their cultural cuts during the election have effectively poisoned the Tory well in Quebec. That Harper was able to elect 10 MPs in both 2006 and 2008 demonstrated that he had built a pan-Canadian party and that he could hope for majorities in the future. However, with the Conservatives' disastrous handling of the Quebec file since the election, they have lost almost half of their support in the province and have even dipped to NDP and Green numbers in some polls. Losing anywhere from five to eight seats in Quebec hurts the party at a time when the Liberals are making great strides in Ontario. Burning that bridge was not a good idea.

Next Election: November 2009. If the polling data continues throughout the summer, the Liberals will not pass up on the opportunity at the end of September/beginning of October. It's perfect campaigning weather. The New Democrats and Bloc Quebecois will probably not be willing to support the government, but we'll have to see how they act this summer. If the two parties make political hay out of the Liberal vote with the government today, they will be forced to vote against the government in September/October. If polling numbers take a turn for the worst for both the NDP and the Bloc, we may have to wait until the Spring of 2010 and the budget for the next election.

Work Force Voting Intentions

Yesterday's EKOS poll broke things down by age, so I thought it would be interesting to look at the voting intentions of working-age Canadians. These are Canadians between the ages of 25 and 64 - in other words, Canadians that are neither at school nor retired. These are the Canadians with most stake in a government, as they are the majority and they pay the taxes.

Voting intentions aren't strikingly different from the overall results, but there are some contrasts. The Liberals still lead but actually have a larger portion of this demographic than they do the country as a whole. The Conservatives also take a larger share of the vote, but only by 0.1 points. The New Democrats and Greens poll worse among this group than they do the country as a whole, while the Bloc Quebecois polls better.

To give the distinctiveness of the Quebec political scene its due, here are the results for that province:The Bloc polls slightly better among this demographic than they do the whole (about 1.3 points), as do the Liberals by 0.4 points and the Conservatives by 0.1 points. The NDP polls worse, about 1.1 points, as do the Greens, at about 0.7 points below the average.

What can we take from this? The oldest age group tends not to support the NDP or Greens, but the youngest group supports those parties in much greater proportions than older voters. So what we have are the mainstream parties - Liberals and Conservatives in Canada and the Liberals and Bloc Quebecois in Quebec - being supported by the "mainstream" of Canadians. This leaves the idealism of the New Democrats and Greens to the young and the pragmatism of the other three parties to the older age groups.

Thursday, June 18, 2009

New Poll: EKOS

EKOS has released a new poll today. It was taken between June 10 and June 16, involved 3,422 interviews, and has a margin of error of 1.7%. Here are the national results:

Liberals - 33.7%
Conservatives - 32.4%
New Democrats - 16.3%
Greens - 9.0%
Bloc Quebecois - 8.7%

This poll closed just before the outcome of the negotiations between Stephen Harper and Michael Ignatieff was announced. There was a significant Liberal drop on Monday and Tuesday to below 30%, perhaps as a result of the election pressure of Ignatieff. It is unfortunate the poll didn't continue until yesterday, as we could have seen what effect the results of the negotiations would have had on the electorate.

Le Devoir is reporting that the Liberal Party has sent out notices for its members to be prepared for an election starting the week of October 5th and ending sometime between November 9 and November 16. So, this summer will be the time to prepare and set the stage for a fall election. This makes the ebb and flow of polling data from now until then very important.

Harper received a 33% approval rating, with 47% disapproving of his performance. That is compared to a 38% to 54% split in April. His highest approval rating was in Alberta (50%), and his lowest was in Quebec (21%). Ignatieff had a 32% to 34% split, significantly worse than his 50% to 28% split in April. His highest rating was in Ontario (36%), and his lowest was in Alberta (21%).

Regionally, the Conservatives led in this poll in British Columbia, Alberta, and the Prairies. The Liberals led in Ontario and Atlantic Canada, while the Bloc Quebecois still leads in Quebec.

Breaking the poll down into demographics, the Conservatives have the lead among male voters (36.5% to the Liberal 33.7%), voters aged 45-64 (35.5% to 35.0%), voters aged 65 or older (40.1% to 37.7%), voters with a high school education or lower (30.7% to 28.2%), voters with a college or CEGEP diploma (37.8% to 28.8%), voters in Vancouver (46.8% to 27.4%), and voters in Calgary (59.7% to 20.9%).

The Liberals lead among female voters (33.7% to the Conservative 28.4%), voters under the age of 25 (23.8% to the NDP's 22.8%), voters aged 25-44 (33.5% to the Conservative 28.8%), voters with a university education (41.7% to 29.1%), voters in Toronto (44.8% to 33.9%), and voters in Ottawa (43.7% to 38.4%). The Bloc Quebecois leads in Montreal at 35.9%, just ahead of the Liberal 34.9% support.

There were some regional variations which favoured the NDP and the Greens. The NDP led among women and voters under the age of 25 in British Columbia, and held the youth vote in the Prairies as well. The Greens were second among those aged less than 25 in Ontario, and were first among those in Atlantic Canada. Elizabeth May needs the youth of the Maritimes to move to Central Nova.

You can go to the bottom of the page for the full regional details of the poll.

This poll would have resulted in an unstable Liberal minority of 127 seats, with the Conservatives forming the official opposition with 107 seats. The Bloc would have been third with 48 seats, the NDP fourth with 25 seats, and the Greens would have won one seat as well.

The short term projection has changed slightly, with the Conservatives losing two seats to the Liberals, who are now up to 135 seats and hold a comfortable minority. However, they nevertheless had the biggest support loss, losing 0.7 points. The Greens had the largest gain, at 0.3 points.

The long term projection has hardly changed at all. Nationally, only the NDP made a move with a 0.1 point gain, while there were no regional losses or gains of more than 0.2 points.

This poll says a lot to three parties. The Conservatives aren't out of it yet, and the volatility demonstrated by the Liberal plunge early this week means that anything could happen during an electoral campaign. This should make the Tories slightly more comfortable with the idea of a fall election. Even if they will definitely lose seats, they could still win a minority government and increase their lifespan for another couple years.

For the Liberals, this poll shows once again that they can form the next government, but it should also warn them about being too hasty. In any case, we'll have to wait and see what the end-of-week numbers were like to see if the Liberals have been hurt long term or if Ignatieff has pulled something good out of his hat.

For the NDP, this was a good poll. They have here some of their highest results of the year nationally and in British Columbia, Alberta, and Ontario, though they are still well below their 2008 electoral result. This translates to not a catastrophic loss of MPs, as this poll projects the NDP would be reduced from 37 to 25 seats. Nevertheless, Jack Layton should try to find a way to avoid defeating the government in October. But, of course, an electoral campaign can change anything and Layton had one of his strongest campaigns in 2008, so he might be banking on his ability to get people thinking about the New Democrats again come election time.

Wednesday, June 17, 2009

No Election This Summer

Liberal leader Michael Ignatieff and the Prime Minister have worked out an agreement that will avoid an election this summer. I'd have to say this was a smart move for everyone. A summer election would further push Canada into a state of democratic apathy.

The Conservatives aren't doing so badly in the polls that Ignatieff has made a mistake in not seeking an election, and they aren't doing so badly that he could legitimately call it an unpopular government that needs to be removed. The Liberals aren't so far ahead that a victory is assured, and they would have risked forming a very shaky minority themselves or giving the Tories another minority, but much reduced in both size and stability. The NDP would have lost half of their MPs, and the Bloc would have spent a lot of money for no real gains or losses.

I imagine we'll be talking about an election again in the fall or winter. It will be interesting to see how things progress in the polls over the summer.

UPDATE: From reports that are coming out, it seems part of the agreement includes an opposition day at the end of September, giving the Liberals a chance to put forward a non-confidence motion at that time if they don't like what transpired over the summer or what the Tories have to say in their economic update. That would put election day somewhere around the end of October or beginning of November - perfect campaigning weather.

Monday, June 15, 2009

What's at Stake

Although Le Devoir is reporting that Michael Ignatieff will support the government and not send this country to an election, in case those sources are wrong here is a look at what each party has to gain or to lose with an election on July 27.

UPDATE: Turns out Ignatieff has some demands which must be met, and it seems unlikely Harper would meet them. Either someone will blink and there won't be an election, or this time next week we're on the campaign trail.

UPDATE UPDATE: Harper responds and looks unlikely to make concessions. The pot begins to boil.

CONSERVATIVE PARTY

Situation: The Conservatives are down in the polls virtually everywhere. Nationally, they've lost anywhere from 7% to 20% of their 2008 electoral support. They're down almost 10 points in Ontario, have lost almost half of their support in Quebec, and some polls even put the party in second place in British Columbia.

Potential Gains: None. They appear to be treading water only in Atlantic Canada and Alberta, and its likely that every province will see a reduction in Tory support and seats.

Potential Losses: It is impossible to foresee a scenario like the last election, as Ignatieff and the Liberals are far more secure than they were under Stéphane Dion and Jack Layton doesn't look to be able to eat away any of the Liberal votes. In a best-case scenario, Stephen Harper could hold on to power but with a greatly reduced minority, and he would likely be in a position where the NDP and Liberals alone could out-vote him. In a worst-case scenario, the Tories would lose power and could even dip below 100 seats.

LIBERAL PARTY

Situation: The Liberals have opened up a beyond-statistical lead over the Conservatives nationally, increasing their support by anywhere from 26% to 42% of their 2008 electoral result. They are up in the polls in every region of the country, and most importantly have a significant lead over the Tories in Ontario and are within striking distance of the Bloc Quebecois in Quebec.

Potential Gains: A best-case scenario would make Ignatieff the country's next Prime Minister. With the polling results we've seen, the Liberals could increase their current caucus by 50 or 60 MPs, putting them in the 130-140 range and probably outside of the voting power of the Conservatives and NDP. More likely is a modest Liberal minority. A worst-case scenario would still see the Liberals increase their caucus by 30 seats or so and put them above the respectable 100-seat level.

Potential Losses: None, at least numbers-wise. The Liberals are virtually guaranteed to increase their support levels and seat totals. What they could lose, however, is the opportunity to form government. The numbers still make a Tory minority possible. That would result in a lost time for Ignatieff. The Tories would be safe for another few years because of election weariness, and the political momentum and capital of the current Liberal tide would be sapped.

BLOC QUEBECOIS

Situation: The Bloc is down slightly in the polls, but not significantly. The polls have shown everything from a 13% loss to a 5% increase over 2008's electoral result. More significant is that the Liberals have placed ahead of the Bloc in several polls.

Potential Gains: What is almost assured is that the Bloc will return to some of the regions of Quebec where the Tories have made gains. This means the Beauce, around Quebec City, and in the Saguenay. These are more traditionally Bloc regions, and it would be an important victory to return to these regions. In a best-case scenario, the Bloc could even increase its caucus in an election by anywhere from one to four seats. A worst-case scenario would still see a return to Quebec City and the Saguenay.

Potential Losses: The Liberals will make some major gains in the province, which means the Bloc could lose some of its seats in Montreal and south of the St. Lawrence. The Liberals could even take a foothold in Quebec City. The Bloc is assured to lose some seats to the Liberals, the only question is whether those can be made up by seat gains over the Tories. In a worst-case scenario the Bloc risks dipping to historic lows in popular support, and could even lose a tenth of its caucus or more. In a best-case scenario, the Bloc would still be pushed out of some of the regions in which it would like to remain: central Montreal, the Outaouais, and the Cantons-de-l'Est.

NEW DEMOCRATIC PARTY

Situation: The NDP has sunk in the polls, losing between 7% and 34% of its 2008 support. The NDP is stagnant in the West but has lost important support in Ontario, Quebec, and Atlantic Canada.

Potential Gains: None. While the NDP can always pull out an upset or two in individual ridings, it is extremely unlikely that the NDP would gain seats or even support in any region of the country.

Potential Losses: The NDP risks losing half of its caucus. The NDP currently has a very respectable 37 MPs, but they would be lucky to get out of an election with more than 20. The NDP is currently a pan-Canadian party, with seats in every region. But even in a best-case scenario, say 17% nationally, the NDP could lose its footholds in Alberta and Quebec. The latter loss would be especially significant. A worst-case scenario would see half of the NDP caucus disappear and Layton reduced to somewhere between 15 and 20 seats. That would be disastrous for a party that has made major strides in the last few elections. From this perspective, the NDP have the most to lose in an election, as though the Tories could lose government, it doesn't look like they can keep it for long anyway. With strong parties on the centre-right and centre-left, Canada is likely to see alternating governments for as long as those two parties remain whole. But for the NDP to return to virtual insignificance would be a tremendous blow.

GREEN PARTY

The Greens are polling relatively well, and would likely increase their national vote level to 8% or 9%. A seat victory is unlikely, but Elizabeth May could pull out a victory in Central Nova if Peter MacKay steps aside, as has been rumoured. Of course, with no real results from the 2008 election, the Greens could stand to lose some support. In any case, the Greens only make themselves relevant during an electoral campaign, so any election for them is a positive.

Friday, June 12, 2009

Projection Update: Harris-Decima

Now that the details of the Harris-Decima poll are available, I've updated the projection.

Firstly, this poll would have resulted in a stable Liberal minority:

Liberals - 140 seats
Conservatives - 94 seats
Bloc Quebecois - 52 seats
New Democrats - 22 seats

The Liberals made significant seat gains from other individual polls thanks to their lead in British Columbia (33% to 30%), their big lead in Ontario (42% to 31%), and their even bigger lead in Atlantic Canada (45% to 30%). The 40% result for the Bloc increased their seat total to 52 seats.

In the short-term projection, the Liberals have gained enough seats to now be considered capable of forming a stable minority government. They are now up four seats to 133, while the Conservatives are done three to 107. The Greens have lost their seat in Atlantic Canada, the NDP is down one, and the Bloc is up one.

In the long-term projection, the New Democrats have lost a seat to the Liberals in Atlantic Canada. That puts the Liberals up to 121 seats, four up on the Tories. There were no big regional gains, but the most significant was the 0.3 point gain by the Liberals in Atlantic Canada.

Now that it seems both Layton and Duceppe have committed themselves against the government's economic update, Michael Ignatieff has to decide whether to send Canadians to an election. Based on the numbers this week, he would be Canada's next Prime Minister. But we've seen how quickly things change, and how individual polls can sometimes show leads within the statistical margin of error. The "Canadians don't want an election" theme is always overstated, after the first week the voters seem to forget about who forced the election. A summer election would be bad for voter turnout, and I'd predict an all-time low in that domain. Of course, Ignatieff can wait until the fall or next spring, but the old cliché is that a short period of time is an eternity in politics. The situation in the fall or next spring might not be as favourable as it is now.

Things for Ignatieff to ponder. If you're reading this Michael, I'd hold off. You have until 2012 to bring down the government, and I'm sure another opportunity will present itself. It's a bit risky right now.

Thursday, June 11, 2009

New Polls: EKOS and Harris-Decima

Two polls were released yesterday and today. The first is by Harris-Decima, and was taken between May 28 and June 8 and involved 2,000 interviews. However, only limited details are available through a Canadian Press article, so I've only entered the information I do have into the projection. I will have to update it with the rest of the polling data when it is available.

The second poll is by EKOS and was taken between May 29 and June 9, involving 6,259 interviews. This is part of their deal with the CBC to release polls every Thursday, and it is a very large poll. These two overlap, and their results are very similar. Here are the national results, Harris-Decima first and EKOS second:

Liberals - 35% / 35%
Conservatives - 31% / 30.3%
New Democrats - 15% / 15.1%
Bloc Quebecois - 9% / 9.2%
Greens - 8% / 10.4%

Of note is that EKOS says the polling data turned hard against the Conservatives near the end of their polling period, coinciding with the Raitt affair. Also, looking at the polling trends, a clear gap is beginning to form between the Liberals and the Conservatives.

The EKOS poll, the details of which can be seen at the bottom of the page, would result in the following seat totals:

Liberals - 138
Conservatives - 101
Bloc Quebecois - 49
New Democrats - 20

The short-term projection has changed, with the Liberals gaining two seats, the Conservatives losing two seats, the NDP losing one seat, and the Bloc gaining one. This puts the Liberals on the brink of having a stable minority government.

The Liberals have gained 0.7 points in the national short-term vote projection, while the Greens have gained 0.2 points. The Conservatives and NDP have each lost 0.4 points.

The long term projection has changed significantly. The Liberals now have the seat total lead, the first time they've had it since I started making projections. They gained three seats to reach 120, while the Tories have lost four seats to drop to 117. The NDP has gained one seat as well, and stands at 22. The Conservatives lost their seats in British Columbia (one) to the NDP and in Ontario (three) to the Liberals.

The long-term national vote projection has also shifted, with the Tories losing half a point and the Liberals gaining 0.4 points. The Greens have also gained 0.2 points. The Liberals now have a significant 1.4 point lead over the Conservatives.

Regionally, the Conservatives lost 0.3 points in Atlantic Canada, 0.4 points in the Prairies, and 0.8 points in Ontario and British Columbia each. The Liberals gained 0.3 points in British Columbia and Atlantic Canada and 0.6 points in Ontario. The Greens also had a good shift, gaining 0.3 points in both Quebec and Ontario, 0.4 points in British Columbia, and 0.6 points in the Prairies.

These are two very bad polls for the Tories. The EKOS one would see the Conservative seat total in Ontario drop to 27. With all the bad press coming out against them recently, they should try to do all they can to avoid an election and hope the tide could turn during the summer. The Liberals, though their lead is narrow, are in the best position of any party. The NDP's result isn't horrible, and is in fact a little better than what we've seen lately, but Jack Layton would be taking a huge gamble if he supported a non-confidence motion, as he will surely see his caucus fall to under 30 seats. He could even lose as many as half of his MPs. The Bloc has nothing to gain in an election, except maybe a return to areas that the Conservatives have won in the last two elections. But they are likely to see their caucus drop by a handful of seats, their portion of the vote drop by a few points, their finances to drop by a few million dollars, and their political capital to take a big hit.

Considering that the Tories and the NDP have to avoid an election while the Bloc has nothing to gain by an election either, it is difficult to see the Liberals being able to corner the two other opposition parties badly enough that they would vote down the government.

Wednesday, June 10, 2009

New Poll: Strategic Counsel

Strategic Counsel released a new poll yesterday, for CTV. The poll was taken between June 3 and June 7, and involved 1,003 interviews.

Here are some of the stories that came out during this period:

JUNE 3
- Natural Resources Minister Lisa Raitt and the documents she left behind at CTV.
- TD Bank forecasts Canada to go $170 billion into debt over the next five years.
JUNE 4
- Minister Raitt's resignation is refused, and an aide resigns instead.
- Liberals and Conservative reject the Bloc proposal to impose Bill 101 on federally regulated institutions in Quebec.
JUNE 5
- Jobless rate in Canada will reach 8.4%.
- Michael Ignatieff makes an important speech in Quebec.
JUNE 6
- Conservatives move to remove the "faint hope" parole clause.
- Unemployment rate reachs 11-year high.

June 7th was a Sunday, so it was pretty quiet. Now that some context has been provided, here are the national results:

Liberals - 34%
Conservatives - 30%
New Democrats - 16%
Greens - 11%
Bloc Quebecois - 9%

Ontario:

Liberals - 44%
Conservatives - 35%
Greens - 11%
New Democrats - 10%

Quebec:

Bloc Quebecois - 37%
Liberals - 35%
Greens - 11%
Conservatives - 9%
New Democrats - 8%

This was a very good poll for the Liberals. They have a national lead beyond the margin of error, a significant lead in Ontario, and are within striking distance of the Bloc in Quebec. This poll would've given the Liberals 65 seats in Ontario, 11 more than they are currently projected to have. They would also win 24 in Quebec, three more than they are projected to have.

This is a bad poll for the Conservatives, as 30% is their 2004 result, as is the 9% in Quebec (translating to no seats). The 35% in Ontario is their 2006 result, which isn't catastrophic but would make holding onto power very difficult, as they would only win 35 seats.

The NDP had a good national result but very bad results in Ontario and Quebec, while the Greens had stellar results in the two major provinces. The Bloc can content itself with 37%, but anything lower than that is very bad news for them.

The short-term projection has changed, with the Liberals gaining six seats and the Conservatives losing six. The Bloc, NDP, and Greens remain unchanged. The Liberals pick up 0.2 points, but the Conservatives lose a full percentage point in the national vote. The Greens and NDP pick up about half a point each.

The long-term projection has not changed, with seats staying the way they are. But, the gap between the Tories and Liberals nationally has widened, with the Conservatives dropping 0.1 points to 32.7% while the Liberals remain steady at 33.3%. The Greens gain 0.1 points nationally, and gain 0.1 points each in Ontario and Quebec. The Liberals and NDP trade 0.1 points in Ontario to the benefit of the Grits, and both the Tories and NDP lose 0.1 points in Quebec.

The Liberals need to start breaking away in the regional polling for them to move into a position to form government. The Tory block of almost 70 seats west of Ontario is difficult to overcome, but if the Liberals continue to poll in the 40s in Ontario and high 30s in Quebec, this could be done. The dip in Atlantic support we've seen recently is costing the Liberals a few precious seats in what would be a very, very close election.

Tuesday, June 9, 2009

Regional Trends: British Columbia

After Ontario and Quebec, British Columbia is the most important battleground in the country. The province has 36 seats, almost 12% of all seats in the House of Commons. For the last three elections, it has been dominated by the Conservative Party.

In 2004, the Tories took 22 seats here with 36.3% of the vote. That dropped to 17 seats in 2006 despite a tiny increase in votes (37.3%), but then roared back to 22 seats with 44.4% of the vote in 2008. The remaining 14-19 seats are fought over by the Liberals and the New Democrats. In 2004, the Liberals took eight to the NDP five (and one independent), but since 2006 the New Democrats have established themselves as the second party in the province, winning 10 seats in 2006 and nine in 2008. The Liberals took nine in 2006 and five in 2008.

The Liberals maintained their votes in 2004 and 2006 (28.6% and 27.6%), but dropped significantly in 2008 down to 19.2%. The New Democrats have a solid base, and moved from 26.6% to 28.6% between 2004 and 2006, and have since dropped slightly to 26.1% in 2008. The Greens toiled in fourth with 6.3% and 5.3% in 2004 and 2006, but had a bit of a breakthrough in 2008, increasing their vote total to 9.4%.

The province can be split up into four regions: the BC Interior (generally the northern and eastern parts of the province), Fraser Valley and the Southern Lower Mainland (east of Vancouver and running along the American border), Vancouver and the Sunshine Coast (Vancouver and the Pacific coast), and Vancouver Island.

The northern and eastern parts of the province (the Interior and Fraser Valley) are dominated by the Conservatives. In 2008, the party took 16 of the 19 seats in these two regions. The New Democrats took two and the Liberals one. This has been a constant since the formation of the Conservative Party, as they took 16 in 2004 and 14 in 2006. Neither the Liberals nor the New Democrats have managed to break into this region, as in 2004 the two parties took one each and in 2006 the Liberals took two and the New Democrats three.

Things are far more competitive in Vancouver and along the Pacific coast. In 2008, the Conservatives and Liberals each took three seats, with five going to the NDP. In 2006, the Tories only took one seat, while the Liberals took six and the NDP four. In 2004, it was again evenly divided, with the Tories and NDP taking three seats and the Liberals taking five. Overall, the region has a slight lean towards the Liberals, who have won 14 seats over the last three elections, compared to 12 for the New Democrats and seven for the Conservatives.

Vancouver Island is even more of a toss-up, but is slightly more favourable to the Tories. They took three seats in 2008 while the NDP took two and the Liberals one. Over the last three elections, the Tories have taken eight seats, the New Democrats six, and the Liberals only four.

Now that the electoral history is out of the way, we can look at the voting trends since December.

The Tories started out the year strongly in British Columbia, holding a clear lead from December to the end of March. Support went as high as 56% (in December) and 50% at the beginning of March. There was a slip in Tory support around the New Year, as the party polled in the high 30s. But neither the Liberals nor the New Democrats were able to separate themselves from the pack, as the Conservative drop coincided with a Liberal and NDP increase to the 20s.

Things changed at the beginning of April, when the Tories and Liberals were tied at 35%-34%. The Tories then dropped even further to third place at 26% (tied with the Liberals). The Greens saw their high-watermark in this poll, at 16%. Since then the Tories have moved themselves back into first place, but their lead is not as clear. While they polled 42% in a late April Ipsos-Reid poll, the party also polled 31% in the massive EKOS poll late last month. The beneficiaries of this Tory instability have been the Liberals, who have moved themselves comfortably into the high-20s, low-30s. This doesn't put them in a position to take the lead, but does put them within striking distance of the Tories when they have a bad poll.

The New Democrats have been stagnant in the high-teens and low-20s, which translates into a big seat loss for them. A downward trend has appeared for them here since the beginning of May, as the NDP has moved from 25% to 15% in a steady decline.

Interestingly, looking at the polling chart you can see that often when the Conservative number is low it is the Greens or the NDP who benefit, not the Liberals. This could just be a coincidence, and a result of, say, Liberals moving to the NDP and Tories moving to the Liberals, but it is a counter-intuitive trend.

As a general conclusion for each of the parties based on the chart, it is clear that the Conservatives have lost a little momentum. They've regained some of it recently, but overall their trend seems to be downward. The Liberal trend is very steady, but is slightly trending upwards. The New Democrats have been steady as well, and appeared to be trending strongly upwards between February and April, but have since started to go downhill. The Greens have maintained themselves between 15% and 5%, which puts them in a good position to repeat 2008's result.

I'm currently projecting the Tories to keep their 22 seats, but the Liberals could reach 12 seats, the highest they've been since the formation of the Conservative Party. I project a disastrous result for the New Democrats with only two seats. Things won't improve for them unless they can get themselves over the 26% mark.

Monday, June 8, 2009

Polling Firm Tendencies (Update)

I've added Nanos Research to the polling firm leaning rankings. The methods of Nanos tend to favour the Liberals and under-poll the Bloc Quebecois. The Conservatives fare worse in Nanos polls nationally and in Ontario, but do slightly better in Quebec. The NDP does slightly better nationally and in Quebec, but worse in Ontario. The Greens do worse nationally and in Quebec, but poll slightly above average in Ontario in Nanos surveys.

These are the rankings - to be constantly added to and updated - of the various polling firms in terms of their tendencies to favour one party over another. This does not necessarily equate to a deliberate bias, but instead is more reflective of the polling methods used.

As you can see, the list is not complete yet.

The following lists show each pollster's average variation from other polling firms. The numbers are the amount of percentage points a particular pollster favours or disfavours that particular party compared to other pollsters.

NATIONAL - CONSERVATIVE PARTY

1. Ipsos-Reid +2.9
2. Angus-Reid Strategies +1.9
3. Léger Marketing -0.4
4. EKOS Research -1.1
5. Nanos Research -1.5
6. Harris-Decima -2.6

NATIONAL - LIBERAL PARTY

1. Nanos Research +2.6
2. EKOS Research +2.3
3. Léger Marketing +2.2
4. Ipsos-Reid +0.1
5. Harris-Decima -1.1
6. Angus-Reid Strategies -1.9

NATIONAL - NEW DEMOCRATIC PARTY

1. Angus-Reid Strategies +0.7
2. Nanos Research +0.4
3. Harris-Decima -0.0
4. Léger Marketing -0.2
5. EKOS Research -1.5
6. Ipsos-Reid -2.3

NATIONAL - GREEN PARTY

1. Harris-Decima +2.1
2. EKOS Research +1.3
3. Ipsos-Reid -1.1
4. Nanos Research -1.2
5. Angus-Reid Strategies -2.2
6. Léger Marketing -2.6

ONTARIO - CONSERVATIVE PARTY

1. Angus-Reid Strategies +3.2
2. Ipsos-Red +1.6
3. Harris-Decima -1.0
4. Léger Marketing -1.4
5. Nanos Research -1.6
6. EKOS Research -1.7

ONTARIO - LIBERAL PARTY

1. Léger Marketing +5.0
2. Nanos Research +2.6
3. EKOS Research +2.1
4. Ipsos-Reid +0.8
5. Angus-Reid Strategies -2.5
6. Harris-Decima -2.6

ONTARIO - NEW DEMOCRATIC PARTY

1. Harris-Decima +0.4
2. Angus-Reid Strategies -0.4
3. Nanos Research -0.7
4. Léger Marketing -1.2
5. Ipsos-Reid / EKOS Research -2.3

ONTARIO - GREEN PARTY

1. Harris-Decima +2.2
2. EKOS Research +1.9
3. Nanos Research +0.6
4. Ipsos-Reid -0.5
5. Léger Marketing -1.8
6. Angus-Reid Strategies -2.0

QUEBEC - CONSERVATIVE PARTY

1. Ipsos-Reid +1.6
2. EKOS Research +0.7
3. Nanos Research +0.2
4. Angus-Reid Strategies -0.2
5. Harris-Decima -2.0
6. Léger Marketing -3.3

QUEBEC - LIBERAL PARTY

1. Nanos Research +5.2
2. Léger Marketing +3.0
3. Harris-Decima +0.4
4. Ipsos-Reid +0.1
5. Angus-Reid Strategies -1.3
6. EKOS Research -4.7

QUEBEC - NEW DEMOCRATIC PARTY

1. EKOS Research +0.3
2. Nanos Research +0.2
3. Ipsos-Reid -0.0
4. Léger Marketing -0.3
5. Angus-Reid Strategies -1.1
6. Harris-Decima -2.3

QUEBEC - BLOC QUÉBÉCOIS

1. Léger Marketing +3.8
2. Harris-Decima +3.0
3. Angus-Reid Strategies +2.7
4. Ipsos-Reid +2.3
5. EKOS Research +1.2
6. Nanos Research -1.5

QUEBEC - GREEN PARTY

1. EKOS Research +3.4
2. Ipsos-Reid -0.7
3. Harris-Decima -1.1
4. Angus-Reid Strategies -1.5
5. Léger Marketing -2.8
6. Nanos Research -2.9

Saturday, June 6, 2009

Liberals Lead in Every Demographic

Behind the national and regional numbers of the most recent Ipsos-Reid poll, the results show something very significant: the Liberals are beating the Conservatives in every demographic.

Nationally, the poll put the Liberals ahead of the Conservatives 36% to 33%. A three point lead is not huge when margins of error are 3.1%, but people get too caught up in the "margin of error". Yes, it does mean that the two parties could be tied, but it also means the gap could be six points instead of three. So let's put that aside for the moment.

The Ipsos-Reid poll divided up the results by age and sex, and in every category the Liberals were ahead.

Among 18-34 year olds, the Liberals had 31% to the Conservative 28%. In the previous Ipsos-Reid poll, where the Tories had a lead over the Liberals (35% to 33% nationally), the margin was only 30% to 29% in this age group.

Among 35-54 year olds, the Liberals have a 35% to 33% lead. That is a huge swing from Ipsos-Reid's last poll at the end of May, when the Tories had a significant 37% to 32% lead in this age group.

This change of fortune continues in the 55+ age group, where the Liberals now have a 41% to 36% lead. At the end of May, the Conservatives were ahead here 39% to 36%.

While before males were split on the two parties at 34% apiece, the Liberals are now ahead 37% to 34%. While at the end of May Ipsos-Reid had women preferring the Conservatives 37% to 31%, that has now swung to the Liberals where they lead 34% to 31%.

That the national lead in Ipsos-Reid polling has swung from the Conservatives to the Liberals is important enough. But that every demographic has moved from the Conservatives to the Liberals is a far more significant fact. If the Liberals can beat the Conservatives in any demographic group, they have a chance to beat them in every part of the country.

New Polls: Nanos and Ipsos-Reid

Two new polls were released today. The Nanos Research poll involved 1,001 interviews and was taken between May 26 and June 1. The Ipsos-Reid poll involved 1,018 interviews and was taken between June 2 and June 4.

During the Nanos poll, the biggest news was the $50 billion deficit. It made top headlines from the 27th to the 29th. From the 30th to the 1st, the "tapes" comment about Ignatieff was made by Harper, Conservative MP Pierre Poilievre made the "tar baby" comment, and Layton announced he didn't want an election this summer. During the Ipsos-Reid poll, the biggest news was Natural Resources Minister Lisa Raitt leaving behind government documents at CTV. From these stories, one would not expect strong polling from the Conservatives.

Here are the national results, Nanos first and Ipsos-Reid second:

Liberals - 37.2% / 36%
Conservatives - 31.8% / 33%
New Democrats - 15.7% / 12%
Bloc Quebecois - 8.0% / 9%
Greens - 7.4% / 9%

For the details of both polls, you can take a look at the bottom of the page.

This last Ipsos-Reid poll would result in 130 seats for the Liberals, 113 for the Conservatives, 51 for the Bloc Quebecois, and 14 for the New Democrats.

The short-term projection has changed. The national vote is as follows, with the difference from last time in brackets:

Liberals - 34.5% (+0.8)
Conservatives - 32.6% (+1.1)
New Democrats - 14.8% (-0.6)
Bloc Quebecois - 8.7% (-0.2)
Greens - 8.4% (-1.1)

The short-term seat projection has changed as well:

Liberals - 121 (+1)
Conservatives - 118
Bloc Quebecois - 48 (-1)
New Democrats - 20 (-1)
Greens - 1 (+1)

The long-term projection has changed a little as well. The Conservatives are up to 121 seats while the Bloc Quebecois has dropped by one to 49. The Conservative seat came in British Columbia, where the Liberals lost one. They made good that last with a gain in Quebec at the hands of the Bloc.

Nationally, the Liberals have widened the gap between them and the Tories, gaining 0.3 points while the Conservatives have lost 0.1. The New Democrats have dropped 0.1 points, as have the Bloc. Regionally, there weren't any huge moves, no party losing or gaining more than 0.2 points.

These two polls are good for the Liberals, especially the Nanos poll. The 5.4 lead they had in that poll over the Conservatives is the biggest we've seen since an EKOS poll in early April. Both polls were bad for the Conservatives, particularly the 8% result the Tories had in Quebec in the Ipsos-Reid poll. The NDP fared better in the Nanos poll, but 12% in the Ipsos-Reid poll would be disastrous.

Friday, June 5, 2009

Polling Firm Tendencies (Update)

I've added Harris-Decima to the polling firm leaning rankings. Overall, Harris-Decima's method is less favourable to both the Conservatives and the Liberals, and is most favourable to the Greens and to the Bloc Quebecois.

These are the rankings - to be constantly added to and updated - of the various polling firms in terms of their tendencies to favour one party over another. This does not necessarily equate to a deliberate bias, but instead is more reflective of the polling methods used.

As you can see, the list is not complete yet.

The following lists show each pollster's average variation from other polling firms. The numbers are the amount of percentage points a particular pollster favours or disfavours that particular party compared to other pollsters.

NATIONAL - CONSERVATIVE PARTY

1. Ipsos-Reid +2.9
2. Angus-Reid Strategies +1.9
3. Léger Marketing -0.4
4. EKOS Research -1.1
5. Harris-Decima -2.6

NATIONAL - LIBERAL PARTY

1. EKOS Research +2.3
2. Léger Marketing +2.2
3. Ipsos-Reid +0.1
4. Harris-Decima -1.1
5. Angus-Reid Strategies -1.9

NATIONAL - NEW DEMOCRATIC PARTY

1. Angus-Reid Strategies +0.7
2. Harris-Decima -0.0
3. Léger Marketing -0.2
4. EKOS Research -1.5
5. Ipsos-Reid -2.3

NATIONAL - GREEN PARTY

1. Harris-Decima +2.1
2. EKOS Research +1.3
3. Ipsos-Reid -1.1
4. Angus-Reid Strategies -2.2
5. Léger Marketing -2.6

ONTARIO - CONSERVATIVE PARTY

1. Angus-Reid Strategies +3.2
2. Ipsos-Red +1.6
3. Harris-Decima -1.0
4. Léger Marketing -1.4
5. EKOS Research -1.7

ONTARIO - LIBERAL PARTY

1. Léger Marketing +5.0
2. EKOS Research +2.1
3. Ipsos-Reid +0.8
4. Angus-Reid Strategies -2.5
5. Harris-Decima -2.6

ONTARIO - NEW DEMOCRATIC PARTY

1. Harris-Decima +0.4
2. Angus-Reid Strategies -0.4
3. Léger Marketing -1.2
4. Ipsos-Reid / EKOS Research -2.3

ONTARIO - GREEN PARTY

1. Harris-Decima +2.2
2. EKOS Research +1.9
3. Ipsos-Reid -0.5
4. Léger Marketing -1.8
5. Angus-Reid Strategies -2.0

QUEBEC - CONSERVATIVE PARTY

1. Ipsos-Reid +1.6
2. EKOS Research +0.7
3. Angus-Reid Strategies -0.2
4. Harris-Decima -2.0
5. Léger Marketing -3.3

QUEBEC - LIBERAL PARTY

1. Léger Marketing +3.0
2. Harris-Decima +0.4
3. Ipsos-Reid +0.1
4. Angus-Reid Strategies -1.3
5. EKOS Research -4.7

QUEBEC - NEW DEMOCRATIC PARTY

1. EKOS Research +0.3
2. Ipsos-Reid -0.0
3. Léger Marketing -0.3
4. Angus-Reid Strategies -1.1
5. Harris-Decima -2.3

QUEBEC - BLOC QUÉBÉCOIS

1. Léger Marketing +3.8
2. Harris-Decima +3.0
3. Angus-Reid Strategies +2.7
4. Ipsos-Reid +2.3
5. EKOS Research +1.2

QUEBEC - GREEN PARTY

1. EKOS Research +3.4
2. Ipsos-Reid -0.7
3. Harris-Decima -1.1
4. Angus-Reid Strategies -1.5
5. Léger Marketing -2.8

Thursday, June 4, 2009

Monthly Picture: May

Time goes on, and now we can average out the polls from May. Five national polls were taken during this month, totalling 15,928 interviews. A further three polls were taken in Quebec alone. Here are the results we get at the national level, with the difference from last month's average in brackets.

Liberals - 33.7% (-0.9)
Conservatives - 31.5% (-0.2)
New Democrats - 15.4% (+0.5)
Greens - 9.5% (+1.3)
Bloc Quebecois - 8.9% (-0.5)

The big winners this month are the New Democrats and Greens. The Liberals and the Bloc are the losers of May. The seat projection for these results is as follows, with the difference from last month in brackets:

Liberals - 121 (-11)
Conservatives - 115 (+6)
Bloc Quebecois - 49 (unchanged)
New Democrats - 22 (+4)
Greens - 1 (+1)

So we end up with a razor thin Liberal minority. The Conservatives and New Democrats were the beneficiaries of the Liberal slip. While in April the Tories and NDP couldn't combine to defeat the projected Liberal government, they can now manage it easily.

The regional results, with difference from last month in brackets:

BRITISH COLUMBIA (four polls)

Conservatives - 36.8% (+2.3)
Liberals - 27.6% (-2.6)
New Democrats - 23.5% (-0.5)
Greens - 10.5% (-0.1)

ALBERTA (four polls)

Conservatives - 54.4% (-4.5)
Liberals - 20.7% (-0.1)
Greens - 13.3% (+2.6)
New Democrats - 10.3% (+2.1)

PRAIRIES (four polls)

Conservatives - 45.7% (+0.8)
New Democrats - 22.4% (+2.3)
Liberals - 21.1% (-6.7)
Greens - 9.7% (+3.8)

ONTARIO (five polls)

Liberals - 40.3% (-3.1)
Conservatives - 35.1% (+0.7)
New Democrats - 14.2% (+0.1)
Greens - 9.3% (-0.3)

QUEBEC (eight polls)

Bloc Quebecois - 36.6% (-0.9)
Liberals - 34.3% (+1.0)
Conservatives - 12.4% (-0.2)
New Democrats - 10.8% (+0.8)
Greens - 5.9% (+0.3)

ATLANTIC CANADA (four polls)

Liberals - 35.5% (-7.0)
Conservatives - 29.6% (+0.9)
New Democrats - 24.3% (+0.8)
Greens - 9.4% (+4.4)

The New Democrats gained points in five of the six regions, with the 2.3 point bump in the Prairies the largest. The Conservatives gained in four of the six regions, British Columbia first among them. The loss of 4.5 percentage points in Alberta, however, is significant. It makes Liberals and New Democrats electable in the province. The Greens also gained in four regions, the 4.4 point increase in Atlantic Canada giving them a seat (Elizabeth May, assuming she runs in Central Nova again). The Liberals, however, lost in five of the six regions, a big reversal of recent trends. And while they only dipped 2.6 points in British Columbia and 0.1 points in Alberta, they lost almost 7 points each in the Prairies and Atlantic Canada. Their 3.1 point loss in Ontario isn't huge, but in such a battleground province it can mean a significant number of seats. This leads one to ask whether the attack ads actually have been working after all.

Wednesday, June 3, 2009

Voter Profiles

Thanks to the incredibly detailed EKOS poll from earlier this week, we're able to determine what kind of voter each party attracts. You might be surprised by some of the results.

The Conservative voter is male, over 45 (and even more heavily over 65), does not live in Quebec, and is economically well off. Those who make more than $80,000 per year are especially enamored with the Tories (38.4%), but the middle class ($40,000 to $80,000) also lean slightly more towards the Tories than the average. Out of the five major metropolitan regions, the Conservative voter is most likely to live in Calgary or, to a lesser extent, Ottawa-Gatineau. There are some regional variations to this profile, however. In Atlantic Canada, the Conservative voter tends to be younger and more middle-class than the rest of the country.

The Liberal voter can be either female or male, but is slightly more likely to be female. He or she is 45 years old or older, and more likely to be over the age of 65. He or she is economically well-off, with the $80,000+ group most heavily weighted. The Liberal voter is less middle-class than the Conservative voter. He or she likely lives in Vancouver, Toronto, or Ottawa-Gatineau. The Liberal voter is more regionally varied, however. In British Columbia, the Prairies, and Ontario, he is male. In Alberta, she is under the age of 25 and middle-class. The Liberals do better amongst the youth in Alberta likely because the Conservatives are the "establishment", and it is more rebellious to be a Liberal here than anywhere else in the country.

The New Democratic voter is female, under the age of 25 and to a lesser extent between the ages of 25 and 44, and makes less than $40,000 per year. She is most likely to live in Vancouver. In Quebec, however, the NDP voter is slightly more likely to be male, and in Atlantic Canada she is middle-aged and more middle-class.

The Green voter is a lot like the NDP voter. She is female, under the age of 25 or perhaps middle-aged, and makes less than $40,000 per year. She is most likely to live in Toronto. In Quebec and Atlantic Canada, however, the Green voter is male.

The Bloc voter is slightly more likely to be female. She is between the ages of 25 and 64 and is middle-class, and probably lives in Montreal.

When you take all of this into account, it is easy to see where the battle lines are drawn. The Tories do well amongst males while the New Democrats and Greens do best amongst females. The Liberals and the Bloc are more gender neutral, and are thus either fighting the Conservatives for the male vote or the NDP and Greens for the female vote. The Conservatives and Liberals are the choice of older people, while the NDP and Greens are the choice of younger people. The Conservatives and Liberals wrestle over the upper class vote while the NDP and Greens fight over the lower class. Each party has its niche, and all have to dip in the large pool of middle aged and middle class voters to make gains. "Establishment" voters, people who vote for the Liberals or Conservatives, are older and richer. "Opposition" voters, people who vote for the New Democrats or Greens, are younger and poorer. It all aligns very well to the political spectrum.

One exception is the Bloc Quebecois. Unlike any other party, they are not the domain of any one group. Only a few more females than males choose them, and they are the favourite party of people who aren't very young or very old, and who aren't very poor or very rich. The "average" person in Quebec, thus, is a Bloc Quebecois voter. Only when you go to the extremes of age and class do you find the other parties doing better. It seems to me an interesting demonstration of how the Bloc Quebecois is less of an "interest" party than the other four. Aside, of course, from Quebec's interests.

These results show, in part, why the Liberals and Conservatives are the parties likely to form government. Older people vote for them, and older people vote in greater numbers. Richer people vote for them, and richer people can donate more money. The NDP and Greens are at a distinct disadvantage because their support comes from people who vote less and financially contribute to parties less. It is a formula for perpetual opposition.

Monday, June 1, 2009

The Pros and Cons of the Leaders

The Angus-Reid poll released yesterday included a question concerning how each of the party leaders are perceived. People had to choose whether they felt certain qualities were associated with each of the leaders. I've broken them down into two categories: good and bad.

Simple enough. I considered whether the survey takers considered a party leader exciting, down to earth, compassionate, honest, efficient, strong, or intelligent as good qualities, and arrogant, uncaring, boring, secretive, dishonest, inefficient, weak, out of touch, and foolish as bad qualities. I averaged them out, and these are the results I got:

People generally respond best to Elizabeth May, Michael Ignatieff, and Jack Layton. On average, about 24.6% of respondents associated her with good qualities. Ignatieff and Layton were close behind. Stephen Harper was further behind, with only 16.9% of respondents attributing good qualities to him. Only 16% considered Gilles Duceppe to have good qualities. However, I don't consider the Duceppe result all that indicative, as people outside of Quebec are likely to have a negative opinion of him simply because of the party he represents and because they don't know much about him. And really, the opinion people from outside of Quebec have of Duceppe is irrelevant.

Negative attributes were more strongly associated with Harper, with 34.9% of respondents. This is a much more significant number than those who considered any of the leaders as having good qualities. People react more strongly to Harper negatively than they do any of the leaders positively. Duceppe follows Harper, but Layton is considered to have more negative qualities than either Ignatieff or May, by a slim margin. May is considered to have the least.

The qualities can be broken down even further into five categories. They are general likability, general unpleasantness, being untrustworthy, being competent, and being incompetent.

The first, which groups the qualities of being exciting, down to earth, compassionate, and honest, is the domain of Elizabeth May and Jack Layton. People just generally like them, and attributed these qualities to them more than anyone else.

The second, which groups the qualities of being arrogant, uncaring, and boring, is owned by Stephen Harper. People just don't seem to like him on a personal level.

The third, which groups the qualities of being secretive and dishonest, is also a Harper trait.

The fourth, which groups the qualities of being efficient, strong, and intelligent, is Michael Ignatieff's. He may not be the most likable person, but people seem to respect his competency.

The fifth, which groups the qualities of being inefficient, weak, out of touch, and foolish, is shared by Layton and Harper. Stephane Dion used to corner the market of incompetence, at least in the public eye. And Layton, while liked, hasn't been trusted to run an economy. But now Harper is being lumped in with those who are perceived to be bad administrators, no doubt in response to the record deficit. Harper drew the descriptions of inefficiency and out-of-touchness, while Layton was considered weak and foolish.

Taking the top attributes, Harper is described as secretive (54%), arrogant (53%), and out of touch (38%). He is not consider open (11%), compassionate (10%), or exciting (3%). This sounds like an unpleasant person.

Ignatieff is thought to be intelligent (53%), arrogant (42%), and strong (31%). He isn't exciting (12%), compassionate (12%), or foolish (9%). Sounds like the typical cold but decisive academic.

Layton is perceived as compassionate (34%), arrogant (32%), and down to earth (31%). But he isn't efficient (12%), secretive (12%), or uncaring (10%). It is difficult to be both down to earth and arrogant, but Layton manages it.

Clearly, Stephen Harper has some personal image issues. Michael Ignatieff is respected, but not liked. Elizabeth May and Jack Layton are liked, but not respected. When you look back at the different strategies each of the leaders has used in the past, you can see that they have all come to the same conclusions as I have and have been trying to bridge the gap. The "sweater-vest" campaign advertisements were an attempt to make Harper more likable. Both Layton and May tried very hard during the last election to demonstrate that their parties had real ideas. It would seem that all of these attempts have failed; Harper's ads came off as phony, the Greens made no significant breakthroughs and people openly feared the possibility of the NDP making cabinet decisions.

This leaves one to wonder whether these three party leaders have reached the end of their re-inventability. People have their opinions about Stephen Harper, Jack Layton, and Elizabeth May, and it may be impossible to change them.

Michael Ignatieff is still relatively new, and we haven't seen any attempt by the Liberals to paint their leader as a "regular guy". That could come, and probably will. Whether it will work is another matter. Canadians seem to be pretty good at perceiving their leaders for who they really are, and in my experience will rarely succumb to parties' manipulative attempts to change their opinions.

Parties who do best play to their leaders' strength. People, and parties, can't be someone or something that they simply aren't.

Mega Poll Projection Change

You'll all remember where you were when you heard about the monster Ekos poll of 2009.

The combination of the largest poll in Canadian history, an Angus-Reid poll, and the end of the month, when all older polls are reduced in weight, has resulted in a significant change in the projection. This Ekos poll is massive - 10,896 interviews between May 7 and May 28. And it is broken down in so many ways that I will have to analyse it at greater length later this week. I outlined the national polling result in a post this morning, so for now let's look at the regional data.

In British Columbia, the Liberals are ahead with 32.3% to the Conservative 31.0%. The NDP is third with 23.0%. Now, this is a bit of an outlier result, despite the 1,254 respondents. The last Liberal lead in BC was in April in an Ekos poll. So, perhaps there is something in the Ekos methodology that favours the Liberals in BC. Or, perhaps respondents were confused by the BC provincial election taking place during this poll. In that election, only the provincial Liberals and NDP were in the race.

The Alberta, Prairie, and Atlantic numbers are nothing special, so you can check them out in the detailed tables at the bottom of the page. The Ontario result, 39.4% to 34.3% in favour of the Liberals, is also not much different from what we've been seeing.

In Quebec, however, the Conservatives pulled a - relatively - strong 16.3%. Most polls have had the Tories in the low teens here. It is possible that the Ignatieff attack ads have been received better by Quebecers. This wouldn't surprise me, because while they contain silly things like his accent, the French ads do contain some actual political topics. However, at 31.5% the Liberals haven't lost much support in Quebec. The Bloc, at 34.7%, is also only slightly lower than the norm.

I'll take a look at the specifics of the poll later in the week, as there is just too much to cover here.

This poll alone, and considering its size this is something significant, would result in the following seat totals:

Liberals - 124
Conservatives - 114
Bloc Quebecois - 47
New Democrats - 23

The short-term, five-poll projection has changed as well. However, since this projection doesn't take into account poll size, it hasn't changed as much. The change from the previous projection (before CROP, Angus-Reid, and Ekos) is in brackets:

Liberals - 33.7% (-1.1)
Conservatives - 31.5% (-0.5)
New Democrats - 15.4% (+0.8)
Greens - 9.5% (+0.5)
Bloc Quebecois - 8.9% (-0.1)

As for the seats:

Liberals - 120 (-7)
Conservatives - 118 (+5)
Bloc Quebecois - 49 (-1)
New Democrats - 21 (+3)

Now, the long-term projection. The Ekos poll has a huge weight and so the poll has swung pretty widely. The seat projection has now changed drastically. From 126 seats the Conservatives have dropped to 120. The Liberals have risen from 111 to 117, while the Bloc Quebecois and New Democrats have remained steady at 50 and 21.

The Conservatives lost two seats each in British Columbia, Alberta, and Ontario. The Liberals gained three in British Columbia and two in Alberta and Ontario. They lost one in Atlantic Canada to the NDP, who lost one of their own seats in British Columbia.

Popular vote swings were large in British Columbia, with support going from the Tories to the Liberals. In Quebec, the Bloc lost almost one point to the benefit of the Conservatives and the Greens. The Liberals almost lost a point as well, in Atlantic Canada to the Greens.

The projection has been classified as an "Unstable Co-Operative Government", because with such a small seat gap between the two major parties, either could govern, but only with the support of other parties or aisle-crossers. This would likely only result in a coalition government or a quick re-election.

This is a difficult poll for every party. Everyone has something to lose. The Bloc will lose some seats and clout. The NDP will lose a quarter or even a third of their seats, which would be disastrous for them. The Conservatives stand to lose power. And the Liberals could lose on the gamble, and end up just outside of power or with a small, unworkable minority.

Now that Jack Layton has said he won't side with the Liberals and force an election at the end of the month, the likelihood of an election is low. With a poll like this, the Liberals would be taking a huge risk by pushing the country into an election. I would imagine they're wise enough to realise this, so it looks like we'll have to wait and see where the chips fall in, well, the fall.

Angus-Reid Details

The details of the Angus-Reid poll are available.

This poll contains a few notable differences from the norm. In Quebec, the NDP is listed at 17%, putting them well ahead of the Conservatives. In the Prairies, the NDP is up to 28%, above the Liberals at 21% and behind the Tories at 38%. Finally, in the Atlantic, there is a three way tie: Conservatives and NDP at 29% and the Liberals at 28%. Of course, the Atlantic polling sample is always one of the smallest, so this is an outlier result.

This is the amount of seats each party would win if this poll were the electoral result:

Conservatives - 120
Liberals - 114
Bloc Quebecois - 48
New Democrats - 25
Greens - 1

Despite being a few points below the Liberals nationally, the Tories still form government on the strength of their 71 seats west of Ontario. The Liberals manage to get themselves within a few seats of the Conservatives on the strength of their 82 seats in Ontario and Quebec. The Atlantic result is what throws this seat projection out of whack, as the three-way tie results in 14/10/7 seats for the Liberals, Conservatives, and NDP, respectively. It even allows the Greens to go through the middle and win a seat (all the more likely now that Peter MacKay might not run in the next election). If the Atlantic poll had been within the norm, this poll would likely have resulted in a slim Liberal minority.

Two New Polls, One of them a Monster

Two new polls were released today, one by EKOS Research Associates and the other by Angus-Reid Strategies. The EKOS poll is a monster: 10,896 interviews taken between May 7 and May 28. The CBC has apparently signed a major deal with EKOS, as they will be releasing polling data every Thursday starting on June 11. Lots of fodder for the projection! Apparently, the polling data swung hard against the Conservatives when the news of the $50 billion deficit came out. It will be interesting to see what the polling numbers will be on June 11.

The Angus-Reid poll was conducted between May 28 and May 29 and included 1,002 interviews. The dominating headline on May 28 was the deficit, but on the 28th and the 29th the media was also reporting on the veiled Harper "tape" threat against Ignatieff, the wrong-doing of CTV in the Dion gaffe affair, and continued questions about the deficit and how much it would grow. To sum up, two bad days for the Conservatives.

Here are the results from the two polls, with EKOS first and Angus-Reid second:

Liberals - 33.5% / 33%
Conservatives - 32.3% / 31%
New Democrats - 15.1% / 17%
Greens - 10.4% / 7%
Bloc Quebecois - 8.7% / 9%

I love EKOS and their decimal-point polls. The consistency between these two polls is very good. But Angus-Reid was taken on the 28th and 29th while EKOS polled throughout the month. Considering the news that was hitting the Conservatives during the Angus-Reid poll, we shouldn't be surprised that the Conservatives have a worse result there. It is even possible that the NDP has benefited in the Angus-Reid poll from disaffected Conservatives who can't bring themselves to vote Liberal.

One thing seems clear: the Conservative attack ads haven't worked. The Liberal number might be a point or two lower than we've seen recently, so it is possible that the attack ads have managed to stop Liberal growth and maybe push some people away from the party. But two low numbers for the Conservatives demonstrates that the attack ads did not send anyone to the Tories, which is understandable. An attack ad doesn't necessarily, and in this case not at all, push people towards the party on the attack. Nothing in those Ignatieff ads tells the viewer to vote Conservative. Attack ads can, however, tell the viewer that the attacker plays dirty politics. It is quite possible that some viewers were turned off by Ignatieff because of the ads. But there is no indication that those viewers went to the Tories in droves. On the face of it, the ads have changed nothing.

EKOS also asked who would make the best Prime Minister. Unfortunately, it seems they limited the answers to Harper or Ignatieff, so I can't add it to my Best PM tracker. In any event, Harper edged out Ignatieff 30% to 26%, with the rest saying "neither".

The regional breakdown in the Angus-Reid poll is spotty, but what has been released is a Liberal lead in Ontario (43% to 36%), a Tory lead in British Columbia (45% to 26%), and a Bloc lead in Quebec over the Liberals (36% to 33%, with the Conservatives at 10%).

Unfortunately, because I don't have the complete details, the poll can't be updated just yet. However, both EKOS and Angus-Reid are relatively quick with posting new information, so hopefully I can have an update before the end of the day or tomorrow.