Monday, February 26, 2007

Indecisve Quebec Campaign Continues

A new Léger poll out today is more favourable to Jean Charest's PLQ. It would deliver 57 seats to Charest, a Liberal minority rather than the CROP poll's PQ minority but continuing evidence of an indecisive outcome.

Sunday, February 25, 2007

Growing number of American liberals

I missed this Pew survey when it was first released in October 2006. But its findings are quite intriguing. Among self-identified Democrats, as recently as 2002 as many Democrats considered themselves conservatives as liberals. Since then there has been a significant increase in the share of liberals. While overall ideological trends appear stable, as I noted in an earlier post, there is other research that reports a strong increase in support for the Democrats.

A key ambiguity in the survey is that is uses the designation "moderate" to capture those in the middle. One problem with this is that the word "liberal" has been demonized for a number of years in popular and media discourse. My guess is that if one parsed the moderates, one would find that, in substance, more are really liberal than conservative, both among Democrats and among the population as a whole. This would account, for example, why overall there has been an even partisan division within the U.S. for the past number of years despite a nominal advantage for conservatives. The trend at the moment is to a Democratic advantage.

Saturday, February 24, 2007

Quebec Election - PQ Starts Ahead

This election looks like a classic set-up for the third party: there is a relatively unpopular Premier and administration (only 42% say they are satisfied with the Charest government) and an even more unpopular leader of the Parti québecois – André Boisclair. The third party, the ADQ, is led by Mario Dumont, who according to this CROP poll is the preferred leader when respondents are asked who they would want to babysit their children or with whom they would want to spend an evening. (Charest leads when the question is who would make the best Premier.)

But I don’t think the ADQ can win. The PQ remains well ahead among francophones and the PLQ dominates the anglo-allo vote (see detailed chart here). The ADQ has only 3% among Anglophone voters who don’t trust Dumont for his support of the YES side in the 1995 referendum, and is five points behind the Liberals among francophone voters. However, the party has made itself as a force to be reckoned with by appealing to more rural, conservative voters, essentially the old Union Nationale base. This may have helped entrench the ADQ but it also prevents it from succeeding overall in a province that is well to the left of it ideologically, among both anglophones and francophones.

So the story is more complicated, but there are reasons for Quebec voters to find the ADQ as less than satisfactory.

What we are likely to see therefore is an indecisive result, most likely a minority government, which Quebec has not had in the modern era of politics.

If one applies this week’s CROP to my model, I get the following outcome:

Party Seats
PLQ 52
PQ 60
ADQ 12
QS 1
Others 0

So this campaign starts off with André Boisclair leading in terms of seats but with less than a majority. The lack of confidence in the alternatives also makes it more likely that the campaign could matter a great deal.

Harper's Character

I think Aislin gets it right.

Tuesday, February 20, 2007

New Strategic Counsel Poll

Aside from allowing the Conservatives to cue their poll questions, the problem with all this leadership focused polling by the media (such as the Globe poll by Strategic Counsel) is that it is misleading. Leadership preferences tend to follow partisan switches not lead them. This Paul Wells post nails it.

Indeed, Joe Clark in 1979, Jean Chrétien in 1993 and Dalton McGuinty in 2003 were all deemed less popular than their opponents at the outset of their campaigns but won anyway.

In fact, I can't believe that most Canadians have any firm view of Dion who has held his position less than three months. Quebeckers perhaps but not English Canadians. This entire discussion amongst the pundit class strikes me as a classic case of Ottawa Rideau fever.

The Liberals have real liabilities going into the campaign - organization, money (meaning an ability to respond in kind to the Tory ads), division on issues (Afghanistan, extending the anti-terrorism law, etc.) but so-called weak leadership impressions at this stage aren't among them.

Sunday, February 18, 2007

Leading Indicator of 2008 U.S. Election

One of the leading analysts of American politics is Charlie Cook. He began to sense weakness in the Bush administration and the Republican Congress long before the 2006 election.

Now his analysis of recent Gallup polls, pinpoints a strong shift to the Democrats underway. Gallup has been asking in all its polls with which party a respondent identifies. Because they have an annual sample of about 30,000 the margin of error is about ± 0.5%. The trend has been shifting the Democrats way for a number of years. I often read, however, about the increasing tendency, especially with younger voters, to call themselves independent. Gallup has been probing these independents for 15 years to see if they lean to the Democrats or the Republicans.

As Cook puts it about Gallup's new numbers in his latest column:

...the real jaw dropper is when independents are asked which party they lean toward. This is important because historically, independents who lean toward a party tend to vote almost as consistently for that party as those who identify themselves with the party. There are just some people who like to call themselves independents but, functionally speaking, are really partisans.

In this category of leaners, Democrats had an advantage of 1.3 points in 2001. The parties were within the margin of error in 2002, when four-tenths of a point separated them and in 2003, when there was just a one-tenth of a point difference.

In 2004, Democrats had a 2.7 point advantage, and it grew to 4.4 points in 2005.

But in 2006, this category exploded to a 10.2-point advantage for Democrats: 50.4 percent for Democrats, 40.2 percent for Republicans. The remaining 9.4 percent did not lean toward either party.

This 10.2-point advantage is the biggest lead either party has had since Gallup began tracking the leaners in 1991.

Perhaps most alarming for Republicans is that for the Gallup interviews conducted during the last quarter of 2006, the Democratic advantage was a whopping 14.2 points: 52.3 percent for Democrats and 38.1 percent for Republicans.

This to TC is an early sign of a potential Democratic blowout in 2008, taking the presidency, keeping the House of Representatives, and ending up in the upper fifties in the Senate. The Iraq war seems well on its way to destroying Republican dominance in the United States every bit as much as the Vietnam war destroyed the Democrats. Their downfall was postponed by the Watergate scandal but they were not able to avoid it. The Republicans appear to be headed for a similar fate.

Friday, February 16, 2007

Nixon or Bush?

So who is the worst liar in American history: George W. Bush or Richard Nixon. I watched a riveting documentary this week on PBS's Frontline documentary series on the state of the American media, which was focused on the Plame case.

I found more of the interview with Carl Bernstein on the Editor and Publisher web site. The answer to the question posed in the title of this post:

Nixon and his men lied and abused the constitution to horrible effect, but they were stopped.

The Bush Administration -- especially its top officials named above and others familiar to most Americans -- was not stopped, and has done far greater damage. As a (Republican) bumper-sticker of the day proclaimed, 'Nobody died at Watergate.' If only we could say that about the era of George W. Bush....

Tuesday, February 13, 2007

New Decima Poll similar to SES

The Conservatives commenced their French language Dion swiftboating ads today, making it that much more apparent that this is a pre-election period. A Decima poll out today has findings close to last week's SES survey, reporting a statistical tie between the Conservatives and the Liberals. TC also heard today of a private poll from two weeks ago also reporting a tie.

Next week brings a Quebec election call with a new poll today giving the Quebec Liberals a five point lead - 36 to 31 over the PQ with the ADQ at 21. These numbers, however, would give us a PQ minority according to my forecaster. The PLQ is too weak at the moment among francophones to win. The one thing that might derail a federal election would be a loss by Charest in Quebec. It would give Harper pause, but would it then give all the opposition parties an incentive to go after him, seizing an advantage based on the wounds Harper would suffer as a consequence of losing a key ally? At the moment an imponderable.

Sunday, February 11, 2007

Léger and SES Polls

There were two national polls out this week from Léger Marketing and SES. When I applied my forecaster to the numbers I found two very different outcomes - the Liberals were ahead based on SES while the Conservatives fell just short of a majority in Léger (but not based on gains in Quebec).

However, if one looks closely at the results and the margins of error, the greatest uncertainty comes from Atlantic Canada where the two polls differ quite widely because their samples are tiny and the margin of error large for both. Elsewhere, although the differences are narrow, the seat outcomes differ significantly because, for example, in Ontario in SES had the Liberals lead the Conservatives while the reverse is true in Léger, but overall the numbers are quite similar in the other regions.

I averaged the two polls and this is what I get:

Conservatives - 128
Liberals - 116
NDP - 21
BQ - 42
Ind. - 1

The Léger numbers are intriguing because if the government believes them or thinks this is what an election result could look like, they would clearly be tempted to get into an early campaign.

Saturday, February 10, 2007

More anti-Kyoto Tories

This post from a Tory blogger is yet another of many examples of how the Conservative base continues in their anti-Kyoto ways. When your base has these attitudes, it limits what Mr. Harper and company can do, and signifies that the opposition parties should not be collaborating with the Conservatives on this issue. If the Conservatives were ever able to obtain a majority (not on the horizon at the moment), they would clearly come under overwhelming pressure from their base to back off.

Friday, February 09, 2007

Ontario Byelections

The results of yesterday's byelections are better news for the Ontario Liberals than might appear to be the case at first glance. Yes, they did lose York South Weston to the NDP, but they held Markham despite efforts by the Conservatives to break into the Chinese community and held their own in Burlington. When I convert the results to province-wide numbers, the outcome would easily produce another Liberal majority.

Looking back over this government's byelection record there are two distinct patterns. One is in low income ridings (Hamilton East, Parkdale High Park, York South Weston, Scarborough Rouge River) where the NDP won the first three and made significant gains in the fourth. This strongly suggests that enough time has passed since the defeat of the discredited Rae government so that the NDP is in a position to recover lost ground in ridings which reflect the long-term pattern of their support. It is clear they can make inroads against the Liberals in poorer urban ridings. This suggests that a strategy similar to the one employed in York South Weston focused on class issues such as the minimum wage could bear electoral fruit.

The other pattern is in the more affluent 905 ridings such as Markham, Burlington and Whitby-Ajax where their vote percentages have been impressive even though they actually did not win two of the three. This bodes well for the Liberals for the October 10 election because the Conservatives simply have to do very well in 905 to win.

The NDP can't win the next election and form a government although they could affect the outcome. They boosted their vote a little in Markham and fell a little in Burlington, even while massively increasing their percentage of the vote in York South Weston. I think this implies that we are seeing continuing strategic voting. Voters who might have the NDP as a first choice but find themselves living in a Burlington appear to be giving up on their party's prospects and voting Liberal.

Addendum: The Greens did not do all that well. The Green vote comes disproportionately from younger, more alienated voters who also tend to have a low turnout. While a recent SES poll suggests 10% support for the Greens in Ontario, the byelections suggest a Green vote of about 5%, similar to the last federal election but much less than the recent media zeitgeist might suggest.

Wednesday, February 07, 2007

Minority Governments

There is a prospect that Canada will elect minority governments in three elections this year - in provincial votes in Ontario and Quebec, and a federal election.

Jean Charest appears anxious to get to the polls as soon as possible to exploit modest improvements in his poll standing and the troubles of André Boisclair. However, applying my forecast model to both the Léger poll and the CROP survey produces minority PQ victories with the ADQ holding the balance of power, and confidence in Charest's government remains limited.

In Ontario, the polls are somewhat more favourable for the McGuinty government. This poll by SES had the Liberals up by eight points, a lead that would produce a Liberal majority, while another by Ipsos-Reid reported a 34 to 33 popular vote lead for the Liberals that would actually produce a Conservative minority government. The Liberal government does seem certain now to face some type of setback in tomorrow's by-elections. A vote of non-confidence in the government would provide more evidence that it will have a tough time winning a second majority.

And all the polls at this point suggest that a federal election would produce another minority government. For example, this Léger poll would produce a Conservative minority of 130 seats with 119 for the Liberals 11 for the NDP and 47 for the BQ. Harper no doubt is interested in winning a majority, but may find it hard to resist the opportunity of going after a new leader still learning the ropes, heading a party short of money and needing time to organize. The money on the attacks would have been completely wasted if the election is not for another year.

Tuesday, February 06, 2007

Fraser Institute Report

The scientists at the blog RealClimate have read the Fraser Institute's effort to discredit the IPCC report and don't like what they see. A few quotes:
Hockey-stick bashing and solar-explains-all advocacy are favorite activities of the denialist camp, so it is no surprise to see both themes amply represented in the Fraser Institute report....

The basic approach taken by the Fraser Institute Report is to fling a lot of mud at the models and hope that at least some of it sticks....

There are so many bizarre statements in the Fraser Institute report that some of us think that spotting them could serve as a good final exam in an elementary course on climate change.

It is worth reading the whole post. It illustrates that, although the climate change deniers are on the defensive, they are far from giving up. Also look at this post on the same subject.

Monday, February 05, 2007

It isn't easy being a green Tory

One of the difficulties of reversing one's traditional ideological views is that there is large blogosphere out there that is likely to undermine you. So following this week's report from the Intergovernmental Panel on Climate Change comes an attack from the (you guessed it) ever reliable Fraser Institute. So what do we find on the Blogging Tories site? None other than this urgent appeal to other like-minded Tory bloggers from Canadian Blue Lemons that "The Fraser Report Must Be Read and Widely Distributed".

John Baird and Stephen Harper want us to believe they care about climate change. Apparently their followers remain to be convinced.