Thursday, March 13, 2014

The Current State of Canadian Politics: Can the federal Liberals win a majority in 2015?

Recently NDP Leader Tom Mulcair discussed his willingness to join a coalition with the Liberals after the next election:
“We’ve always said we’re ready to work with other parties. We’re a progressive party. We want to get results,” NDP leader Thomas Mulcair told reporters when asked if he would be willing to form a coalition with Mr. Trudeau after the election. However, Liberal leader Justin Trudeau rejected the idea suggesting the Liberals would focus on a "winning" strategy.
This incident tells us that the NDP thinks there is a strong possibility that the NDP and the Liberals together can form a majority in the House of Commons after the next election, while the Liberals perceive that they have a chance to form a majority government on their own. With everyone's attention on the Quebec election, perhaps this is as good a time as any to ask whether or not the federal Liberal Party can realistically expect to win a majority in the 2015 federal election.

Based on my tracking of the polls and my seat projections TCNorris has reached some conclusions about this. Based on the trend line of the past year a Liberal majority is unlikely (there is one posssible scenario I will discuss below). Equally a Conservative majority is unlikely, but a plurality is possible. The most likely outcome would see the NDP plus the Liberals holding a majority of the seats in the House of Commons. There would be no way for the Conservatives to construct a majority through the support of other parties. Quickly now, when did this situation last appear in Canadian politics?  The answer is 1972.  In other words it has been over forty years since we have seen a situation at the national level where there is a potential for governance based on Liberal-NDP cooperation (this is what happened in the 1972-74 parliament). This could include either a coalition or an accord.

The chart below summarizes the numbers (they date back to June 2013) that form the trend line referred to above:
It is the percentage of seats won that matters. To help clarify I have expressed what it would mean both in the House as it will be in 2015 and what it would mean for today's House if it was still in place. A majority in the current 308 seat house is 155, in the 338 seat House it is 170.

The two parties do have some experience with working together. The NDP supported the Trudeau Liberals in exchange for various concessions in the 1972 to 1974 parliament. The Liberals and NDP attempted to form a coalition after the 2008 election, but needed the support of the Bloc to make it a possibility. The two parties did coalesce in Saskatchewan when the NDP failed to win a working majority in the 1999 election. The NDP was a very dominant partner in that arrangement. In Ontario the two parties negotiated an accord in 1985 that ousted the Ontario PCs from power after 43 years in office. The agreement permitted the Liberals to form a government for two years in exchange for implementing some NDP priorities.
Some such as former Harper advisor Tom Flanagan believe that the public would oppose such a coalition in 2015, as they did in 2008. This strikes me as mistaken. Given the growing antipathy to Harper on the centre and left it seems likely that the public attitude welcoming change that prevailed in Ontario 1985 is the more likely outcome. The public will judge a coalition based on the context, not on some preconceived notion of whether it is a good or bad idea.

If we look more closely at the messages the polls are sending about the state of politics in Canada right now in the various regions it helps us understand our current situation.

Atlantic Canada
In Atlantic Canada, the Liberals have taken a big lead.  Their average support in the polls is at 50%, an astronomically high number putting them 27 percentage points ahead of their nearest competitor, the Conservatives. Were this repeated elsewhere a majority would be in sight. Why are the Liberals so high?

Let me suggest a few factors. One is simply the fact that the Liberal brand in Atlantic Canada has been relatively strong over time. The current state of provincial politics also helps (there are good provincial poll numbers available in Atlantic Canada from Corporate Research Associates). There is a newly-elected Liberal government in Nova Scotia still enjoying its honeymoon. As Nova Scotia is about 40% of Atlantic Canada this alone has a big impact. In addition there are unpopular PC governments in both Newfoundland and Labrador and New Brunswick, both likely to lose the next provincial elections there to the Liberals. The incumbent Liberal government in PEI remains popular. In polling there is a spillover between the federal and provincial realms. Some respondents become confused about which level of government is being asked about. Given its strong brand this helps the Liberals in Atlantic Canada, and likely hurts the NDP, which still suffers from the unpopularity of the former Dexter government in Nova Scotia. Finally, it is clear that the Harper government is quite unpopular here.

Quebec dislikes the Harper government more than any other province or region so the Conservatives are not going to increase their representation but opposition is split among the Liberals, the NDP and the Bloc Québecois with the latter two parties doing disproportionately well among francophones, while the Liberals are the dominant force among Quebec anglophones and allophones. Liberal strength has grown here but they still trail among francophones.

Ontario is the biggest province and also where the federal Liberals have gained the least since 2011. The troubles of the provincial Liberals are a factor here. Ironically the situation in Ontario suggests one possible federal Liberal majority government scenario. For it to occur the Wynne Liberals would need to lose the expected spring election to the Hudak PCs. For maximum effect a PC majority would likely be necessary. My suspicion is that Tim Hudak would make the Conservative brand name quite toxic within a year in Ontario. Both the federal Liberals and NDP could benefit. If the federal Liberals could increase their Ontario support significantly at the expense of the Conservatives (with NDP gains held to a bare minimum) a majority might come in sight.

All this depends on a set of events that have not yet begun to take place. And Ontario alone would not be enough. The Liberals would still need to finish a strong first in Quebec and must do exceptionally well elsewhere including B.C. I cannot stress enough that this is an unlikely scenario, but one cannot completely rule it out. It is also possible but not likely that the NDP will find a way to get ahead of the Liberals. This could happen if the public acquires a stronger sense of Justin Trudeau's weaknesses.

The Prairie Provinces
The Liberals are poised to make significant gains in Manitoba in part because of unpopularity on the part of the NDP provincial government. Something close to the status quo prevails in Saskatchewan, although the NDP seem likely to gain a few seats here as a result of the redrawn boundaries.

Remarkably, the Conservatives are experiencing significant losses in Alberta. On average the party is down about 14 points since 2011. It is enough to give the Liberals a few seats if it holds through an election. I do think it is slightly overstated. The "Other" category is high in some surveys and I suspect that these are respondents who confuse federal with provincial politics, and volunteer the Wild Rose Party as their preference.

British Columbia
This is where the popularity of the Harper government has declined the most. It is the Liberals who have gained the most from it although their 13.4% share of the popular vote in B.C. in 2011 was quite low by historical standards. For example, their current poll average of 30.6% is close to the 28.6% share the Liberals achieved in B.C. in 2004. NDP support is down a little since 2011.

One possible liability for the Liberals here is a difficult term for the BC Liberal provincial government. Although it is a coalition of federal Liberals and Conservatives, the government is led by Christie Clark, a federal Liberal. Based on recent history it does seems likely that the B.C. government will get into political trouble between now and 2015, with the NDP the beneficiary.

TCNorris does not think that a Liberal majority in 2015 is a realistic expectation. Justin Trudeau and company should keep their minds open to other governing possibilities. Thomas Mulcair to his credit just gave us a preview of what conversation we should expect after the next election: what form should Liberal-NDP cooperation take.

Thursday, March 06, 2014

The enigma of Quebec politics

As we head once more into a Quebec electoral campaign I am reminded of when I went door to door with a federal Liberal candidate in the 1988 election in one of those mixed anglo/allo/francophone constituencies that dot the island of Montreal. We came upon a group of francophone men standing in an open garage facing the street and chatting on a gray rainy afternoon.

Their response to the vote-seeking politician was guarded and non-committal. The candidate observed as we walked away that he found that francophones were generally reluctant to give their political views. He explained it was because of the division within Quebec between those who believed in Canada and those who supported independence. A wide gulf between the two visions is found within francophone Quebec families and this often makes politics a sensitive topic best avoided.

I think this partly explains why Quebec has so frequently been the source of so many political surprises, including the defeat of the Lesage regime in 1966, the first victory of the Parti Québecois in 1976, and most recently the unexpectedly strong showing of the Quebec Liberals in 2012. It also accounts for why, after the divisions of the 1995 referendum, so many francophone Quebeckers were allergic to the prospect of another referendum campaign.

When the votes were tallied in the 2012 election the outcome wound up being closer than anticipated, generally because the polls underestimated Liberal support (in a few cases PQ support was greatly over-estimated), a phenomenon that has appeared many times in the past. The final election survey by CROP, while understating the Liberals, accurately anticipated that the PQ would capture 32% of the popular vote.

When CROP, which has long enjoyed a good reputation as a Quebec pollster, released a poll in February giving the PQ 40% with the Quebec Liberals trailing behind at 34% and the CAQ (Coalition Avenir Québec) a distant 16%, election fever began to warm up a cold Quebec winter.

When a political party forms a minority government, it is always looking for an opportunity to turn that into a majority. No doubt the PQ did some of their own polling to firm up their view that such an opportunity was at hand and has now called an election for April 7, 2014. However a poll released the same day the election was called by Léger Marketing gives the PQ just a two point lead over the Quebec Liberals. The PQ cannot be certain that it will win a majority. My own model that converts poll numbers into seats suggests the CROP February poll would have given the PQ a small majority, while the new Léger poll would leave them just one seat over a majority.

Here are my projections for the CROP February Poll and the just released Léger poll:

A key factor here is the CAQ vote, which has dropped over the past year and a half.  It is low enough now that the party would win just a handful of seats. This means the chances of either a PQ or Liberal majority are high unless the CAQ recovers substantial ground during the campaign. I estimate they need 23% to win at least 10 seats (they won 19 with 27% in 2012). Although they start off weak this time, the CAQ also started out in a similar position during the 2012 campaign averaging 21% in the opening surveys but improving their position significantly over the course of the campaign.

The two polls share one characteristic in common that should give rise to additional doubt about the what April 7th holds in store. This characteristic was not a feature of the CROP survey in 2012 that accurately predicted the PQ's 32% vote share. Most Quebec polling comes from Léger and CROP and in recent years they have relied extensively on online surveys, a methodology that I believe should still be regarded as experimental. As it turned out in 2012 so did CROP. If you read the methodological statement in its last election poll (at the bottom of the news story), one finds that CROP used traditional telephone polling methods for its final election poll. I hope we will see the firm do the same this year and that we will see more telephone polling generally.

Given that almost all the Quebec polling we have seen since the last election has been online and the results even there are close, there remains considerable uncertainty about what we will see come the evening of April 7.

Charter of Values and the Prospects for Quebec Independence
A key focus of debate will be the so-called proposed Quebec Charter of Values. While a majority say they support it, a Léger poll on the topic from January reports that it ranks fourth in a list of issues that respondents say will be the most important vote determinant in the next election, with health care and economic/fiscal issues ranking ahead of the Charter. It should be noted that this type of response in a poll can be misleading. For example, if the public is pessimistic and thinks that what will happen in health care or the economy won't be affected much regardless of who wins, a lesser consideration where the outcome will matter, such as the Charter of Values, could be more important.

A key finding for me was that among francophones in this survey only 30 per cent strongly favoured the Charter. Those who say they "somewhat" rather "strongly" support something are expressing uncertainty about how they feel. It is the "strongly"category that really matters when judging how the public feels about issues. Further, it is likely that those who strongly favour the Charter consist mainly of core PQ voters. It remains unclear how many of this group currently favour other parties and might therefore be tempted to switch by the Charter.

In the longer run the PQ would like to see the Charter overturned by the Supreme Court of Canada and thus hopefully, from their perspective, provide the political basis for a Quebec referendum on independence. Given that the Charter has seriously divided the pro-independence movement in Quebec, it seems highly unlikely that this could work even if the PQ achieved its hoped-for scenario. For example, former PQ Premier Jacques Parizeau is among those who have denounced the Charter of Values.

Support overall for independence is not only weak, running far below 50%, when you look at the demographics of support for independence (on page 9 of this January 2014 Léger poll) by age, it is clear that it is concentrated disproportionately among the baby boom, roughly those aged 45-65. Support is weaker among both younger voters and older voters.

The polls are close, the political significance of the Charter of Values is still unexplored territory, and support for sovereignty is weak. Quebec strikes me as in a transitional phase of its history and has not yet found a clear sense of itself going forward. It remains an enigma.