The Liberals inch closer to majority territory: 338Canada

Philippe J. Fournier: The latest 338Canada projection puts the Liberals as 50-to-one favourites to win the most seats and the Conservatives failing to turn the tide post-convention

Even though several parts of the country are entering a third wave of the pandemic with abrupt rises in infection rates, there is reason for medium-term optimism, as most Canadians should be fully vaccinated by late summer if things go as planned. (Quebec Premier François Legault even went as far as publicly predicting that all Quebecers willing to receive the vaccine will get their first dose by June 24.)

Hence, there could be a window for federal Liberals by summertime to try what three provincial legislatures (British Columbia, New Brunswick, and Newfoundland and Labrador) have managed since last summer: upgrade from a minority to a majority.

The Conservative Party’s national convention has come and gone, and we now have post-convention polling numbers from several firms to analyze. The least we can say is Erin O’Toole’s party did not benefit from this heightened media exposure, which, rightly or not, centred around the rejected motion stating that climate was real and that the CPC should act upon it. Although the movement in voting intentions has been modest in the aggregate, other indicators such as low leader favourability for O’Toole and still reasonably high government approval (58 per cent according to Léger) could signal difficult times ahead for Erin O’Toole.

Here is a short recap from the past two weeks:

  • The latest Léger/Canadian Press bi-weekly update measured the Liberals up seven points on the CPC, 35 to 28 per cent. Looking at Léger’s regional breakdown, we notice the CPC slide below the 30 per cent mark appears to due to the abnormally low support in Alberta with only 41 per cent. While other polling firms have also measured CPC support below its 2019 results in Alberta, could it be that the Conservatives have lost close to 30 points in the province since then? (See my column on this topic here). Additionally, we note that Léger’s latest survey has Jagmeet Singh’s NDP at 22 per cent nationally, a number somewhat higher than the party’s current average of 18 per cent. Léger has been bullish on the NDP for the better part of the last year, regularly measuring its support at or above the 20 per cent mark. Obviously, we do not mean to insinuate that Léger’s NDP numbers are wrong, but they are definitely worth pointing out: Whether the NDP stands at 15 per cent or above 20 per cent could have a huge consequence on every party’s standings in terms of seats. Here is Léger’s latest report.
  • Abacus Data had been in the field in mid-March and measured a steady but modest 4-point lead for the Liberals. Post CPC-convention, Abacus Data now has the Liberals jumping to an 8-point lead, 38 to 30 per cent, including a 7-point advantage over the Bloc Québécois in Quebec, and a massive 16-point lead over the CPC in Ontario. With a virtual three-way-tie in British Columbia, Abacus’s numbers would most likely translate into a decisive LPC majority. Abacus also has Erin O’Toole’s personal numbers in a downward spiral: O’Toole’s net impression (positive impression minus negative impression) sank to a minus-14 nationally, including an 18-point drop in Ontario and 17-point slide in British Columbia since September 2020 (shortly after he won the CPC leadership). Although the Prime Minister may not break any records with his own net impression (Trudeau currently stands at minus-1 according to Abacus), his main opponent’s numbers have been significantly worse. See Abacus’s full report here.
  • Mainstreet Research‘s latest federal numbers for iPolitics had the Liberals at 37 per cent, a three-point lead over the Conservatives, indicating a notably tighter race than other pollsters of late. However, Mainstreet also had the NDP at a dismal 10 per cent nationally. While this specific data may be a bit of an outlier, we estimate that such a poor result for the NDP would, hypothetically, help the Liberals win a few close races, especially in Ontario and British Columbia, probably resulting in a strong plurality for the Liberals. Find Mainstreet report here.
  • Finally, Nanos Research (whose paywalled numbers may be found here) has the Liberals leading the Conservatives by double digits, comfortably placing the LPC in majority territory. As for the NDP, Nanos measures a level similar to the party’s 2019 results, thus roughly putting the NDP half-way between Mainstreet’s and Léger’s numbers.

We add these latest polls to the 338Canada model and present today this updated federal projection (you will find all federal polls listed on this page). Here are the weighted averages of the main five parties: The Liberal Party leads the fold with 37 per cent support nationally, three points higher than the party’s 2019 results:

The Conservatives take second place with an average of 30 per cent, four points below their 2019 score. The New Democrats currently sit at 18 per cent, but, as mentioned above, there is significant disagreement among Canadian pollsters on where the NDP stands, which explains the projection’s wide confidence intervals (a bell-curve stretching from 14 to 21 per cent, 19 times out of 20). The NDP projection is uncertain specifically because the party’s numbers have been all over the place of late.

Finally, the Bloc Québécois is holding its own with 7 per cent nationally (29 per cent in Quebec on average), marginally below its 2019 results. The Greens remain mostly steady with an average of 6 per cent.

How do these levels of support translate into seats? The Liberals have slowly creeped back up above the majority threshold (of 170 seats), but barely: the LPC wins an average of 172 seats, including 40 in Quebec, 80 in Ontario, 27 in Atlantic Canada, and 15 in British Columbia (all averages).

The Conservatives win an average of 108 seats, including 31 in Ontario, 30 in Alberta, 19 in Manitoba and Saskatchewan, and 13 in British Columbia.

For the Conservatives to win an election with these numbers, they would have to win all close races (toss up and leaning districts), and hope for a favourable vote-splitting between the Liberals and NDP. Oh, and the polls would obviously have to underestimate the CPC considerably, which is possible, but it has not occurred in the previous two federal elections.

Here are the seat projection probability densities for the Liberals and Conservatives. Notice that, even considering these projections’ high uncertainty, both curves now barely overlap. In short, if an election were held this week and these were the last available numbers, the Liberals would be 50-to-1 favourites to win the most seats.

Here is the probability density of the seat difference between the Liberals and Conservatives. Red columns, the LPC win the most seats; Blue column, the CPC does. As you can see, the numbers are heavily tilted towards the Liberals:

While it is not hard to imagine Liberal strategists rejoicing at these numbers, the current state of the pandemic in Canada renders the idea of a spring federal election unfathomable. L’actualité political bureau chief Alec Castonguay reported last week the Liberals were now eyeing a September election in the hopes that, by then, a vast majority of Canadians will have received at least one if not two doses of the COVID-19 vaccines. The vaccine rollout has been picking up the pace in March, albeit unevenly across the country, and should continue to do so in April.

Growing optimism among fully vaccinated voters throughout the country should, theoretically of course, improve the Liberals’ electoral fortunes, especially if Erin O’Toole is unable to turn the tide on his public perception.

Follow 338Canada on Twitter

* * *

Details of this projection are available on the 338Canada page. To find your home district, use the list of all 338 electoral districts here, or use the regional links below:

Looking for more?

Get the Best of Maclean's sent straight to your inbox. Sign up for news, commentary and analysis.