Summer polling can be quirky sometimes. I remember the summer of 2019, mere months after the SNC-Lavalin saga had rocked the federal Liberals, a string of polls showed Andrew Scheer’s CPC had taken the lead in popular support. Scheer was even featured on the cover of L’actualité magazine, along with the ominous title “The next Prime Minister?”
To be fair, Scheer’s lead in the polls had been propelled by the first scandal to put a dent in Liberal support since Trudeau’s first victory in 2015, not necessarily by the quirkiness of summer polling. Heading into and throughout the summer of 2019, Scheer appeared to be the favourite to win the fall election and become Prime Minister.
To wit: A Léger poll fielded in June 2019 measured the CPC leading in popular support with 38%, against 29% for Justin Trudeau’s Liberals — including a stunning ten-point lead in Ontario (39% CPC, 29% LPC).
Sounds familiar?
As the fall 2019 campaign was about to get under way, this CPC lead would gradually fade, and both Liberals and Conservatives would remain neck-and-neck for the duration of the campaign. When all the votes were counted on October 21, 2019, the Liberal had secured a victory in terms of seats, but had lost the popular vote to Scheer’s CPC by one point (34% CPC against 33% LPC).
Then came the pandemic, the lockdowns, the vaccine sagas, the omicron variant.
Fast-forward to June 2021. A string of polls from top-rated firms put the Liberals squarely in majority territory:
EKOS: LPC 35%, CPC 26%.
Abacus Data: LPC 37%, CPC 27%.
Ipsos: LPC 38%, CPC 26%.
Nanos Research: LPC 37%, CPC 25%.
Mainstreet Research: LPC 38%, CPC 27%.
A third term, but most importantly a second majority, was within reach for Trudeau’s Liberals.
Later that August, undoubtedly encouraged by these numbers, the Prime Minister requested dissolution of Parliament. From day one in mid-August to early September, the Liberals fell in the polls and, before the televised debates took place, it looked like Trudeau’s prime ministership was coming to end.
The Liberals nevertheless turned it around to avoid defeat. The September 2021 election ended up with eerily similar results to the 2019 election: Among the 338 electoral districts, only 22 changed colours. All parties finished a few seats away from their 2019 seat count.
Summer polls, once again, had fooled us (and them).
Oh, this does not suggest that these summer polls were wrong. Not in the slightest. But the country’s mood changed during the campaign once voters started paying attention. It bears repeating: Polls do not predict the future, they describe the very recent past.
So, what are we to make of the latest summer polls?
We won’t know until later this fall at the earliest, but the numbers released by pollsters in recent weeks indicate that this time it could be different. Only 26% of Canadians believe the country is going in the right direction according to Abacus Data’s latest poll, the lowest point in years, and government approval stands at a measly -23 rating (31% approve, 54 disapprove), also an all time low for this government.
With polls measuring the CPC ahead by margins of 6-10 points from coast to coast in July and August (see the complete list of federal polls here), the latest 338Canada federal projection (updated August 13, 2023) puts the Conservatives right at the cusp of a majority with an average of 169 seats.
The Liberals are now only projected at 111 seats - their lowest average since I began tracking seat projections in 2017.
Two regions have shown significant movement in the latest months: Atlantic Canada and Ontario.
In the last three federal elections, the Liberals won 32, 26 and 24 seats (out of 32) from the Atlantic provinces, giving the LPC a convenient cushion of seats akin to the CPC’s in Alberta. However, unlike Alberta — which remains squarely in the CPC fold — recent polls from Abacus Data, Léger and Ipsos all indicated that the Conservatives had closed the gap in Atlantic Canada.
However, a word of caution: As you can see on the chart above, Atlantic Canada numbers are quite noisy, since this regional sub-sample is generally limited to barely a hundred respondents or fewer in most polls —leading to a massive amount of uncertainty (all four Atlantic province account to about 6% of Canada’s population). Hence, we will need more time and data to confirm whether this is a real trend.
The numbers in Ontario have also shown improvement for the Conservatives over the summer. Unlike Atlantic Canada, Ontario polling samples generally range from 300 to 600 respondents, leading to much less uncertainty in the data. Nanos Research (data paywalled), Abacus Data and Léger all gave the CPC leads ranging from 4 to 10 points over the summer. Ipsos had both parties tied at 38% in the province.
In both Ontario and Atlantic Canada, the Conservatives would record significant seats gains in a general election should those numbers hold.
…but there is no election in sight now, is there?
As stated above, these projections are not a prediction for the next general election, whenever it may be held (although October 2025 appears increasingly likely). Nevertheless, they show that this is the deepest hole the Trudeau Liberals have found themselves in since taking power in 2015, and one wonders whether they can turn it around once more.
Summer polls do not predict the future — not even the near future, as we saw in 2019 and 2021 — but all indicators point to an electorate in an increasingly foul mood towards its government. At some point, magic runs out for any governing party.
The next two years will feel like the longest of Pierre Poilievre’s life. To be continued.
Visit 338Canada.com for details of this federal projection. See all projections, provincial and federal, by visiting this page.
Musique à mes oreilles. Merci pour les bonnes nouvelles !
"The Liberals are now only projected at 111 seats - their lowest average since I began tracking seat projections in 2017."