Weekly Writ for June 26: Another canary in the Liberal coalmine

What happened in Toronto–St. Paul's and what it means.

Welcome to the Weekly Writ, a round-up of the latest federal and provincial polls, election news and political history that lands in your inbox every Wednesday morning.

Well, that escalated slowly.

Yes, it wasn’t until just before 4 AM early Tuesday morning that the Conservatives’ Don Stewart finally edged ahead of Leslie Church and the Liberals in the Toronto–St. Paul’s byelection. When Elections Canada finally finished the count — the ballot had 84 candidates on it, making for an interminably long counting process — Stewart had won by 1.6 percentage points, taking 42.1% of the vote and winning a seat that hasn’t voted blue since 1988.

It’s a historic win for the Conservatives and a big loss for the Liberals, who have only once before since coming to power lost one of their seats. That was in Chicoutimi–Le Fjord in 2018, and that wasn’t a safe seat. Toronto–St. Paul’s, Liberal since 1993 and won by Carolyn Bennett in 2021 by a margin of 23.9 points, was supposed to be a safe seat.

Turns out, there might not be many safe Liberal seats left.

Stewart’s share of the vote jumped 16.8 points from the Conservatives’ performance here in the last election, pushing the party above the 40% threshold for the first time since the 1980s. The Liberals fell 8.7 points to 40.5%, a total swing of 25.5 points between the two parties. There are 55 other seats in Ontario in which the Liberals’ margin of victory over the Conservatives in 2021 was less than 25.5 points.

This means all those seats would be at risk of flipping blue if the Toronto–St. Paul’s swing was replicated across the province. That includes 11 of 12 seats in Brampton and Mississauga, all of Oakville, Burlington and Milton, every seat around Kitchener-Waterloo and nearly all of Ottawa. All that would remain for the Liberals would be the downtown cores of Ottawa and Toronto, most of Scarborough and a few holdouts scattered about the rest of the province. Seat projections generally put the Conservatives in the 210-seat range and, notably, Toronto–St. Paul’s is usually not one of them.

Putting these results into historical context shows how significant they are. This is the Liberals’ lowest share of the vote in this riding (and its predecessors) since 1984, when Brian Mulroney won the largest majority government in Canadian history. With the sole exception of the 2011 election, when the Conservatives took 32.4% of the vote, that party hasn’t done better than 27% since the merger of the Progressive Conservatives and Canadian Alliance in 2003.

The New Democrats’ had a rough night, dropping nearly six points and falling to just 10.9% of the vote. That is the party’s lowest share here since the 2000 election. But the party’s 5.9-point drop is identical to the NDP’s average drop in byelections held since the last federal election, which suggests this is more of a typical byelection under-performance for the NDP than any particular sign of malaise.

If we compare the results to the benchmarks I set on Monday, which are based on where the provincewide polling would situate each of the parties, we do see a significant under-performance for the NDP and an over-performance for the Conservatives.

It might be tempting to see in this a move from the NDP to the Conservatives, but the reality is that many NDP voters stay home in byelections in which their party is not a real contender. All of the reporting that had been done in Toronto–St. Paul’s highlighted former Liberals swinging over to the Conservatives, and that seems to be the more plausible explanation considering the wealthier, urban profile of the riding.

The Liberals met their polling benchmarks, and were even on the higher end of them. But that’s not good news for the Liberals, as their current polling benchmarks have them losing nearly 100 seats. Church needed to beat the polls by a wider margin to both win the riding and give hope to other Liberals that things might not be as bad as they seem.

That Stewart beat the Conservatives’ polling benchmarks suggests that Pierre Poilievre’s support in the polls is no mirage — the results of real voters casting real ballots shows the polls are the real deal. It’s possible, though, that the specific circumstances of this contest might have boosted Conservative fortunes. Toronto–St. Paul’s has one of the highest concentration of Jewish voters in the country, and while a relatively small proportion of the riding as a whole they might have been a more motivated cohort considering the rise of anti-Semitism in the country and criticisms from the Jewish community of the perceived inaction of the Trudeau government in response.

Any dismissal of the results because it’s “just a byelection” would be unwise, as the results corroborate all the other signals indicating that Justin Trudeau and his party are in serious trouble. One dead canary in a coalmine might be a coincidence, but as the dead-canary count piles up it becomes foolish, reckless and dangerous not to get the heck out of the mine.

Okay, but what if it was just a summer byelection? Not so fast — turnout was 43.5%. That’s anemic in a general election but quite respectable in a byelection. Turnout has averaged around 34% in federal byelections held since 2021. What’s more, Stewart actually received 1,968 more votes than his Conservative predecessor in the last general election. The Liberals lost 11,464 votes.

Undoubtedly, many Liberals stayed home. But the Conservatives got new voters to support them and received more individual votes in this lower-turnout byelection than they have in any higher-turnout general election since 2011.

Lastly, there is the oft-repeated excuse that byelections generally go badly for incumbent governments. That’s not entirely true. In nearly 500 byelections held since 1867 (for which data is available), the governing party has lost an average of just 2.9 points. That doesn’t suggest byelections are routinely punishing sessions for governments. It is true that governments tend to lose support in byelections more often than gain it (in 57% of byelections the government’s vote share has gone down, in 43% it has gone up), but it is far from an ironclad rule of Canadian politics.

So, what does this mean for Prime Minister Justin Trudeau’s future? One byelection defeat does not a resignation make, but the results in Toronto–St. Paul’s add to the growing volume of what is a loud warning for the Liberals: something needs to change. If Trudeau decides he wants to walk past another canary and stay on as leader, his caucus might need to start asking themselves whether they want to keep digging or find their own way out of the coalmine.

Thanks to everyone who joined me and Philippe J. Fournier for our byelection results livestream on Monday night. It was a marathon, and even though it lasted more than five hours we still couldn’t make a call when we finally went to bed. I’m floored that total viewership of the livestream is now over 130,000 — and counting!

Now, to what is in this week’s instalment of the Weekly Writ:

• News on Naheed Nenshi’s landslide, Denis Coderre makes it official and two sitting MPs won’t be running again.

• Polls show the Conservatives continuing to enjoy healthy leads, while B.C. United crumbles in British Columbia.

• Majorities for Pierre Poilievre and David Eby if the elections were held today.

• Riding profiles of the Conservative leader’s tour through Quebec.

• The 1917 election in Saskatchewan for the #EveryElectionProject.

IN THE NEWS

Naheed Nenshi wins Alberta NDP leadership in landslide

It wasn’t even close — so much so that even the jokes that had been made about this race being over months ago under-stated the scale of Naheed Nenshi’s victory in the Alberta NDP leadership race.