Jagmeet Singh vs. Maurice Duverger
Jagmeet Singh mishandled the CASA, but widespread public confusion about it signals deeper challenges for Canada’s third parties.

Jagmeet Singh played the confidence-and-supply agreement (CASA) all wrong, but Canadians didn't understand it, either. That poses longer-term problems for Canada's third parties — the NDP, the Bloc Quebecois, the Green Party of Canada, etc. With his resignation last night, it is worth evaluating both Singh’s leadership and his significance in Canadian political history.
Though the party was decimated in Monday’s elections, winning single-digit seats and not making official party status, the NDP helped pass some of the last parliament’s most popular policies. It might also surprise future historians just how much the NDP featured in the political discourse of the late Trudeau years. Singh’s failings have been described at length. But he is also a tragic figure, as his downfall is part of a stunning reversal in Canadian politics. After decades of fragmentation of the political sphere, with more parties taking up more airtime, Canada has returned to true two-party dominance.
Over the last decade, there was movement away from the big-two parties: They won a combined seventy per cent of the vote in 2015, and then sixty-seven per cent in 2019 and 2021. Far from what we see in the United States, Canada’s major parties barely reached two-thirds of the popular vote between them. Last night, they won closer to eighty-five per cent. This reversal is also visible in the strength of the smallest parties: In 2019, there were six leaders on the debate stage, and in this election there were four.
This seems like evidence of French political scientist Maurice Duverger’s proposition that single-winner electoral systems structurally incentivize two-party systems. Over time, fear of the spoiler effect (the “vote-splitting” that preoccupies Canadians) will lead to a national two-party system. Stephen Harper took a Duvergerian approach by uniting the right in 2003 — and treating the Liberal Party as an outdated rival to be destroyed, nearly succeeding by 2011. Justin Trudeau might have ended this discussion altogether by keeping his promise to move away from first-past-the-post.
Yet Canada has been in flagrant violation of Duverger's Law for decades. Third parties are well-established and not going away. The NDP never permanently replaced the Liberals, nor have the Liberals ever been able to get rid of the thorn in their left side. Political scientists like Daniel Bochsler have adapted Duverger’s Law to Canada by arguing that regional diversity creates multiple localized two-party systems: While in most ridings the race is between the Liberals and the Conservatives, that is not the case in Edmonton, Sherbrooke, or even Saanich. Third parties establish themselves with a regional base, and then justify their existence by extracting concessions from a major party.
This was a unique election. Foreign policy and Canadian sovereignty were top of mind, disempowering third parties from appealing to their regional bases. Ironically, anti-Americanism made our party system more American. In future elections, in which our sovereignty is not threatened, third parties might have a chance to make a comeback. Duverger’s Law is too teleological, but it is also undeniable that Canadians voters fear spoilers and love bandwagons, especially in times of crisis.
This election supports Duverger’s Law not just because the major parties’ vote share, but because voters rejected third parties’ rationale for existing. If the function of the NDP in a single-winner system is to win enough seats to influence a minority government, then they executed their role successfully through the CASA. The actual outcomes of that agreement are popular and aligned with the party’s social-democratic principles. Yet they were punished by voters for their participation in the agreement at every turn.
Early on, the CASA was described as “government by blackmail” by conservatives. Here they echoed Stephen Harper’s critique of a planned coalition between the Liberals, NDP, and Bloc Quebecois in 2008. Inter-party collaboration, negotiated as it is in backrooms between party leaders, is anti-democratic in their eyes. Ironically, the Bloc too critiqued the CASA as a “false majority.” Nevermind that the combined popular vote of the CASA participants actually surpassed 50 per cent, unlike if the Liberals governed on their own.
As opportunistic as this rhetoric was from Poilievre and Harper, it also resonated with large swaths of the Canadian electorate. Many reacted with incredulity to the idea of inter-party collaboration: Every time Singh criticized a government decision, he was asked why he was continuing to “prop them up.” The straightforward answer — that not every complaint was worth bringing down the government, and that he could criticize the government while still using his influence to extract concessions from them — baffled people. In other Westminster democracies like New Zealand, these concepts are understood. Most embarrassingly for Canada, journalists tempted by a “gotcha” against the already unpopular NDP leader also repeatedly asked the “propping up” question, pretending not to understand how leverage works.
Singh played a role in creating this dynamic by never being explicit with his intentions. He squirmed when the CASA came under scrutiny, which made his critics salivate. In the 2019 leaders’ debates, Elizabeth May stated explicitly that her goal was to lead a small caucus of Green MPs in extracting concessions from a minority Liberal government (she took a swipe at Andrew Scheer along the way). Her candor and the straightforward argument for electing more Greens were rewarded with the best ever election results for her party.
On the other hand, Singh was also the subject of disdain when he expressed doubts about triggering an election because of the potential of a Conservative victory. This ire came from NDP partisans as much as anyone else, who felt that the NDP should be attempting to form government themselves, not paying attention to other parties’ polls. At this point it was already the media consensus that Singh was unserious, and so admitting that he was not campaigning to be prime minister was made out to be proof of that.
But it wasn’t unreasonable for Singh to prefer a Liberal minority over a Conservative majority. Going to an election in the fall likely would have meant the end of the dental and pharmacare programs. In multi-party systems, small parties must pay attention to their potential coalition partners’ polling numbers. In Canada, admitting this was electoral poison for the NDP — tantamount to revealing that it was merely an accessory to the Liberal Party.
The NDP leader was stuck between a rock and a hard place. Ultimately, Singh’s approach was a ham-fisted middle ground between May’s 2019 approach and campaigning to be prime minister, essentially: You’ve seen what we can accomplish with a small caucus, now give us a majority government. He managed to square the circle of simultaneously appearing unambitious and delusional.
Jagmeet Singh’s failures are undeniably his own. But this defeat (which they share with the Bloc and the Greens) also revealed Canadians’ Duvergerien impulses. Inter-party collaboration is not politically rewarding for the junior partner. But if every party’s goal is to form a majority government, then three parties are clearly too many, much less four or five. In such a system, what role is there for third parties? Critics of first-past-the-post, and supporters of smaller parties, should not underestimate the importance of these questions by putting all of the blame on Singh.
Disclaimer: Gabriel Blanc was previously the Ontario Representative on the Young Greens of Canada Council in the Green Party of Canada. All views expressed in this editorial are his alone.
Gabriel Blanc (he/him) is the Editor-in-Chief of The Bell from Toronto, Ontario. He wrote for the Brown Political Review as a staff writer while pursuing his undergraduate degree at Brown University. He has since written for several online publications, including his own blog. While working at UNICEF, he contributed to human interest stories in English and French.
This is the logic of first past the post. An entirely different logic emerges under proportional representation