Why the Liberals are cheering for a weak but enduring NDP—and why the CPC should be paying attention

0:00Welcome to Hubheadlines. Today's program features the weekly rap by the hub's editor at large, Sean Spear, who analyzes the big stories shaping politics, policy, and the economy in the week that was. Today, he discusses why the liberals are cheering for a week but enduring NDP, and why the CPC should be paying attention. Last weekend's NDP convention, with its viral clips of equity card arguments going global, looked out of step with the median Canadian voter. This has led to a lot of commentary about how a weaker NDP benefits the liberals by reducing vote splitting on the center left. But it's also true that a week yet enduring NDP benefits the liberals. It can absorb voters whose preferences are furthest from the median, thereby permitting the liberals to maintain a more electrically efficient position. This points to a broader question about what one might call an optimized opposition.

1:05In a first past the post system, the electoral success of a major party doesn't just depend on its own positioning, but also on how the wider political ecosystem organizes and distributes voters across parties. The interrelationship between the federal NDP and the liberal party offers a case study. For decades, the NDP has served as an institutional home for voters, activists, and ideas that sit to the left of the Liberal Party's electoral center of gravity. That arrangement has often enabled the liberals to operate as a more electrically efficient center left party, able to reach toward the median voter without fully internalizing the preferences of their coalition's most ideologically committed elements. The NDP, in this sense, doesn't merely compete with the liberals. It also performs a sorting function within the broader center left ecosystem. The NDP conventions perceive distance from mainstream opinion underscores this role. It's precisely because the party can accommodate positions that are weird or unrepresentative

2:07that it reduces the pressure on the liberals to do so. The existence of a separate vessel for those views allows the liberals to maintain a broader and more electrically viable coalition. The relationship isn't static or always beneficial. There are periods when vote splitting on the left clearly costs seats, but the point is that fragmentation has both costs and benefits. It's a trade-off rather than a pathology. The more interesting question is whether a similar logic applies to the right. The standard view is that fragmentation among conservative voters is inherently self-defeating. The experience of the Reform PC split in the 1990s looms large in this regard. But the contemporary case is different. The People's Party of Canada is a marginal actor rather than a co-equal competitor for government. At low levels of support, a flank party like the PPC can impose fewer direct electoral costs while still delivering some of the coalition sorting benefits observed on the left. A useful way to think about this is in terms

3:11of two offsetting effects. The first is a vote-splitting cost. In a first pass the post system, a party that loses even a small share of votes in closely contested writings can forfeit seats. This cost rises sharply with the flank parties vote share because electoral margins are often thin. Once a rival party's support reaches the low-to-mid single digits, it can exceed the margin of victory in a meaningful number of writings, particularly in suburban battlegrounds, such as those in the greater Toronto area. The second is a coalition management benefit. A major party must assemble a coalition that's broad enough to win elections, but coherent enough to campaign effectively. When that coalition includes voters whose preferences are far from the median, it can impose constraints on messaging, candidate selection, and policy emphasis. It may require a party to adopt rhetoric that's less appealing to swing voters. Elevate candidates who reflect internal coalition pressures

4:12rather than broader electoral considerations or prioritize issues that are salient within the coalition but less resonant with the median voter. A small flank party can relieve some of that pressure by attracting those voters, allowing the main party to present a more moderate and electrically appealing profile. The interaction of these two effects implies that the relationship between flank party support and major party performance is non-linear. At low levels of support, the coalition management benefit may dominate. A small amount of vote leakage reduces internal heterogeneity without materially affecting seat outcomes. At higher levels of support, the vote splitting cost becomes decisive. The flank party begins to flip seats. The optimal level of fragmentation, if one exists, lies somewhere in between. It's therefore useful to think in terms of ranges rather than a single point estimate. When flank party support rises into the 3-5% range, the risk of seat loss becomes significant

5:13because it often exceeds the margin of victory in competitive writings. By contrast, when support falls below roughly 1%, the direct electoral cost largely disappears, but so too may any coalition management benefit. The empirical question, then, is whether this intuition is born out in recent elections. Evidence from recent federal contests suggests that a PPC vote share in the range of 3-5% is likely too high. In 2021, the PPC's support frequently exceeded the conservative margin of defeat in competitive suburban writings, particularly in the greater Toronto area, making it difficult to argue that its presence was anything other than electrally costly. At those levels, the vote splitting effect is no longer marginal and instead becomes determinative. But it doesn't follow that the optimal level of PPC support is zero. The 2025 election provides a useful counterpoint. With PPC support collapsing to well below 1%,

6:16the direct electoral cost of right-wing fragmentation largely disappeared. Yet in many suburban writings, the conservative still fell short. The binding constraint in those contests wasn't vote splitting on the right, but insufficient support among median voters, as reflected in persistent conservative underperformance in suburban writings. This raises the possibility, admittedly difficult to quantify, that fully reabsorbing the right flank may carry its own costs, particularly if it requires rhetorical or policy adjustments that reduce appeal in key swing constituencies. This is where the analogy with the NDP becomes instructive, albeit imperfect. Just as the NDP can absorb some of the center-left's least median-friendly preferences, a small right-wing party may perform a similar function for the conservatives. The crucial difference is scale. The NDP is large enough that its vote share can be electrally consequential in its own right, whereas the PPC is only plausibly useful to

7:16the conservatives when it remains small. Putting these pieces together, the most defensible conclusion is that there may be a low non-zero level of right flank support that's strategically tolerable and perhaps even modestly beneficial. The available evidence suggests that the most plausible equilibrium lies in the low single digits, where some degree of voter sorting can occur without materially affecting seed outcomes, closer to 1 or 2% than to 5. Below that threshold, the coalition management benefits may outweigh the electoral costs. Above it, the calculus quickly reverses. None of this should be overstated. The uncertainties are significant, and the counterfactuals are inherently difficult to observe. But the broader point stands, fragmentation isn't an unqualified liability. In a system like Canada's where elections are decided riding by the effects of fragmentation depend critically on its magnitude, its geographic distribution,

8:18and the substitutability of voters across parties. The NDP's convention may have drawn attention for its ideological distance from the mainstream, but it also highlights a useful political reality. By absorbing a coalition's least-median-friendly voters, a smaller party can make a larger one more electrically effective. That's it for today's edition of Hub Headlines. We hope you enjoyed the program. Hub Headlines is produced by Alicia Rao. This program was narrated by automated voices. Thanks for listening.