Analyzing a Sea Change in Canada

Even for observers who were the most optimistic about the New Democratic Party’s chances, the results of Monday’s general election in Canada were nothing short of astounding.

Here on FiveThirtyEight, I argued that a slew of factors could hold the New Democrats back, including the inherent obstacles for national third parties in first-past-the-post systems, the difficulty of making electoral headway in a province like Quebec where separatist sentiment is strong, and the fact that rapid gains in polls late in the electoral game often overstate the level of support a party will get in the voting booth (for reasons of soft support and possible selection bias in polling).

But the New Democrats succeeded in proving the skeptics (including me) quite wrong, by winning more than 30 percent of the popular vote nationally and more than 100 seats in the 308-member Canadian House of Commons. The success means that the party vaults ahead of the Liberals to become the primary opposition to the Conservatives, who won nearly 40 percent of the national popular vote and at least 167 seats, enough for a majority.

Yesterday’s post laid out a series of scenarios for how the New Democrats might fare. Based on a series of challenges that they would need to overcome, particularly in the province of Quebec, I doubted that they would be able to live up to their polling figures or that their seat-to-vote ratio would be able to match those of the other three major parties, all of them greater than 1.0 for the last several elections.

In the end, though, the New Democrats not only matched my best-case scenario for them, by attracting votes to equal their polling average of 31 percent, but they also won an extra seven seats, with a seat-to-vote ratio of slightly higher than 1.1, the average for the Liberals and Conservatives in the three elections from 2004 to 2008.
How did they achieve this?

To begin with, the party performed incredibly well — let’s be honest, historically well — in Quebec.

Even the ordinarily very efficient Bloc Québécois, the provincial nationalist party, fell victim to the essential “orange-wash” of the province (orange is the New Democrats’ signature color). The N.D.P. overperformed its polling average in Quebec by two points, and with a remarkable seat ratio of 1.8, managed to turn 42.9 percent of the popular vote in Quebec into victory in nearly 80 percent of Quebec’s “ridings,” as legislative districts in Canada are known.

Several factors contributed to the powerful performance. First, the majority of the races that the New Democrats won were three-way contests with the Bloc Québécois and the Liberals or Conservatives. There were even some that could be characterized as 4-way races, with the Bloc and both of the other national parties. Because the Democrats’ gains in support were distributed relatively evenly across the province, they were able to capture these seats (usually from the Bloc) with 30 to 45 percent of the vote, winning convincingly over split opposition.

Second, the party overperformed even its high expectations in the popular vote in Quebec, while the Bloc and the Liberals fell short. As mentioned earlier, the New Democrats did 2 points better than the polling average predicted — though if we exclude the online pollster Angus Reid (we learned in the British election that online polls tend to oversample younger voters), the New Democrats overperformed by 4 percentage points.

Cleaning up in Quebec, though, would not have been enough by itself to propel the New Democrats up to second place nationally. That required strong performances in two other important provinces.

One was British Columbia, where the so-called second wave of polling strength for the party was seen. In that province, the New Democrats actually underperformed the final polls by 4 points, indicating that their support was softer there than in Quebec:

But the N.D.P. countered that erosion with a high level of efficiency: the party focused its resources on the ridings where it had a chance to win, and did so with great success, again typically where the vote was split several ways (especially outside Vancouver and on Vancouver Island, where the New Democrats won several close contests). In the end, they achieved a 1.0 seat-to-vote ratio in the province, while the Liberals and Greens were left behind with piles of futile votes.

Finally, in Ontario, Canada’s largest province, though the Conservatives once again dominated the landscape, the contrast in results between the New Democrats and Liberals is a telling case study.

The New Democrats and Liberals won practically the same number of total votes, but the New Democrats won twice as many seats as the Liberals in the province.

Overall, the same tendency that has cost the party seats in past elections — votes spread evenly over large areas rather than being bunched in high concentrations — yielded a triumphant result in this election. The New Democrats’ gains in support were enough to give it victory after victory in ridings where the vote was split several ways, overpowering the Liberals and Greens in every province and dominating the Bloc Québécois in Quebec.

While Canada remains subject to the vagaries of the single-member, first-past-the-post system, which tends to hinder national third-party representation, the voters in 2011 showed a strong willingness to shake up the equation. At the national level, Canadian politics will still be a two-party game, but thanks to the decline of the Liberals and the collapse of the Bloc Québécois, it will now be a different pair of parties.