Traditionally, the Canadian system was one of the least polarized in the world. For more than a century, it was dominated by a Liberal party that was unabashedly centrist. The secret of Liberal success was Quebec. From the late 19th to the late 20th century, that province was the pivot for government. It typically delivered almost all of its seats to the Liberals, such that Quebec alone put that party half way to a majority.
The Liberals’ chief rival was a conservative party that for most of the century carried the usefully contradictory moniker of Progressive Conservative (PC). For the PCs to capture a majority, Quebec was no less critical. So although the PC candidates outside Quebec became seriously conservative in the 1970s and 1980s, strategic imperatives moderated the party’s platform. Moderation of the party’s position applied to the existential question of Quebec’s role within Canada, but also to bread-and-butter issues, on which Quebeckers stood rather to the left of other Canadians. (Suitably stylized, this might sound familiar to Americans. The marriage of opposites that was the old Democratic party was arguably critical to the lack of polarization in the US system before the 1970s.)
But this meant that when the PCs won, their coalition was radically incoherent, including francophones and francophobes, social and economic conservatives and technocrats from North America’s most interventionist jurisdiction. The party ruled on this basis from 1984 to 1993, but in the latter year the coalition exploded. Failure to deliver on constitutional promises led key Quebec Members of Parliament (MPs) to form the Bloc Quebecois. For the first time, Quebec voters were offered a secessionist choice in a federal election, and the new party won a majority of the province’s seats. In the four western provinces, the Reform party broke through on the right flank. The breakthrough happened during the campaign and was not inevitable. The PCs were all but eliminated from Parliament but continued to divide the overall right-wing vote. The system stretched rightward, but the center still held, as the Liberals were back in power. They were helped by the fact that the party to their left, the labor-oriented New Democratic Party (NDP), lost about half the base that it had accumulated over the previous three decades.
There followed a decade of maneuver and strategizing on the right. Reform renamed itself the Alliance, a signal that its leaders grasped the imperative to grow. They did not abandon core conservative objectives, however, and when they finally merged with the PCs, it was on Alliance terms. The name of the new Conservative Party is unambiguous. With the right now united, the Liberal grip on power loosened and in 2006 the Conservatives took power. They began with a feeble plurality but strengthened it in 2008 and secured an outright majority in 2011.
Over this same period, the NDP recovered, in effect stretching the system’s left flank. If on the eve of the 2011 election it still seemed a minor party, this was an artifact of the electoral system, and the 2011 election reversed the Liberal-NDP difference. The key for the NDP was a breakthrough in Quebec, which in turn induced shifts in the rest of the country. The NDP’s newfound competitiveness in Quebec helped make it a strategic choice elsewhere, too. The 2011 shift, like the 1993 one, has elements of accident. Accident or not, by weakening the Liberals, it enhanced polarization in Parliament.
Did this reflect shifts in the electorate? Unfortunately, there is no comprehensive indicator that goes back to the years before the fragmentation of 1993, or even before the reconsolidation of 2004. We can, however, get some purchase on events since 2004. The 2004 Canadian Election Study (CES) included the 11-point left-right scale that is now standard in the Comparative Study of Electoral Systems (CSES), and repeated it in 2008 and 2011.
Over that span, party groups remained pretty much fixed in place ideologically, as the graph just above shows. (In the graph, the 0-10 scale has been reset and centered for interpretive clarity.) Party identifiers are slightly more polarized than voters, not surprisingly, but within groups is there no obvious trend of increasing polarization. Similarly, the overall ideological dispersion in the sample has not grown: standard deviations of 0.38 to 0.41 depending on whether the calculation is based on the whole sample or just on voters or identifiers. Instead, the vote of the flanking parties, the Conservatives and NDP, has grown, while that of the Liberals in the center has been cut in half. So the big battalions now are on the flanks, rather than in the middle.
The exact numbers should be taken with a grain of salt, perhaps, but the basic point is that the Canadian system has polarized as a result of shifts in the menu of choices. Canadians’ ideological locations are probably better sorted by party than they were 30 years ago, especially on the right. But this sorting was not ineluctable. The shifts embody a large dose of contingency, with Quebec as a major source, amplified by strategic induction through the electoral system.
This is the latest post in our ongoing series on political polarization. The previous posts are listed below. -Dan Hopkins