There is no ‘convention’ that the party with the most seats gets to govern

Open this photo in gallery:

Gracious, I seem to have hit a nerve. I had wondered how the Conservatives would react to the suggestion that Justin Trudeau might continue governing after the next election, with the support of the NDP, even if he failed to win the most seats.

Would they recognize how fundamentally federal politics had changed in the wake of the Liberal-NDP supply-and-confidence agreement? Or would they go to the mats, manufacture a phony crisis out of it, and try to brazen it out?

I shouldn’t have asked. Of course they’ll try to turn it into a crisis.

I had expected to hear the hardcore partisans in the leader’s office make this argument: that the Prime Minister, unless he finished election night with a plurality of the seats, was obliged to resign on the spot; that it was not open to him to remain in office even long enough to meet the House and test whether he had its support.

I had expected them to try to paint any such attempt as an unconstitutional power grab, a dictatorial attempt to “steal the election”; to ratchet up the sense of crisis, even to invoke the threat of violence – not threatening it themselves, you understand, but merely pointing out the “potential” – to reinforce their point.

And I had expected them to harness the resentments of their supporters to this end: not only of experts, but of expertise. Which is to say: knowledge. Which is to say: facts.

Thus if every expert in constitutional law asserts that a prime minister holds power on the basis of his or her ability to maintain the confidence of the House of Commons; that the prime minister before the election is therefore still the prime minister after the election, regardless of the result, at least until the new House has met and passed judgment, unless he chooses to resign before then – if all this commands the universal consensus of those who study the subject for a living, that is to be taken not as a sign that this is in fact the case, but as evidence of the class hatred of experts for Conservatives.

What I had not expected was to hear the party’s own intellectual class providing cover for this. Yet ever since I first raised the question there has been a steady stream of commentary from the conservative side claiming that, whatever pointy-headed “academics” might say, there is a “convention,” an “unwritten rule,” a “century of precedent,” that the leader of the party with the most seats always gets first crack at governing after an election. Or at any rate, that this is what “people believe.”

What is the factual basis for this claim? The question obviously does not arise when one party or another wins a majority. Neither is it an issue, in a minority Parliament, when the incumbent party wins the most seats. That would describe eight of the 13 minority Parliaments we have had since the 1925 election.

So we are really only talking about five parliaments, and five elections: 1925, 1957, 1963, 1979, and 2006. If precedent means anything, it was set in the first of these: though he won 15 fewer seats than the Conservatives, Mackenzie King carried on governing, with the support of the Progressive Party, for another nine months.

If the prime ministers in the other four made the opposite decision, then, it was not because of any precedent, but because of a prudent calculation that they were unlikely to survive in power. In 1963 and 2006, this was obvious: the governing party was so far behind their opponents they could not hope to find enough support to carry on. In 1957 and 1979, it was closer. But the calculation was the same.

In all four cases the government would have needed the support of not just one party but two to make a majority. That’s hard to keep together. The next election, however, might well produce a Parliament in which the Liberals need only the NDP to put them over the top. And, unlike all previous examples, they would have the advantage of having spent the previous three years in a formal governing agreement with them.

In any case, conventions are not something you can just declare unilaterally. To be considered a binding rule, rather than just an observed pattern, all sides must have agreed it’s a rule and to be bound by it. There’s no evidence that there has ever been any such all-party accord.

What will govern events, as always, is the numbers. If the Liberals are too far behind the Conservatives in seats, Justin Trudeau may well decline, like his predecessors. At some point, a thing may not be viewed as legitimate, even if it is constitutional. To that extent, what “people believe” is relevant. But it will still be his decision.