NOTE: an earlier version of these comments, framed around a potential vote on the finance committee report, may be found here.
The Confidence Convention and the May 2005 Vote on the Public Accounts Committee Report
Canadian parliamentary democracy is hinged upon the fundamental principle that the government of the day must enjoy the confidence of the House of Commons. The principle of collective responsibility provides the basis of deciding who has the right to form a government after an election. Canadians do not vote for a prime minister or for a cabinet, they vote in general elections to elect 308 individual members of parliament. The leader of the party who is able to command a majority on confidence votes in the House of Commons is the person who has the right to be prime minister. It is a firm constitutional convention that prime ministers must either resign or call an election if they lose a vote of confidence in the House. The May 10, 2005 vote on the public accounts committee report is essentially a vote on whether the government should resign, and as such it should be considered a clear vote of confidence.
The confidence convention
There are three generally-agreed categories of confidence votes:
If a government loses a confidence vote it has only two choices: to resign or to call an election. Generally speaking a government would only resign if it lost a confidence vote within a few months of an election. For example, Ontario premier Frank Miller resigned in 1985 after losing a vote on the first speech from the throne right after provincial election, and the Lieutenant Governor appointed the leader of the opposition as the new premier (who then won a subsequent vote of confidence).
Occasionally a government is defeated on a vote that calls into question whether it still enjoys the confidence of the legislature. In the most prominent of these loses, governments have introduced their own explicit motion of confidence for the House to vote on; this is an important response which is needed to settle whether the government still has a right to govern. For example, the Saskatchewan government introduced (and won) a confidence vote in 2001 after their legislature's finance committee had defeated a supply motion. Prime Minister Pearson introduced (and won) a confidence vote in 1968 after losing a supply vote.
The May 10, 2005 Vote
The House of Commons vote on the Conservative Party's motion has generated considerable controversy over whether it is a vote of confidence. The question is whether the vote constituted one of the first category of confidence motions listed above. The motion passed by a 153-150 vote reads:
At first glance the motion appears to be simply a procedural matter, sending a matter back to the finance committee with an instruction to amend its report. However, it should not matter what procedural context a vote of confidence occurs in.
What makes a vote one of confidence is the content of the motion. In order to qualify as a confidence vote, a motion has to contain wording that either states the lack of confidence explicitly, calls upon the government to resign, accuses the government of gross impropriety or incompetence, or questions the authority of the government to remain in office. However the motion is worded, the essence of a confidence motion is to embody the House's judgment that the government is unfit in some way to govern. That specific wording can take many forms, and examples of this variety are found in Canadian provincial and federal precedents.
The fundamental basis of a confidence vote is that the elected members of the legislature express their collective view of the government. If that view conveys a loss of confidence or states that the government should resign, then the government must either resign or call an election.
The wording of the motion passed on May 10, 2005 indicates that it should be considered a clear vote of confidence. What is important in this motion is that the House had to collectively express its view on whether the government should resign. One could not vote for the motion without agreeing that the government should resign, which is the essence of a non-confidence vote. While the wording of the motion is convoluted, the essential content is a clear expression of non-confidence.
The current motion is also strikingly similar, in procedural terms, to that proposed by H.H Stevens on June 22, 1926. That motion also recommended that a committee report be amended and precipitated the whole King-Byng crisis, when the Governor General refused a dissolution to King on the grounds that he should not avoid a censure motion then before the House but not voted on; this was the Stevens' motion. For information on those events, see: House of Commons Debates, 1926, Vol.V, June 22 to June 25.
In light of the past precedents, and especially the relevance of the 1926 motions on the Customs Affair, the current motion appears to be clearly a vote of confidence which would normally require the government to resign or call an election after losing the vote. It is a fundamental blow to the government's authority for a majority of the House to agree on a motion that it should resign.
The fact that several members of the House were missing from the vote on May 10, 2005 might evoke comparisons to the 1968 defeat of the Pearson government's supply motion. In that instance the prime minister argued that several members were unable to return in time for the vote called at short notice. He then adjourned the House for a few days and reconvened with an explicit vote of confidence with all members present who could attend; he won that vote and carried on governing.
The precedents indicate that the proper constitutional course of action is for the current government to introduce its own confidence motion within a few days, if it does not immediately accept the results of the vote as a definitive loss of confidence. It would be against previous precedent and principles of responsible government to simply carry on as if this was merely a procedural vote.
* Associate Professor
Prof. Heard's publications on constitutional conventions include the book Canadian Constitutional Conventions: The Marriage of Law and Politics, Toronto: Oxford University Press, 1991 and several articles: "Recognizing the Variety Among Constitutional Conventions", (1989) 22 Canadian Journal of Political Science 63; "Constitutional Conventions and Election Campaigns," (1995) 18 Canadian Parliamentary Review (3) 8; "Constitutional Conventions and Parliament,"(2005) 28 Canadian Parliamentary Review (2) 7 (forthcoming).