Canada 2019

It is the final Friday before Shemini Atzeret, also known as Election Day in Canada this year, And what an interesting campaign it has been! The polls have moved quite a lot, especially recently. The New Democrats (NDP) seem to be enjoying a surge. Not on anything like the scale of 2011, but still something notable, as it was not long ago that there was talk about the Greens possibly passing them for third place. The Greens have slipped somewhat, as has been the case in past campaigns. No longer do they look likely to win as many as four seats; two (which is their current number) looks most likely.

The striking thing is that the poll aggregate at CBC (compiled by Éric Grenier) shows both major parties–incumbent seat-majority Liberal and opposition Conservative) barely above 30% of the vote (31.7-30.8 at my latest check). From 1949 to present, the largest party has never had a vote percentage below 36.3% (in 2006). So if there is not a late surge of strategic voting, this will be quite a record-breaker.

Projecting seats under FPTP is always a challenge. The CBC Poll Tracker currently has the Liberals significantly favored, despite being marginally behind in votes, 133 seats to 123 (but with wide confidence bands on both).  That would be 39.3% of the seats for the largest party, which would also break the record (from 1949 on) set in 2006 (40.3%, or 124 in what was then a smaller parliament).

Despite being both a plurality reversal and a record low vote percentage (and an extremely close vote margin), the advantage ratio (%seats/%votes) of 1.278 for the largest seat-winner would be just about average. Over 22 elections, the mean advantage ratio has been 1.2897. (Note: I am calculating this as the share of the largest seat-winner over its vote share, not over the share of the largest vote-winner, when those diverge.) For those who know Canadian electoral history, I will note that advantage ratios of around 1.2-1.3 have occurred in 1965, 1968, and 2008 (among others). Thus even if the specific vote totals may be very unusual, the workings of FPTP, given the actual votes, is fairly “typical” for Canada.

As for the other parties, I mentioned the NDP surge. But just as noteworthy is the surge of the Bloc Quebecois, which may turn out (again) to be the single most important factor in preventing a majority of seats. The BQ is currently polling just under 7% nationwide, while the Greens are just over 8%.

Of course, the BQ and Green fortunes will diverge in seats. It is very helpful for votes-seats conversion to be a regional party under FPTP, and not useful to be relatively more dispersed. So the BQ is currently estimated to get 38 seats, about the same as a much larger national party, the NDP (41) and vastly more than the also larger–in votes–Greens (2).

Regarding those surges I mentioned. The BQ was, according to the polling aggregate on only about 20% in Quebec as recently as one month ago. Now it is up to almost 30%, and just behind the Liberals’ percentage in the province (31%, having been 37% a month ago). The Conservatives have really crashed in Quebec, down from 22% a month ago to just under 16% now. The latter puts them not too far ahead of the NDP, who are now on about 14% in the province.

Nationally, the NDP was at only about 13% a month ago, but is approaching 19%. A rising vote share tends to lift the seat share–even for a national third party under FPTP. While a month ago, the Poll Tracker had the party at only 15 seats, its 41 projected now represents an increase by a factor of 2.7 when its votes have increased only 1.27 (19/15). The party would still be significantly under-represented by the electoral system, but it has reached a point where it gains a lot of seats by a small increase in votes (assuming it holds and that Grenier’s swing assumptions are reasonable, etc.).

The NDP has also pulled narrowly ahead of the Liberals in the polling aggregate in British Columbia, although still well behind the leading Conservatives.

As for the Greens, their slide has been quite abrupt. They were over 10% as recently as the first of October and were projected to win 4 seats as recently as 16 Oct.

The sixth party in the picture, the far-right Peoples Party of Canada, looks likely to win only the seat of its leader, Maxime Bernier. The riding (district) is Beauce, in Quebec, in which Bernier has held as a Conservative since 2006 until defecting from that party in 2018. (I see the Rhinoceros Party has found a candidate with the same name to put up against him.) For months, the PPC has been at either zero or one seat in the projection.

As for who will form a government, the Liberals seem best placed, even if the result is as short of majority as the Poll Tracker projects. It is possible that they will be weak enough to have to form a coalition with the NDP, even though probably the Liberals would prefer a minority government. On current numbers, Liberal+NDP would be a very bare majority. The coalition or a minority government might need working arrangements of some sort with the Greens and/or BQ as well.

It is much harder to see how the Conservatives can form government, even if they end up edging out the Liberals for a seat plurality. Conservative leader Scheer has already begun the spin just in case, claiming this week (incorrectly) that the party with the most seats gets the first shot at forming a government.

NDP leader Jagmeet Singh has said he would try to form a coalition with the Liberal Party if the Conservatives have the most seats. And PM Justin Trudeau would have the legal right to attempt to work out such a deal and meet parliament to try to retain office. Presumably, Singh (and the Green leader, Elizabeth May) would attempt to extract a concession that 2019 be the last election under FPTP.

33 thoughts on “Canada 2019

  1. On the other hand, LeanTossup has the Liberals well ahead, maybe even able to win a majority on their own (on just 32.5% of the vote). That model also has the Greens winning five seats and the NDP only 25.

  2. It’s quite simple really. All parties of movement will propose electoral reform without defining it. They will then spend some months arguing over different proposals with the supporters of each proposal arguing that no change is better than any competing proposal. There will be several attempts to reinvent the wheel in the course of the argument. They will then conduct Canadian electoral referendum number 297 accompanied by Canadian Fear Uncertainty Doubt campaign number 297 and then all fall about in horror and surprise when FUD wins yet again.

  3. Many moons ago I proposed here that “most seats forms government” seems to be the new default for Canada. I was, rightfully, given some grief for my claim. I cannot remember if I defended it or not, but I do seem to have noticed something. Of course, both I was and the CPC is wrong. In a three party system where no one is willing to work together, first past the post for party leaders may make some sort of sense. As would appointing the seat leading party if no one else can seem to form a government and the incumbent doesn’t want to try. But that is not what Canada has now.

    Back in 2017 after the B.C. election and subsequent coalition deal, I would have pressed for Christy Clark and her Liberal government to be sacked when she refused to resign on notice that a deal between the NDP and Greens had been signed. She decided to met parliament, and then was forced out when she lost the vote on the speech from the throne. She followed the law and custom and remained in office believing that the NDP and Greens would not or could not vote her out. I realized I was wrong when I saw Brian Gallant try to survive in New Brunswick last year, at least until the Conservatives managed to secure a confidence and supply deal with a minor party.

    Allowing the leader with the most seats to form a government makes sense if he has a majority or if no one can bring his government down on the Floor. It won’t work if that leader is automatically appointed before anyone else gets a say. That would lead to either an Indian style race to see who can convince the viceroy that they have the largest caucus/coalition behind them, risking involving the Crown in politics. Or it would lead to farcical situations. Does anyone think appointing a Scheer government next week only for it to fall and for Trudeau to be summoned back to Rideau Hall in a few weeks? Or for a string of elections that result in a minority government being returned for winning the most seats every time, but never with a majority?

  4. Before Miller and Cherry v Prime Minister I would have chastised MSS for writing that Trudeau has a legal right to meet the parliament. Now I probably would not.

    Miller and Cherry is not binding law outside the UK, but I’d expect apex courts in Westminster systems to pay enormous attention to it and probably to follow it. If then, the governor-general did not allow Trudeau to remain in office until the new parliament assembles, I’d think Trudeau would now have a reasonable prospect of success before the supreme court of Canada of getting that reversed on a Miller/Cherry argument. Ditto if Trudeau or Scheer attempted a Harper-style prorogation to avoid facing a hostile parliament. So yeah, after Miller/Cherry it is probably more accurate than not to say a Westminster prime minister has the legal right to meet a new parliament; except obviously where another party wins an absolute majority of the lower house.

    • That seems like something of a stretch. Miller/Cherry didn’t deal with the process of appointing a Prime Minister, and unlike the United Kingdom Canada does have written rules for appointing a Prime Minister (or at least to select those members of the Privy Council who will advise him). Those rules are relatively clear: the Governor-General has, according to Canada’s supreme law, unfettered discretion. Insofar as Miller/Cherry might be relevant, I struggle to see how the key to the case (that Johnson had attempted to stymie Parliament) would fit into a situation in which the Governor-General decided to remove Trudeau: how, exactly, would appointing Scheer reduce Parliament’s ability to act any more than keeping Trudeau in office?

      • The Canadian constitution, like the Australian constitution, is silent on how the powers of the Crown are to be exercised, except for a preamble clause:

        WHEREAS the Provinces of Canada, Nova Scotia, and New Brunswick have expressed their Desire to be federally united into One Dominion under the Crown of the United Kingdom of Great Britain and Ireland, with a Constitution similar in Principle to that of the United Kingdom:

        The Canadian constitution mentions the prime minister only in relation to constitutional conferences required by the Constitution Act 1982. There is no mention of ministers at all, except in some transitional provisions about pre-Confederation provincial ministers. In Canada, as in Australia, New Zealand and until recently Britain, the constitution says nothing about how the powers of the Crown are to be exercised.

        The high court of Australia and the supreme court of Canada have both ruled that the respective constitutions create a representative democracy with a government responsible to the parliament.

        It has always been assumed that the powers of the Crown are not subject to judicial review. For that reason, for example, there was no attempt to apply for judicial review of the governor-general’s decision to dismiss Whitlam. In Cherry/Miller the UK supreme court held unanimously that:

        69. This court is not, therefore, precluded by article 9 or by any wider Parliamentary privilege from considering the validity of the prorogation itself. The logical approach to that question is to start at the beginning, with the advice that led to it. That advice was unlawful. It was outside the powers of the Prime Minister to give it. This means that it was null and of no effect: see, if authority were needed, R (UNISON) v Lord Chancellor [2017] UKSC 51, para 119. It led to the Order in Council which, being founded on unlawful advice, was likewise unlawful, null and of no effect and should be quashed. This led to the actual prorogation, which was as if the Commissioners had walked into Parliament with a blank piece of paper. It too was unlawful, null and of no effect.
        70. It follows that Parliament has not been prorogued and that this court should make declarations to that effect. We have been told by counsel for the Prime Minister that he will “take all necessary steps to comply with the terms of any declaration made by the court” and we expect him to do so. However, it appears to us that, as Parliament is not prorogued, it is for Parliament to decide what to do next. There is no need for Parliament to be recalled under the Meeting of Parliament Act 1797. Nor has Parliament voted to adjourn or go into recess. Unless there is some Parliamentary rule to the contrary of which we are unaware, the Speaker of the House of Commons and the Lord Speaker can take immediate steps to enable each House to meet as soon as possible to decide upon a way forward. That would, of course, be a proceeding in Parliament which could not be called in question in this or any other court.

        The same reasoning is at least arguable before courts in Westminster countries where the powers of the Crown are not regulated by law. as noted by Professor Anne Twomey, whose book was cited extensively by parties to the proceedings and whose chapter on prorogation was read by all lords justices:

        They would certainly give serious consideration to it, as this is the only precedent on the prorogation of parliament in a Westminster-style system of government, and the unanimous judgement of a significant court.

        Moreover, the UK court’s reasoning is very similar to existing Australian cases in which courts have ruled that the common law must be interpreted in a manner that is consistent with constitutional principles.

        This means that Australian governments should, in the future, be quite careful when proroguing parliament. They will need to ensure they do not do so for unnecessarily long periods of time and to prevent parliament from fulfilling its legislative and scrutiny functions, especially during periods of political controversy.

      • Right, but that principle gets Trudeau as far as the Supreme Court. He would then have to argue that the principle that the incumbent Prime Minister has the right to meet the house is actually binding. I just don’t see how that could be done. As you say, there’s no formal constitutional requirement, and I struggle to see how this particular rule is intrinsic to Canada’s status as a representative democracy with a government responsible to Parliament.

      • The argument would be that the convention that a prime minister has the right to meet a new parliament where there is no alternative government with a clear majority is as essential a component of the Westminster system as the principle that prorogation cannot be used to subordinate the parliament to the executive. I don’t say that argument would necessarily succeed, but it would be a very strong argument. And as Wilf notes all the precedents from Canada and elsewhere are that a prime minister has that right. What is different after Cherry/Miller the Westminster conventions are at least arguably subject to judicial review.

        We are all going to have to do some thinking about precisely what that means for democracy in uncodified Westminster systems.

        PS I forgot to provide a link to the Twomey article.

      • Alan,

        “The consensus view before Miller/Cherry was that the exercise of a reserve power was not justiciable. That view was reflected in the opinion of the Scottish court of session at first instance… The divisional court for England and Wales, the divisional court in Northern Ireland, and the government submissions to the UK supreme curt all took the same view.”

        How? Why? The Meeting of Parliament Act 1694 requires Parliament to meet at least once every three years. So a prorogation can’t last more than three years. Ignoring the newer rulings, how could anyone reach the conclusion that the prerogative power of prorogation is non-justiciable when there is a law explicitly limiting it? Again, the same can be asked about Australia and Canada, which constitutionally require parliament to meet at least once a year.

      • jd

        Justiciability is horribly complex, although it is considerably simpler post Cherry/Miller. No-one has ever brought a case under the Meeting of Parliament Act. If they had the courts, until now, would almost certainly have ruled that the issue was non-justiciable.

        The mere existence of a constitutional or legal provision does not inherently mean that rule can be enforced by the courts. For example, none of us are unfamiliar with the phrase ‘Treason, Bribery, or other high Crimes and Misdemeanors’. But the supreme court ruled in Nixon v United States that the way the house of representatives and the senate deal with impeachment and removal is absolutely non-justiciable.

    • Whence the notion that prerogative powers are categorically immune to judicial review? Britain, Canada, and Australia all have requirements that parliament meet at least once in some period. In the latter two, this is written into the constitution. So surely prerogative powers are limited and therefore justiciable from the start, regardless of how small the limitations are?

      • The terms are confusing. The royal prerogative is defined by the NZ Cabinet Manual:

        1.12 The royal prerogative is the discretionary power held by the Sovereign under common law. Most of the Sovereign’s prerogative powers have been delegated to the Governor-General. The Sovereign has retained a number of prerogative powers, including the conferment of certain honours. See paragraphs 1.55 – 1.64 for details of the New Zealand Royal Honours system.

        The situation in Australia and Canada is slightly different because while the governor-general performs functions that are prerogative in the UK. the functions are constitutional in the two countries. The prerogative can be limited or abolished by statute. The Fixed Term Parliaments Act 2011 in the UK, for example, abolished the royal power of dissolution. As it happens, I think abolishing the royal power of dissolution was very poor constitutional design. The reserve powers in Australia and Canada, strangely enough, cannot be limited or abolished except by constitutional amendment where in Britain and New Zealand an ordinary act of parliament is enough. To add to the confusion, British lawyers tend to talk about personal prerogatives, those exercised by the Queen without advice. The classic examples are appointing and dismissing a prime minister, and dissolving or proroguing.

        A better term is reserve powers, which gets over the prerogative/constitutional distinction. Although Twomey’s The Veiled Sceptre: Reserve Powers of Heads of State in Westminster Systems was extensively cited by all sides in Cherry/Miller they otherwise kept on using the older language about personal prerogatives.

        The consensus view before Miller/Cherry was that the exercise of a reserve power was not justiciable. That view was reflected in the opinion of the Scottish court of session at first instance which held:

        [17] The exercise by the Sovereign of the power to prorogue upon receipt of advice from the prime Minister is governed by constitutional convention alone. The courts cannot enforce a political convention. The sanction for non-observance of a convention is political, not legal.

        The divisional court for England and Wales, the divisional court in Northern Ireland, and the government submissions to the UK supreme curt all took the same view.

        The court of session in its appellate jurisdiction took a different view, as did the UK supreme court.

        Henry has questioned why the convention that a prime minister has the right to meet the new parliament if it is not clear there is an alternative prime minster with a parliamentary majority is central to the Westminster system. The easy answer is that all the precedents say that convention exists and even after Cherry/Miller the courts’ job is upholding the conventions as they are now, not writing a new set.

        A better answer is that if the Crown can appoint a prime minister who does not hold a clear parliamentary majority, then the Crown is substituting their own decision for that of parliament. Even if (I am not putting this forward as a likely scenario) the NDP, the Bloc and the Greens made public statements that they would back a Conservative government, Trudeau would still have the right to test that when the house first met. Parliament must always win. The reserve powers exist to ensure that parliament always wins.

      • It is unclear to me exactly what the principle that the Prime Minister should only be appointed if they have a clear majority has to do with the precedent that the incumbent Prime Minister has the right to test the confidence of the house, particularly if that incumbent Prime Minister does not have anything like a clear majority after an election (and if they are in a considerably worse parliamentary position than the Opposition Leader). It is also unclear to me how it is a win for Parliament if a majority of MPs tell the Governor-General that they would like Scheer to be appointed but the Governor-General is obliged to keep Trudeau in office with all the powers of the executive until he gives a Throne Speech.

    • On the question of whether the incumbent PM has a specifically legal right to meet parliament in the Canadian case, I am relying on an excellent article by Philippe Lagassé (in particular, p.6, second column).

      • Lagassé certainly establishes that there is a convention that a first minister has the right to meet a new parliament. Before Cherry/Miller the consensus view was that constitutional conventions were not enforceable in the courts, a point Lagassé also makes. This particular convention will almost certainly never be litigated because it’s just about impossible to imagine the Crown ever breaking it.

  5. Several points here:

    The authority for the incumbent Prime Minister to meet the House in Canada is undoubted by anyone but Conservative Andrew Scheer, who knows better but is making mischief. It is noted on the Library of Parliament website: “If no party gets a clear majority, the cabinet that was in office before and during the election has two choices. It can resign, in which case the Governor General or lieutenant-governor will call on the leader of the largest opposition party to form a cabinet. Or the cabinet already in office can choose to stay in office and meet the newly elected House — which, however, it must do promptly. In either case, it is the people’s representatives in the newly elected House who will decide whether the “minority” government (one whose own party has fewer than half the seats) shall stay in office or be thrown out.” We did not need Miller and Cherry. Nor has anyone argued that the PM is obliged to follow an opposition Accord; the Ontario Premier in 1985 went through the doomed motions of meeting the House and losing the first non-confidence vote with not a squeak of outrage from anyone (it isn’t final until the MPPs vote on the floor of the House.)
    “All parties of movement will propose electoral reform without defining it” was no doubt a tongue-in-cheek statement, but for the record, the NDP platform states “A New Democrat government will bring in mixed member proportional representation that works for Canada – and we will do it in our first mandate in government. We’ll establish an independent citizen’s assembly to recommend the best way to put it in place for the next election to ensure both local representation and a federal government that reflects the voters’ choice of parties. Once Canadians have the opportunity to experience the new voting system and compare it to the old one, we will hold a referendum to confirm the choice.” Jagmeet Singh has elaborated that “the opportunity” means “two elections.”
    NDP insider Brian Topp has helpfully written about the 2008 Liberal-NDP Coalition (“An Accord on a Cooperative Government”) in a book titled “How We Almost Gave the Tories the Boot” which I expect all insiders have re-read. It says the NDP’s “scenarios committee” quietly met 10 days before that election and made plans. If it met this year on the same schedule, shortly after that meeting Jagmeet answered a reporter’s question about a coalition with the Liberals not with the expected “that’s hypothetical, I’m running for Prime Minister ” but with, in his trademark breeziness, “Oh, absolutely.” The Bloc will do no such thing, but this morning Erie Grenier gives the Liberals 133 and the NDP 41 for a relatively comfortable 174 (170 being a majority), so the Bloc’s 38 will be rendered powerless unless the Liberals choose to operate bill-by-bill and try to play the NDP and Bloc off each other, arguably unwise. If those numbers drop slightly, Jody Wilson-Raybould and/or Jane Philpott may find themselves able to take sweet and just revenge for their arbitrary exclusion from the Liberal family, and Elizabeth May may have the share of the power she declines to exercise (she keeps saying she will never whip a caucus member’s vote, meaning any accord with the Greens would require unanimous consent of her little caucus for any government action) so she will just sit back and hold the government to account rather than dirty her green hands.

    • If the NDP policy is MMP then the citizens assembly would presumably be bound to that policy and could not be described as independent. As citizens assemblies have been such a successful mechanism for achieving electoral reform in Canada, no doubt we can once again expect this one will make a proposal that will romp home.

      • In 2016, the Liberals had a majority, and opposition to a referendum meant that they could pass any system they liked. If the Liberals are dependent on the NDP for a majority, they will presumably be expected to support a more proportional electoral system, which goes directly against their partisan interests. Under such circumstances, I don’t think it would be totally unexpected for the Liberals to have a sudden conversion to the idea of a referendum.

    • I am puzzled by the NDP position that there should be a citizens assembly solely on how to implement an electoral system the party has already decided on (MMP). Of course, there are plenty of variations on MMP, but I share Alan’s point that it is hardly “independent” if the most important decision has already been made for it.

      Alas, this question appears rather moot now.

  6. What about Ireland?

    “The Taoiseach is appointed by the President upon the nomination of Dáil Éireann, the lower house of the Oireachtas (parliament), and must, to remain in office, retain the support of a majority in the Dáil.” I am quoting Wikipedia. Is this a better model than the unwritten that the monarchy/governor general negotiates with the various party leaders and appoints a Prime Minister.

    https://ecpr.eu/Filestore/PaperProposal/ba2bdffd-a910-4bae-9c05-bb832db684a5.pdf

    • Canada seems to have some success with minority government, depending on how one defines success. I personally don’t have a problem and see advantages to minority governments and negative parliamentarianism, which is difficult to achieve in Ireland. Parties there need to actively support or at least actively abstain from an investiture vote, do they not?

      Whatever the case for Ireland or one’s views on minority governments, I do not believe that that the Governor General negotiating with party leaders is part of the Westminster system in Canada or anywhere else in the Anglophone world. Party leaders sometimes can present the case that they have majority support. One can find something like that in Australia, New Zealand and India. In Canada, the incumbent first minister either faces parliament or gives way to someone else.

  7. In relation to the discussion here I have two completely unrelated questions. What happened in NZ at their last change of government? Did the Parliament vote against the Nationalist party (who were, if I recall, only a seat or two short of a majority and the clear minority government favorite)? Or did the switch happen before Parliament sat, and, if so, was it with or without the advice of the Prime Minister. I suppose it was not without the advice. I think it was quite a surprise to many observers that Jacinta Ardern was able to pull it off.

    And completely unrelatedly, the Canadian party system is not unified or federalised, but as far as I know, each province has its own set of parties with separate membership, which may be quite strongly opposed to the federal party of the same name. How did this come about? Was it made inevitable by the relative strength of the provinces, or did it make the provinces relatively stronger compared to the centre? Did language differences have any role? or was Quebec merely one of ten provinces in the process? I guess I don’t ask “what was the boring history” but something more along the lines of “what is the opinion about whether, had certain contingencies not happened, the whole thing would’ve collapsed anyway”.

    • The incumbent prime minister resigned once it was known that Jacinda Ardern (the most popular political in Australia, but that’s another story) was in a position to form a government. The election was on 7 October. The incumbent prime minister, Bill English resigned on 26 October and was succeeded by Ardern. Ardern met the new parliament on 8 November.

  8. Constituent assemblies are successful in Canada? The BC one, advised by an advocate for STV, voted about 98% in favour of STV. The Ontario one, advised by advocates for MMP, voted about 98% in favour of MMP. All this proved is that constituent assemblies are sheep to be driven, and the key factor is who is appointed to drive them.

    • I suspect Alan was being sarcastic, but yes, relying on a constituent assembly for a matter that is relatively distant to the ordinary voter like electoral systems has the flaw of relying very heavily on the willingness of those in charge of the assembly to provide genuinely balanced information to the members.

      • You are quite right. I misunderstood Alan’s irony/sarcasm for serious praise. As you say, however, the point holds.

  9. It is a reverse plurality election. Liberals form government despite not being the plurality winner in the popular vote. Will this lead to electoral reform of PR?

  10. Pingback: Canada 2019: Results and a good night for the Seat Product Model | Fruits and Votes

  11. There is now a post-election/post-holiday planting. May I suggest that new comments go there (unless they are direct responses to an earlier comment or my pre-election remarks).

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.