Governing with consent

by Aaron Wherry

Last week, Mark Donald heralded a “tide of ennui.” This week, Andrew Coyne writes, somewhat less satirically, of our “deeply, deeply cynical political culture.”

On those notes, some math. Namely, the mandates of each government in our history, expressed not as a percentage of seats won or votes cast, but as the percentage of possible votes. In other words, what percentage of eligible voters actually chose to support the government that governs them.

First, the ten strongest mandates in Canadian history.

1. Borden, 1917 42.8%
2. Diefenbaker, 1958 42.6%
3. Laurier, 1900 40.2%
4. Mulroney, 1984 37.7%
5. Laurier, 1904 37.6%
6. Macdonald, 1882 37.5%
7. Mackenzie, 1874 37.4%
8. St. Laurent, 1949 37.0%
9. Macdonald, 1878 36.8%
10. Bennett, 1930 36.7%

And now, the ten weakest.

1. Harper, 2008 22.2%
2. Martin, 2004 22.4%
3. Harper, 2006 23.5%
4. Chretien, 1997 25.8%
5. Chretien, 2000 26.2%
6. Clark, 1979 27.2%
7. King, 1921 27.6%
8. Laurier, 1896 28.4%
9. Diefenbaker, 1957 28.9%
10. Chretien, 1993 29.3%

Twenty of our 40 governments received between 29.5% and 36.6%. Without reprinting the entire list, the trend line should be fairly obvious.

(Note: All of the above with printed the standard caveat about my math skills. Email any and all corrections. Voter turnout numbers are here. A handy guide to election results is here. Calculation used: Popular Vote , divided by 100, multiplied by Voter Turnout)

Governing with consent

  1. Good list, Aaron! It really highlights the decline in voter turnout during the last forty years – a worrisome trend. For anyone who is interested in the raw voter turnout numbers, Elections Canada provides a handy list:

  2. Why is Coyne complaining? He asserts all the time that Canadians are too cretinous to want or deserve better from our politicians and, well…reality is unfolding as it should.

  3. I remember 84 and the B man … worked my keyster off for that guy then .. ah to be so young again … sigh … Whwers Trudeau wassa matter here?

    • Lucky you didn't decapitate youself.

      • How does a ?

  4. This list is really about voter turnout, not popular support.
    The five "weakest mandates" are the last five elections.

    The only standouts that bucked the tide of voter turnout are Mulroney in 84 on the plus side, King and Laurie on the minus side.

    You could do a similar list, but instead of listing the most voters for, you could list the percentage of voters against. Those elections with the least number of voters opposed to the winner would then be the strongest mandates.

    Guess what? You'd get the reverse, Harper would be on top.

    • Only 37.5% of eligible Canadian voters voted against Stephen Harper in the last election.

      • The problem with that statement is that a vote for someone isnt necessarily a vote against someone else. A voite, as we can read it, only speaks to who you say you want, since that is the question being asked at the ballot level.

        This is why that statement is a little simplistic and the implication that there isnt consent of the governened is not a conclusion one can draw from current voting data. You can only get that by driving questions to A-B choices. So in France when they do the run offs then you get to that, but notice participation drops there for the run offs.

        Legitimacy and consent are subtle and supple concept's not easily captured in mathematical proxies…..but it is interesting to say the least

        • My tongue was planted firmly in cheek when I wrote that. I completely agree that a vote for someone isn't necessarily a vote against someone else. Similarly, a vote for nobody isn't necessarily a vote against somebody. That's why it really only makes sense to measure a politician's "mandate" as a percentage of votes that were actually cast, rather than as a percentage of possible votes.

          • My apologies. We are in broad agrement, one would almost say there is a broad consensus.

          • Heh.

          • I agree. Using votes not cast as proof of anything is silly.

          • Except as a rebuttal of sorts to the ridiculous "no legitimate mandate" arguments some try to bring up against the winners of low-turnout elections. If some want to use such a dumb argument, then the "insufficiently-outraged-against" index does have some worth after all…

    • One other point is that all the the top 5 come in the post Bloc era when 7-10% of all votes nationally have been siphoned off to a perpetual opposition party. So some of the recent drop in proxy is explained by that.

      • True, it's harder to gain a large number of votes when there are more parties in play (vote splitting), so the addition of a regional party has an effect. Even the existence of the greens, who get over 5% of the vote, has an impact, since those votes would mostly be going elsewhere otherwise.

    • Your suggestion that "the list is really about voter turnout, not popular support" makes an assumption about causation that I am not sure is substantiated.

  5. top 5 "weakest" I mean

  6. This is an interesting and important analysis–thank you, Aaron.

    I think, however, that this analysis unfairly conflats the decline in overall voter turnout (i.e. animosity toward politicians in general) with the increase in support for "alternative" parties (i.e. animosity toward the two old line parties).

    As the chart below shows, there's a remarkable correlation between support for the government as a percentage of votes cast and as a percentage of the population–the red and blue lines match up almost perfectly! While decreasing voter turnout certainly exacerbates the downward trend in the government's "mandate", the driving force seems to be declining support for the older parties more than a rejection of politics in general.

  7. Failing to be arsed to show up to cast a ballot leads to implied consent to go along with those who could be arsed.

    • Yes, I agree as well.

  8. Another great example of this phenomenon is that in the last election the so-called "third parties" almost won more combined votes than either of the two traditional parties. The Conservatives got 37.6% of the vote on election day while the combined total of the third parties was close behind at 36.2%. This is only the second time in Canadian history that this has even been close to being true (the first being at the height of the "Progressive" movement).

    <img src="; border="0" alt="Image and video hosting by TinyPic">

  9. Welcome to "popular discontent" as the new Coca-Cola — everywhere out of favour

  10. Great list….love this kind of stuff. I think Borden needs an asterix beside him though give the times:

    To force the Liberals into the Union government, and to ensure its victory in the upcoming election, Borden changed the country's election laws. Soldiers would be allowed to vote simply for or against the government, with their ballots being counted in ridings of their choosing. Borden also gave the vote to some women: military nurses, and the wives, mothers, sisters and widows of soldiers would all be allowed to vote in the 1917 election. To further determine the election outcome, those immigrants from countries with which Canada was at war who had become Canadian citizens after 1902 were stripped of their right to vote. The government also committed itself to exempting farm workers from conscription.

    • Exactly right, CBP. Borden's support is so high because he eliminated much of the likely opposition vote and gave the franchise to women who would support him. A lesson for our present leaders!

  11. I think this is what people call "progress". Good job progressives!

  12. What this reflects is the legacy of the great Mulroney which was to unite us in our perpetual diviseness.


    I wrote about this a few months ago and included a downward sloping trendline for all mandates since WW2. With two notable outliers (1958 and 1984), there has been a steady trend downwards in the size of the government mandate.

  14. The trend over the last few years also reflects that we currently have more viable political parties than at any other point in our history. In the first several elections, there were really only two parties. From the 1920s forward, there were three or four – and usually the third and fourth parties between them wouldn't have amounted to more than 20% of the vote and 10ish percent of the seats.

    Since 1993, there have been five viable national parties contesting each election. with the third through fifth place parties (by popular vote) taking in excess of 30 percent of the vote and as much as one third of the seats.

    Thus it is not entirely surprising that the winning party is winning with a smaller share of the votes cast even before we work in the implications of the declining voter participation.

    • There are not five viable national parties, …unless of course you wish to clasify the Bloc as a "national" party. It is most definitely a regional party in the normal sense and its existence distorts the overall numbers.

Your email address will not be published. Required fields are marked *