14

On electoral reform, what are Canada’s options?

Confused by all the proposals on the table? We break down the current system and four popular alternatives.


 

Liberals are determined to change how Canadians vote at the federal level, and the burgeoning debate has pundits talking about all kinds of alternatives to the first-past-the-post system that Canada has used for 149 years. There may be no consensus among academics or politicians about the option that’s best for the country, but everyone has their favourite.

Electoral systems matter. How we pick our representatives has a huge influence on the makeup of our legislatures. Let’s learn a bit more about the most likely alternatives to FPTP. Keep in mind, there are many, many variations on electoral systems, and we’ve only outlined a few. To make it easier, we’ve included cute illustrations. (Everything is easier to understand with adorable drawings.)


First past the post

Before we look at the alternatives, let’s sketch out our current electoral system. FPTP is a simple and common structure: whichever candidate garners the most votes wins. It’s used in about 80 countries in the world, including the United Kingdom and the United States. However, a lot of people don’t like it—including our current Prime Minister.

fptp_test

Pros: It’s easy to understand and simple to implement.
Cons: Winners don’t need an absolute majority, just more votes than anyone else. Candidates often win with fewer than 50 per cent of votes. The majority of a constituency may not vote for the eventual winner, and winning parties rarely win a majority of votes nationally.


Proportional representation

This is a popular alternative to FPTP, and forms of it are more widely used around the globe. The idea is to ensure the share of a party’s representation is proportional to its share of votes. Because the system allows more parties to win seats, the system often requires parties to form governing coalitions in order to accomplish anything.

pr

Pros: It’s easy to understand and allows small parties a greater likelihood of winning seats. May allow for greater consensus via cooperation among parties.
Cons: Can create a government comprising many small parties where consensus is difficult to achieve. Though rare, it may allow a foothold for extremist parties. Since seats are filled from a party list, voters may have no geographic connection to their representatives.


Ranked ballot

Sometimes called preferential voting, the ranked ballot system allows voters to list candidates in order of preference. Below we detail the system in use in Australia: the instant-runoff ranked ballot system. Simply put, if no candidate wins more than 50 per cent of votes, a series of runoff counts follow, where second and third preference votes are allocated until there’s a winner.

ranked_fix

Pros:May reduce tactical voting, as voters don’t fear wasted votes. Candidates in instant-runoff systems have also been known to participate less in attack campaigning, as second- and third-choice votes can be valuable.
Cons: Apathetic voters may be prone to simply ranking candidates in the order they appear on the ballot, giving preference to the first name listed.


Single transferable vote

Single transferable vote is a complex system of ranked voting. It achieves nearly proportional representation through ranked ballots in constituencies with multiple seats. It’s not a widely used electoral system, but in 2005 and 2009 it was put to a referendum in British Columbia. Voters rejected the system both times (a simple majority of voters did vote in favour of STV in 2005, but proponents failed to achieve the required 60 per cent majority).

stv

Pros: Fewer votes are “wasted,” and most voters can point to a representative they helped elect.
Cons: Votes take much longer to be counted and can be influenced, similarly to ranked voting, by voters who simply list favourites in the order names appear on the ballot. The system is also difficult to explain and understand.


Mixed member proportional

Mixed member proportional is another complex electoral system, but it has found success in both Germany and New Zealand. MMP offers a hybrid voting method that incorporates parts of both proportional representation and FPTP. In 2007, voters in Ontario rejected a proposal to adopt MMP for provincial elections.

mmp

Pros: Allows smaller parties a chance to obtain representation. Achieves proportional representation and maintains geographic constituencies.
Cons: A complicated system that is difficult to explain and understand.


Editor’s note, June 21, 2016: This post has been updated to clarify certain elements of alternate voting systems.


Your turn! How you think Canada should reform its electoral system … if at all?

  • These thoughts will not be added into the comments, which are below. We will gather these reader opinions to share in a future article.

 

On electoral reform, what are Canada’s options?

  1. Props to Maclean’s for attempting to give us some information on this crucial issue. And the graphics are adorable. Unfortunately, a great deal of the “information” is inadequate or actually misleading.

    For example, “whichever candidate garners the most votes wins” is true of every voting system, and is hardly adequate as a description of first-past-the-post. The essential element of FPTP is that only one MP is elected in each riding, with the result that most of us are “represented” by somebody we voted against.

    And so on. For better info on voting reform: http://FairVote.Ca

    • Hi Wayne,

      Thanks for the feedback. In the graphics, we seek to simply explain the electoral mechanisms for each system (some of which really are quite complex!). Underneath, we list a few of the pros and cons; you’ll notice that a con for FPTP is certainly that often candidates don’t represent the majority of their constituents.

      Cheers!
      Amanda

    • Unfortunately, Fair Vote has already determined that the outcome should be some form of proportional representation. Given that my first choice for a new system is the ranked ballot, that leaves me in the cold.

      I’m not trying to get into an argument about which system is best (or least worst). Just pointing out that Fair Vote is not unbiased.

      • Ah, but Jim, the ranked ballot can be used in just about any proportional system, and Fair Vote hasn’t ruled that out at all. So, unless there’s a reason why you don’t want fairness to voters, or national unity, or all regions of the country in both government and opposition, we can all get what we want.

        Fair Vote is only biased because we have incredibly detailed discussions based on evidence, and the evidence points to a form of proportional representation. We are unbiased (as an organization) on which form that takes.

        As to the complexity of the systems, it may be more of a challenge for Elections Canada–who will not likely have the outcome determined before British Columbia’s polls close anymore. I don’t see that as such a bad thing, and it is certainly easy enough to vote under any of the systems. And it isn’t so hard that EC will be unable to do the job. You know, electricity is complex, too, but electricians can handle it.

        • There is more than one way to provide “fairness to voters”. Ranked ballot will certainly be a lot fairer than FPTP wthout the problems that many of us see with proportional representation. It will also allow people to vote for a new or smaller party knowing that their second and third choices will come into play if their first choice does not work out. The proportional representation groups are nearly fanatical with their adherence to their ideology. It reminds me of religious groups who believe their way is the only way. Anybody can find “evidence” to support their cause. Just look at those desperately trying to justify holding onto FPTP.

  2. You have described the MMP model rejected by Ontario voters in 2007, but no one proposes that model today. The Law Commission of Canada recommended a vital improvement on the MMP model used by Scotland and Wales: that you can cast your second vote for your party’s regional candidate you prefer, so all MPs have faced the voters and been elected. This is the open-list MMP model which the House of Commons debated on Dec. 3, 2014, when half the Liberal caucus voted for it. So you have not yet given your readers accurate and relevant information about MMP. I trust you will fix it.

  3. This article makes a number of common errors. E.g. the author says that a “con” of Plurality Voting is that “Winners don’t need an absolute majority”. But whether or not there’s a candidate favored by an absolute majority has nothing to do with the voting system—that’s simply a state of voter preferences.

    Amanda mentions Instant Runoff Voting (IRV), one of the countless ranked voting methods out there (and generally the worst; see Borda or Condorcet if you want sensible ways of tabulating ranked ballots). This system gives the superficial appearance of guaranteeing a “majority winner”, but this is purely illusory. Consider this hypothetical election.

    % of voters their ranking
    35% W > Y > Z > X
    17% X > Y > Z > W
    32% Y > Z > X > W
    16% Z > X > Y > W

    Instant Runoff Voting selects candidate X as the winner, beating W in the final round, 65% to 35%. But a huge 67% majority of voters would rather have candidate Y than X. AND Y received nearly TWICE as many first-place votes as X, 32% vs. 17%. So any notion that IRV has elected a “majority winner” is patently false. Note also that an even larger 83% super-majority of voters would rather have candidate Z than X (and Z got just a little fewer first-place votes than X).

    Fact: IRV only guarantees that the winner is preferred to *at least one other candidate*.

    Amanda then says that one of the pros of IRV is that it, “Reduces tactical voting, as voters don’t fear wasted votes.” The latter claim is false. IRV can indeed punish voters for “wasting” their first-place ranking on a “spoiler” who can’t win. See a simple presentation by a math PhD who co-founded the Center for Election Science.
    https://www.youtube.com/watch?v=JtKAScORevQ

    It would take many pages to correct the multitude of other mistakes here. But perhaps the most effective use of keystrokes here is to point out that the systems discussed here are ALL antiques, invented well before the advent of modern social choice theory (e.g. the Gabbard-Satterthwaite Theorem, Arrow’s Theorem. There are better and simpler methods such as Approval Voting, Score Voting (and proportional variants of both these systems that are simpler than STV), and Asset Voting. Warren Smith, a Princeton math PhD whose work was featured in the book “Gaming the Vote” elaborates on them here.
    http://scorevoting.net/CanadaOverview.html

    It is really quite unfortunate that (apparently) no one with really solid expertise on electoral system design is even active within Canada so much as explaining these basic electoral system design concepts. It’s analogous to designing a ship to take us to Mars, but only involving scientists whose knowledge contains nothing more recent than the 1930s.

    Clay Shentrup
    Co-founder, The Center for Election Science
    Clay Shentrup

  4. Thank you, Amanda. A couple of thoughts. Ranked ballot is also simple and easy to understand although I will say the way the graphics were used it made it seem more complicated than it really is. Second, it would have been instructive to point out that the major political parties (most recently the Conservatives selecting Rona Ambrose) use ranked ballot or instant run-off to select their leaders. And although people will try to differentiate selecting a leader from selecting an MP, there is no difference. One winner is declared from 5 or 6 candidates standing for election.

  5. I recently explained how John Stuart Mill, reflecting on Hare’s System of Personal (and Proportional) Representation – that is STV – founded the philosophy of Peace-making Power-sharing (the title of my first e-book on election method).
    A commentor told me: “Think for yourself. It will do you good.”
    That remark is no more outrageously misleading than the tenor of the Canadian electoral reform debate.
    I have just published my third book on election method and science: Science is Ethics as Electics.
    https://www.smashwords.com/books/view/643703

    The other book is: Scientific Method Of Elections.
    All three books are free from Smashwords.
    (Amazon charge a minimum fee.)
    Richard Lung.

  6. Amanda,

    Maclean’s may not be aware of Australia’s Seventh post – Double Dissolution 45th General Elections of the House of Representatives and the Seventh Full Senate Elections (and Territorial Senate Elections since 1975) that is going on as we debate Electoral Reform. This was the seventh Double Dissolution since Federation on the 1st January 1901, and the first since 1987. Pretty good record for rare use of the Governor General’s extraordinary Constitutional power of Double Dissolution.

    On Sunday the 8th of May, Australia’s Prime Minister Malcolm Turnbull visited His Excellency General (Retired) the Honorable Sir Peter J. Cosgrove at Government House, Yarralumla in Canberra, Australian Capital Territory, to formally advise in a written argument requesting His Excellency to exercise the Governor General’s extraordinary Constitutional power under Section 57 of the Constitution of the Commonwealth of Australia, to dissolve the Senate and the House of Representatives simultaneously effective 9 A.M. on Monday the 9th of May 2016 with Polling Day on Saturday the 2nd of July 2016.

    In your magazine article above, you had simply explained the Preferential ranked ballots, or the Instant Run — Off ballots, that Australia uses for the House of Representatives in the one hundred and fifty Commonwealth Electoral Divisions; whereas, the Senate Elections uses the Hare – Clark System of Preferential Quota – Proportional Representation, or the Proportional Representation – Single Transferable Vote, electoral system was changed just in March 2016 to abolish the Group Voting Tickets, and returned the power of preferences to the electors through the institution of Optional Preferential Voting in Above The Line, and most importantly, in Below The Line voting. For further information on Australia’s electoral systems, readers and Maclean’s could checkout the following websites:

    For House of Representatives Elections: http://www.aph.gov.au/About_Parliament/House_of_Representatives/Powers_practice_and_procedure/00_-_Infosheets/Infosheet_8_-_Elections_for_the_House_of_Representatives

    For Senate Elections: http://www.aph.gov.au/About_Parliament/Senate/Powers_practice_n_procedures/Senate_Briefs/Brief01

    The Australian Electoral Commission: http://www.aec.gov.au/

    Documents Relating to the Double Dissolution: http://web.archive.org/web/20160508072551/https://www.gg.gov.au/sites/default/files/files/IPS/Documents%20relating%20to%20calling%20of%20the%20double%20dissolution%20election.pdf

    For excellent coverage of the Australian General Election, please see the Australian Broadcasting Corporation’s resident psephologist and election analyst Antony Green’s well-written blogs: http://blogs.abc.net.au/antonygreen/

    Senate Elections: http://blogs.abc.net.au/antonygreen/senate-elections/

    Double Dissolutions: http://blogs.abc.net.au/antonygreen/double-dissolutions/

    Sincerely yours,

    Ronald A.McCallum
    B. A. (History and Political Science)

  7. One comment on the couple of ranked options that you’re worried about being influenced by people sequentially down the list:
    This can be mitigated by printing ballots with the names placed in the list randomly, as I believe is done in the elections for the House of Reps in Australia. If 100 people in a riding number their lists from top to bottom, the preferences all go a different, completely random way.

    • Are you serious ? Do you truly believe these thoroughly corrupt Liberals would EVER “randomize” the ballots ? Rest assured, the Liberal candidate would appear at the top of EVERY single ballot printed by Liberal-chosen printing companies ! Of course, they would mumble some vague, nonsensical, sort-of-promises that the NEXT election would be different… Sunni ways !

      • Wow, Andrew. Aren’t you full of sunny ways yourself!! Your comment is a little rich given the nastiness and deceit of the previous 9 and a half years. You are still so under the influence of Conservative “funny ways” (robo calls, electoral fraud, phony Fair Elections Act etc) that it is hard for you to imagine that not everyone else is like that. Hopefully electoral reform will prevent a Harper clone and the accompanying mindless minions (of any political stripe) from ever holding this country hostage again. It was like a dark shroud had been lifted off this country on Oct 20th last fall. Try to relax, get out and get your anger under control. The witch is dead!!

  8. Sorry I’m late to the party on this. Took me a while to get myself registered. Fishing intervened :-)

    Just a few points of contention with some assumptions put forth in this piece.

    “FPTP is a simple and common structure: whichever candidate garners the most votes wins. It’s used in about 80 countries in the world”

    The US and UK are then stated as examples, leaving some readers to assume that this is a widely used system in major democracies. Yet the list of countries using FPTP isn’t exactly an exclusive club of major democracies. Aside from Canada, US, UK and India, they are many small countries like Antigua, Bahamas, Cook Islands, Dominica, Gambia, Saint Lucia, and Virgin Islands. Simply stating that 80 countries use it is not evidence of its appropriateness as an electoral system. There are over 70 countries in which it’s still technically illegal to be gay, but that obviously isn’t okay. http://www.bbc.com/news/world-25927595

    Furthermore, some of the countries using FPTP have some questionable democracies, with Freedom House ranking at least 25 of them as either “Not Free” or “Partly Free” and 17 of them ranked by Democracy Index as “Authoritarian” or “Hybrid Regime” (not quite authoritarian but not quite democratic).

    Canada, UK and US are the only 3 OECD nations using FPTP. All the rest use some form of proportional representation or hybrid thereof, save for Australia and France. Australia uses instant-runoff for its House and STV for its Senate. France uses two-round runoff voting.

    Within proportional representation, although most countries using it employ party lists, it does not by definition need to be done this way. The German state of Baden-Württemberg has a unique solution for this. They use the “best runner up” model to determine who fills the “list seats”. In Canada because of geographic population disparity, likely we’d have to base that on percentage of vote won, not vote count, but it could be done.

    I also have to disagree that MMP is by definition a difficult system to understand and explain. Any friend or family member I’ve ever explained it to understands it perfectly well. Besides, maybe other MMP systems in other countries are difficult to understand. But we could keep things simple. If people don’t want the ballot to change, don’t change the ballot. Still one vote, one candidate. Just that, like in Baden-Württemberg, a vote for a candidate counts for them AND their party. People in Canada rarely vote by local candidate anyway. Our system is very party focussed and will likely stay that way. If the ballot doesn’t change, and people still elect a representative in their riding who represents them, the method of apportioning seats seems rather esoteric to me. Most people if you told them that whoever gets the most votes in their riding wins, and in the end the House of Commons will have as many MPs for each party as their percentage of the vote dictates, that seems pretty simple to me.

    Finally, this isn’t so much a disagreement as it is a thought. I agree that in ranked voting the alphabetic advantage may pose a problem. But surely we must have the technology to randomly scramble name order. This may be difficult with printed ballots, but with new electronic voting technology it would be easy as coding the system to do so. https://www.usenix.org/conference/evtwote13/workshop-program/presentation/bell

    Overall, I really like the article. It is great particularly for those coming at this with little previous exposure.

Sign in to comment.