Objectives of Proportional Representation

Home Forums Fair Vote Discussion Forum Objectives of Proportional Representation

Viewing 15 posts - 1 through 15 (of 48 total)
  • Author
  • #25948
    David Nash

    I believe that we need to define the objectives of proportional representation in detail, so that we can measure the effectiveness of the various systems in meeting these objectives. There may be some contention about these objectives. We need to be upfront about these issues, and hopefully resolve them.

    I leave it open to others to start the ball rolling

    Robert Jarman

    One would be to make it so that more people actually have voted for one of the people who end up elected, or at least put it in the top ranking if we use STV.

    Another would be to make it so that a candidate or party is rewarded when they become more popular and vice versa. The Alberta Party in 2019 got way more votes but lost all their seats in the provincial legislature.

    On a similar topic, the amount of votes it takes on average to elect a member of the legislative body should be more similar. The Greens needed 603 thousand votes to win one seat. The Liberals got about 11.5 times that many votes but got 184 times as many seats, both figures from 2015 federal election.

    The caucuses of each party should also be more diverse. The Liberals have almost no caucus from Alberta or Saskatchewan, and the Conservatives have no seats or MPs from the territories or the Maritimes. This would also go for the committees in the parliament itself as well, and the cabinet. If the Reform Act is applied to a caucus, then it also goes for the election of the caucus chairs and votes of confidence in the party leadership as well. If other reforms like the way that the New Zealand Labour caucus elects their share of cabinet ministers, this also means that the caucus in Canada would be more diverse and so elect a more interesting and useful range of cabinet ministers.

    There would also be a reason for many people to join the constituency associations of many more areas and become a party member, and likely people would be able to join a party they agree with more given that in a proportional system, more have a chance of being elected and getting seats. This would shape their AGMs, their constituency boards, those they nominate as candidates, their delegates to conventions and those who attend party AGMs/congresses/conventions, and the votes during votes of confidence in the leadership, and in the case of parties with some form of a standing council or committee in addition to an executive board like the NDP and the communist party, also shape the representation on that. Take a look at how many people join parties and what kind of people join constituency associations.

    A possible measuring system could also be the filling in of vacancies. For an MMP or list system, a vacancy on the list can be filled by the next candidate on the list, ranked using an open list, this would be the next most voted candidate on that list. For the local candidate in MMP; if we use a ranked ballot to elect the local candidates, then we could use the countback method. For STV, you can use the countback method and within a few days of the vacancy in either case get a vacancy filled. If neither of those methods work, such as if the candidates have moved on and found other jobs or the list is exhausted, one option would be to let the board of the constituency association of the party in MMP or a party list system elect with a ranked ballot a replacement. If none of that works, then a ranked ballot could be used to fill the vacancy in a byelection, but that would become a lot less common. It would probably also limit a party in terms of trying to deliberately have their candidates resign so as to let someone else in like a party leader who gets defeated in their own riding or is a new leader.

    It would also be harder to regionally concentrate rewards and penalties because the caucus is not as concentrated from specific areas and likely excluded from major areas. Swing ridings will not be very important although they can have significance in some ways such as if a small party depends on winning a single seat in an area to get list seats under the same rules for MMP as New Zealand, but the result on the whole would be more important.


    Im not sure Davids Question is framed correctly because PR is not just one system and any objectives we come up with are going to be valued differently by different people.

    For example, I value individual voting choice and selection with the most fair counting system possible even if its complex or done by a computer. For others they might value fairness of party seats in regards to the popular vote over anything else.

    I think this is why Citizen Assemblies are important because they discuss all the possible objectives of PR and then determine how different people in different locations place thier values.

    I think this would be a much more useful discussion if we could come up with a list of all possible objectives in a table and then rank with a percentage how closely a specific PR system aligns with that Value. Then you can reach out to the people and hold polls to figure out what systems are favoured where by what margins.

    You could also ask why do some people support some objectives over the others, how could a new system be designed to alleviate fears or rise up to what people want?

    David Nash

    I had been thinking in much more general terms, some of my thoughts are listed here.

    First and foremost, if we are to get support for PR, we have to propose a system that people understand. I think it follows that it must be a fairly straight-forward system. The process of voting has to be absolutely clear, and it follows that it must be as simple as reasonable.
    The process of counting is less central but it has to be possible to encapsulate in a clear description.
    The actual physical process of counting needs only be defined in a subtext.

    Second most important is that we must define how the system functions to give better representaion. In my mind there are two aspects: A. How does the system guarantee that opinions are represented in proportion their prevalence in the society? B.If it does so, how does the system guarantee that there is representation of regional opinion and differences in regional opinion?

    Not necessarily third most important, but certainly important because it is the central nub of many objections to PR, is the question of limiting the proliferating small parties, often with the implied notion that small parties will be extremist and destabilising. I think this is important from a Public Relations point of view.

    I believe that we need to be able to define the role of political parties in the system, both what it should be and how it will be modified by the system we propose.

    This is my own particular hobby horse, so it should have no place in my present contribution, but I do think it needs to be introduced into the discussion. Any reform of the electoral system should be concurrent with a reform of the process criteria by which Candidates are selected.

    Robert Jarman

    Some factors that determine the success or lack thereof of a PR system also has to do with elements just beyond the way that the Commons and a legislature are simply elected. A prime minister with the ability to call elections basically whenever is still going to be able to screw with votes of confidence and limit the success, and so that is why it was important when in 2011, the Fixed Term Parliament Act in the UK limited the PMs ability to call new elections without a motion of no confidence, and this lack of ability to do that was a part of how the LibDem and Conservative coalition was stable whereas in Canada, minority governments have had difficulties.

    In Canada, I also imagine that a fair number of governments under a PR system would be confidence agreements, not always true coalitions, and so the influence on the cabinet itself can be more limited than would be true in a real coalition, barring other aspects in the confidence agreements like changes to relevant laws that would restrict the cabinet and enhance the parliament as a whole.

    Proportional representation cannot exist as an isolated factor in the way our system runs. We also have to study other factors into this. New Zealand was useful because it was not in combination with that many other reforms at the same time, although parties updated their constitutions and bylaws to update the rules for how the lists were to be assembled.

    David Nash

    I think that it goes without saying that the prime objective, is of course, proportionality.

    jim in oakville

    I have seen 5% mentioned before a party would be entitled to a proportional seat.
    With 338 seats I feel the threshold for 1 seat should be 0.3%
    With an eligible voting population of 26 million and a turnout of 80% (we got 79% 3 times and PV is supposed to increase our participation) .3% is 62,400 votes. I don’t believe a PR system should ignore that many voters.

    I look forward to your replies.

    Robert Jarman

    That would not be correct unless we have a single nationwide constituency like the Netherlands, Slovakia, or Israel, or where the list seats are allocated in a single constituency at large in an MMP fashion like New Zealand.

    But constitutionally, every MP must be from a specific province or the territories, and realistically, we are going to be dividing Canada into multi member constituencies. I can only imagine areas even approaching the point where a threshold would become relevant would be in the largest of the largest cities like Toronto, and even they are likely to be subdivided. Scotland is divided into regions with 7 members of the Scottish parliament who are chosen from a closed list which supplements in an MMP style any single member constituencies which in the region will number between 7 and 9, for a total of 14 to 16.

    If we define the size of the constituency, for MMP this is the total for both list and local seats, as N, because you logically would have to get at least 1/Nth the vote to win a seat (or a close fraction of it), only if the number of seats exceeded 20 would a threshold of even 5% become relevant, as 5% is equal to 1/20. If you get 5%, you will get 1 seat, and if you do not have at least 5%, you would not get a seat.

    Really, we could even do without a threshold at all unless perhaps for some provincial and municipal MMP systems like in PEI where they may use a single at large proportional list region, where we might use a threshold, but even then, it probably would not have a large effect.

    jim in oakville

    Robert, I think you missed my point.
    If a party gets 0.3% of the votes they should get .3% of the seats or 1 out of the 338 seats.
    5% of the votes should equal 5% of the 338 seats or 17 seats.

    Feel free to contact me on Facebook or email
    Jim Hemphill
    [email protected]

    David Nash

    Your arithmetic is correct, Jim, but not your knowledge of the way most successful PR systems work. Indeed, you are advocating, by implication, a system which has been chaotic in some instances, and one which does rightly draw criticism from many opponents of proportional representation, on the ground that it would result in the election of representative of too many small political factions. Unfortunately those critics ignore the alternatives that are available and operating successfully in many parts of the world.

    To be fair, it does work in some places (for example, I believe in the Netherlands) but it is less than effective in Israel, for example. It seems to depend on whether the country is a fairly stable and uniform in its social objectives. It is, constitutionally impossible in Canada, as well as, in all probability, being disruptive to such a large and diverse country.

    Both of the most widely discussed systems proposed for this country, STV in multi-member constituencies, and MMP, avoid this pratfall, although they would most definitely ignore the votes of small minorities. They both have the advantage that they avoid the false pattern of regional political division normally produced by our “first past the post” system and more important avoid its fatal flaw, the election of majority governments without majority support, which is the common case these days.

    David Nash
    780 435 5488
    3512 117A Street NW
    Edmonton AB
    [email protected]

    Robert Jarman

    Very few places have a single nationwide constituency. Not that many places have electoral districts that are so large that a 5% threshold would become relevant, and are more commonly found in places that are huge metropolitan areas like Rio or in Bagdad, or when using a hybrid form where some seats are chosen in one form and the others are chosen in a list based system like Ukraine, Mexico, and New Zealand which all have a single nationwide constituency for the list seats only.

    If you divide a country into multi member regional constituencies, a likely outcome if Canada adopts MMP in any form, it would become necessary to win a droop quota or hare quota (depending on the exact math of the system) to win a seat, and that would apply within each region. Only if a region exceeded 20 people from the sum of the list and local seats would a threshold of 5% exclude a party that would otherwise win a seat.

    jim in oakville

    Not to seem argumentative, but the Green Party has only made that 5% threshold once in ten elections and I would certainly consider them a serious party. But only in the last 3 elections have they won even 1 seat.
    In 2008 they got 6.8% which would have got them 21 seats proportionally not the single seat they got.

    How would you justify dropping them from 21 PR seats in 2008 to just one seat in 2011 because they dropped from 6.8% to 3.9% and still pretend to be proportional. This has to be allowed for or we open our PR systems to legitimate criticism. We have to stop playing into the hands of the No side.

    It brings up also a problem for a regional party like the Bloc Québécois that have 10 seats but only ran in Quebec yet got 4.7% of national vote. Would a PR system make any allowance for regional parties in Quebec or the Prairies?

    David Nash

    Jim, I know you have a point, but so do the people who reject proportional representation on the basis that every group of people who can raise roughly 60,000 votes (or less if turnout is low) should be represented in Parliament. After all, the Nazis rose to power under the “pure” form of PR that your arguments are based on. They have a valid point which can be made by pointing to clearly dysfunctional possibilities in a fragmented parliament.

    I would guess that most provinces would return at least one Green and some (BC Ontario and Quebec) would return several under STV or MMP. I’d like to see someone do a simulation based on previous voting patterns. Remember, there are also gains for the Greens to be had under a proportional system, since people will be free to vote positively for the party they favour, rather than looking over their shoulders to see which candidates they wish to defeat, so called “strategic” voting.

    Robert Jarman

    Invoking the Weimar Republic is a bad argument for a number of reasons. The Weimar republic was not a parliamentary system, it was the first (as far as I know) semi presidential system. A well designed such system can lead to power sharing but nobody really knew just what a semi presidential system should do at the time. We know in hindsight and looking at what has happened in most semi presidential systems with similar distributions of power as was the case in the Weimar Republic what kind of powers worked and what failed.

    The President of the Deutschereich was directly elected in a two round system. If nobody had a majority on round one, a second was held, however they did not formally establish the rule that it was a two two system and so sometimes candidates failed to get a majority in the second round. The president had a 7 year term, without a term limit I might add, and he could veto legislation which needed a 2/3 vote to override. He also appointed the chancellor unilaterally, and importantly, could dismiss the chancellor unilaterally. The Reichstag could dismiss the chancellor too, a power they did not have during the time of the Kaiserreich, but this left the chancellor with dual loyalties to the president and to the parliament rather than a tool of power sharing in both. The president also had pretty broad power to call an election without many limits.

    The President also had vast emergency powers, in article 48. This is a provision with few of the modern limits on power that we might know. It was used to intimidate the opponents of the Nazis during a number of the last multi party, but not truly free, elections in the Weimar republic. There were also not provisions to protect the constitution from violence like many modern constitutions, which often forbid dissolutions of the legislature during emergencies, forbid amendments to the constitution or any laws that have the same weight as one (as the infamous Enabling Act did) during times of emergency, forbid changes to the electoral laws during times of emergency, require a 2/3 vote of the legislature to sustain the emergency or else losing effect within a few weeks to a couple of months maximum of the emergency being declared, requiring the President to have the countersignature of a prime minister whom the president cannot dismiss without a vote of no confidence succeeding, and stronger limits on what rights cannot be limited even during an emergency. Hindenburg did not just use it as a form of the Reichstag Fire Decree, he had consistently been using it since 1930, while the Nazis had almost no seats.

    Also, Hindenburg died in 1934, just two years into his term. Had he lived until the end of his term, he may have provided more resistance to the worst of the powers of the Nazis. The president dying in office is not supposed to happen, and even when it does, a new election is supposed to be held to elect a new one. Hitler seized the office of president and combined them into a single position as the Führer in a rigged referendum.

    There were also few contingency provisions that many modern constitutions have to limit negative effects of poor electoral outcomes. Modern Germany has a runoff system to elect the chancellor if a majority coalition cannot be formed. Germany also requires that votes of no confidence also come with a candidate to replace the chancellor and only if the current chancellor gets a majority against and the new candidate gets a majority in favour can a replacement occur. A dissolution of the Bundestag can only happen when a vote of confidence in the chancellor, either at the beginning of the term after an election (ergo requiring runoff ballots) or at some point down the line, produces a majority against the chancellor but no majority in favour of someone else, and the president has the decision as to whether to call one or permit a minority government. Many other countries have other limitations like needing say 20% of the legislature to initiate a vote of no confidence and often making it so that if the motion fails, they cannot attempt another motion during the same session, making smaller parties unable to arbitrarily use their power without good reason.

    Modern semi presidential systems, designed to promote cooperation between a president and parliament, often specify something along the lines of the president shall nominate a prime minister from among the likely parliamentary majority and the parliament shall vote on it. If approved, they are the prime minister. If not, the speaker of the parliament or any other group of members of parliament nominate someone who they believe is likely to retain the confidence of the parliament. If this candidate is not approved by a majority, a new election is held.

    The veto of the president also made it quite hard for standard laws to be passed, which would have created a much more formalized system of law that would have provided some stability without regard as to who specifically was in power. A lower threshold, say needing an absolute majority to override, with the ability to introduce amendments or a line item veto on non monetary bills (provided the presidential decree powers and military powers were limited), would have been a much better compromise for the fractured politics at the time.

    The proportional thresholds are also a risk in that once a party gets enough support, and as the Brexit Party has shown us, this can happen with brand new parties too if the atmosphere is conductive to it, the thresholds imposed by laws can leave potentially genuine and moderate opposition, even though small, out of the picture. Having thresholds would have for example limited the ability of nationalist parties in Wales to fight for seats, most nationalist parties of which oppose the most risky party.

    You also are not accounting for the massive economic collapse that was not the fault of the population of Germany, on multiple occasions. Hyperinflation to pay off war debt assigned by the surrender compact with the Entente in 1923, the November Revolution and the near starvation imposed by British blockades in 1918, the Great Depression, and the countless strikes and attempts at violent overthrow, including the Beer Hall Putsch and the March Action which was a communist revolution that was truly massive on a scale most people today in Canada cannot understand. All this just a few years after one of the most devastating wars in human history and a devastating influenza pandemic that killed millions of people. Very few countries with any form of government, proportional or not, can survive challenges like these effectively. Even moderate politicians in the Weimar Republic hated the limits imposed by the Treaty of Versailles, which imposed crippling economic limits and distributed much of the corporate and industrial wealth out of Germany and limited the military to the point where it had a hard time putting down that communist revolution in the March Action, and so helped to have it ignored and undermined in secret (such as sending tank trainers to the Soviet Union in the late 1920s), which evaded limits that would have constrained the war machine of the future Nazi Germany had they been observed.

    David Nash

    Robert, I have to suppose that your analysis of the situation in Germany in the 1930s is a response to my one liner: “After all, the Nazis rose to power under the “pure” form of PR that your arguments are based on.”

    I enjoyed your detailed analysis of the history of the origins of Nazi power. It my well be that you are entirely correct in implying that “pure” PR was by no means the only, or even a major factor, in the rise of Nazism. I did not claim that it was. I just commented that the first Nazis were elected under such a system, which I believe to be true, although I am open to correction on that, too.

    What is true is that many people believe that a “pure” system can lead to the rise of extreme parties. It is true also that substantial proportionality can be produced by the design of the electoral systems so as to exclude very small political factions. I for instance disagree that a movement that can garner only 60,000 votes or less from the entire country necessarily should be represented in parliament.

    In contrast, I believe that fair provision must be made for the election of independents, even those running as representatives of a tiny party. However, I believe that such individuals should be subject to the requirement to obtain a threshold of support at a relatively local level, from people likely to have a personal knowledge of the individual’s potential as a representative. This would, I am sure apply to Elizabeth May. And if we Had had a PR system in past elections, I guess we would have had several more Green Party MPs in the past.

Viewing 15 posts - 1 through 15 (of 48 total)
  • You must be logged in to reply to this topic.