July 11, 2019 at 11:30 am #26643
This is coming from the discussion of Greece on the mailing list.
So, when I did my talks for the referendum, I’d say that there are proportional countries (like Ireland, Germany and the Scandinavian countries), winner take all countries (like Canada, France and the United Kingdom), and then semi-proportional systems that are some mix of the two. I’d then say that some of those mixes are well thought out, and others not so well thought out. I think we can leave it at that unless pressed, and if pressed Greece is a great example of a poorly thought out electoral system.July 11, 2019 at 12:49 pm #26644
With a general public audience, I think Ryan’s approach is best.
For serious discussions with political scientists and those designing or evaluating electoral rules, the usage of the Gallagher Index makes for greater clarity.
For that reason, I’ll cross-post here part of what I wrote to the pr_discussion_group earlier today:
“With regard to definitions of “proportional”, so far there is no commonly accepted standard that a certain average Gallagher Index (GI) is or isn’t “proportional”, but I’d like to see one gradually develop to replace the current criterion, which as you know is simply based on the structure of the electoral system and not on the actual outcomes of elections in practice.
“This latter approach fails to give a clear answer in the many countries, like Greece, that have little add-ons to what would otherwise be a PR system and which muddy the waters, making it difficult to tell if they have PR simply from their electoral rules.
“A few people have recommended a GI of 5 or less, which is equivalent to an excellent level of proportionality. That’s a good goal, while others feel 7 or less is very good, but those goals are not a means to define “PR”.
“However, if you look at the various lists compiled of countries which are usually considered by the experts to have “proportional representation” and then look at their GI’s you will see that countries with an average GI above 9 are not called PR, while countries called “PR” have a GI under 9. There’s a clear cut border at 9. However, that doesn’t mean the GI can be used willy-nilly to identify PR in reverse, as the case of the U.S.’s low GI shows. (In the whole world there are only a few such exceptions where the average GI doesn’t identify PR.) Becker’s Composite GI [Byron Weber Becker] is much better, and I don’t know if there are any cases in which it fails to classify PR.
“With regard to the low G.I. (low apparent disproportionality) in the USA, it’s not any lack of primaries within the minor American political parties that causes the U.S. GI to be low, but rather the fact that the primaries serve, psychologically, as a run-off, and, once those results are publicized, everyone knows that with rare exceptions only the Republicans and Democrats have a chance, so they get almost all votes; this rigid concentration on just two parties in turn forces those parties to set policies designed to attract half the vote, which in turn, averaged over the whole USA, results in a low GI (but not a low composite GI): There is nothing in their system to discourage the two main parties from setting policies designed to catch a bit more than half the votes even at the expense of doing poorly in some states.”
In spite of the few exceptions such as the USA, the Gallagher Index (least squares index promoted by Michael Gallagher) remains the best and most commonly used quantitative measure of proportionality of the dozen or so indices that have been tested.July 11, 2019 at 4:22 pm #26645
Ah, I see, Northern Ireland reduced district magnitude after the 2016 election.
Still, the proportionality is pretty fair. Maybe use a small balance seat system if Canada, by nature of it’s large geography has to use relatively small district magnitudes using the best runner ups.
As for France, I’d also point out that executive power is split between the Parliament and the President with a smaller role for the prime minister than is seen in Canada, although the president is far more limited than the American president and cannot dismiss the government or cabinet members, needs the countersignature of the PM for a good chunk of things, the PM generally creates a coalition with a relatively diverse, at least by the standards of majoritarian systems, set of parties and the PM and cabinet members cannot be members of the parliament, the PM can be chosen from outside any given party and the president is free to select who they want to be PM although they know that the Assembly can dismiss them at any time, the president has a very limited suspensive veto where they can order a second vote but nothing more, and is most influential with foreign policy and the military.
Local mayors are also elected in a runoff system with the local candidates elected in what as far as I can tell is a majority bonus system with two rounds, and legislative power is also split. Even candidates from the same party for president and the local officials can differ, and the local candidates in particular can, because of the runoff, often be trying to appeal to different coalitions. The president of France, Macron, won by creating a coalition united against Marie Le Pen of the Front National, which majorly shook up the two party system which had dominated a lot of French politics before this.
However, in individual parliamentary constituencies, the candidates from the same party often had to contend with different types of coalitions, some needing the support of more rightist voters to defeat a centre left coalition and vice versa, given that REM is fairly centrist, it can swing either way.
- This reply was modified 1 week, 3 days ago by Robert Jarman.
You must be logged in to reply to this topic.