Stress Test for a Proposed Electoral System

For the serious discussion of weighty matters and worldly issues. No off-topic posts allowed.

Moderators: Azrael, Moderators General, Prelates

kevbo86
Posts: 3
Joined: Mon Mar 13, 2017 3:13 pm UTC

Stress Test for a Proposed Electoral System

Postby kevbo86 » Tue Mar 14, 2017 12:55 am UTC

They call it Local Proportional Representation (modified version of STV). They have the support of several prominent Member of Parliament from all parties in Canada. It is not exactly like anything currently used as far as I know and so I am concerned about flaws/vulnerabilities. Curious to know where I can go to get a bunch of math/tech folks to take a look and purposefully try to break/manipulate the proposed system to expose any flaws/vulnerabilities.

Feel free to join a thread of the same title I created on the "voting" section of reddit.
Last edited by kevbo86 on Thu Mar 16, 2017 2:09 am UTC, edited 1 time in total.

User avatar
Qaanol
The Cheshirest Catamount
Posts: 3024
Joined: Sat May 09, 2009 11:55 pm UTC

Re: Stress Test for a Proposed Electoral System

Postby Qaanol » Wed Mar 15, 2017 7:05 pm UTC

I’m interested in voting reform and I’d like to take a look.

You can post a link after you’ve made, I believe, 5 posts on the forum.
wee free kings

kevbo86
Posts: 3
Joined: Mon Mar 13, 2017 3:13 pm UTC

Re: Stress Test for a Proposed Electoral System

Postby kevbo86 » Thu Mar 16, 2017 2:00 am UTC

I understand. Post number 2! :P

kevbo86
Posts: 3
Joined: Mon Mar 13, 2017 3:13 pm UTC

Re: Stress Test for a Proposed Electoral System

Postby kevbo86 » Thu Mar 16, 2017 2:07 am UTC

Alternatively, join a thread of the same title I started in the "Voting" section of reddit.

User avatar
Qaanol
The Cheshirest Catamount
Posts: 3024
Joined: Sat May 09, 2009 11:55 pm UTC

Re: Stress Test for a Proposed Electoral System

Postby Qaanol » Thu Mar 16, 2017 6:16 pm UTC

The link is http://localpr.ca/details/. I haven’t read it yet though.
wee free kings

User avatar
aldonius
Posts: 105
Joined: Sun Jan 17, 2010 2:33 am UTC

Re: Stress Test for a Proposed Electoral System

Postby aldonius » Tue Mar 21, 2017 11:11 am UTC

This is weird. Why would you have 'local' MPs if they can be elected by people from outside your riding? If you're willing to coalesce ridings into a region, why not just use some variant of STV or MMP?

Key idiocy IMHO: as a candidate, you can lose the election even if your party wins an absolute majority in your riding. @kevbo86, you identified this in your Reddit thread, good job over there.

There's a worked example under FAQ>Fairness. Four ridings, four MPs, 100 voters per riding. Conservatives got 210/400 votes (and an outright majority in Ridings B, C & D). Liberals 96/400. NDP 94/400. Quota 81 votes. The NDP win Riding A, because their candidate there got the preferences from the worse-performing NDP candidates in B, C & D. Presumably the votes for the Liberal and Conservative candidates in Riding A get redistributed back to those in B, C and B, C & D, respectively. The Liberals have a quota overall, so to satisfy proportionality they have to win either B or C, overriding the Conservatives (who got more than double the Liberals primary there).

For crying out loud, it's hard enough convincing people that having two classes of MPs is OK with MMP (specifically, that a candidate who lost a local election can come in on the list).

User avatar
Thesh
Made to Fuck Dinosaurs
Posts: 5247
Joined: Tue Jan 12, 2010 1:55 am UTC
Location: Colorado

Re: Stress Test for a Proposed Electoral System

Postby Thesh » Thu Mar 23, 2017 4:04 pm UTC

aldonius wrote:This is weird. Why would you have 'local' MPs if they can be elected by people from outside your riding? If you're willing to coalesce ridings into a region, why not just use some variant of STV or MMP?


It is a variant of STV, which is similar to and arguably more desirable than MMP. And the reason is for the same reason as MMP: politics. They want proportional representation, but they don't want more MPs, and they want local representation, and they want to vote on candidates and not parties. Using straight STV without increasing the MPs results in ridiculously large districts that leaves the whole idea of local representation behind for many. Too small of districts, and they become less proportional unless you reduce the number of seats. MMP, on the other hand, makes the party itself more important than the candidates.

This provides many of the benefits of STV, while ensuring that everyone has a local representative without the need for increasing the number of MPs.

aldonius wrote:Key idiocy IMHO: as a candidate, you can lose the election even if your party wins an absolute majority in your riding. @kevbo86, you identified this in your Reddit thread, good job over there.


That's by design. The majority of the riding doesn't become the important thing - the riding is to ensure everyone has a local representative, but the objective is to ensure that as a whole they are proportionally represented.
Honesty replaced by greed, they gave us the reason to fight and bleed
They try to torch our faith and hope, spit at our presence and detest our goals

Chen
Posts: 5075
Joined: Fri Jul 25, 2008 6:53 pm UTC
Location: Montreal

Re: Stress Test for a Proposed Electoral System

Postby Chen » Thu Mar 23, 2017 8:07 pm UTC

Thesh wrote:That's by design. The majority of the riding doesn't become the important thing - the riding is to ensure everyone has a local representative, but the objective is to ensure that as a whole they are proportionally represented.


Do the people really have a local representative if the person they, in majority, vote for doesn't necessarily end up representing them? That seems at best to be following the letter of what they want (local representatives) rather than the spirit of it.

User avatar
Thesh
Made to Fuck Dinosaurs
Posts: 5247
Joined: Tue Jan 12, 2010 1:55 am UTC
Location: Colorado

Re: Stress Test for a Proposed Electoral System

Postby Thesh » Thu Mar 23, 2017 8:49 pm UTC

Depends on how you define "local representative" - in this case, and the case of MMP, this just means that there is someone that was elected from your local area. If you agree that proportional representation is more important than the majority of each district having an honest-to-god representative, than this is a decent compromise. If you think that someone who wins a majority of a running should win the majority, then in many circumstances it becomes indistinguishable from first past the post; if all of the runnings in a district have the same majority party, should you expect that one party to get 100% of the seats? No, of course not, because then the district as a whole wouldn't be representative of the people there.
Honesty replaced by greed, they gave us the reason to fight and bleed
They try to torch our faith and hope, spit at our presence and detest our goals

reval
Posts: 59
Joined: Fri Sep 23, 2016 2:56 pm UTC

Re: Stress Test for a Proposed Electoral System

Postby reval » Fri Mar 24, 2017 1:54 pm UTC

Trying to square winner-take-all with proportional representation keeps not working. Winner-take-all gets you a local representative who may actually care what you think, but if you're in the local minority, you get nothing. Proportional gets you a national voice for minority positions, even they can't do much, but it also gets you all-powerful party machines that trample individual representatives into dust. Trying to square the two systems usually results in party dominance.

In the German system (mentioned at localpr.ca as an MMP example) you get two votes. The first vote elects your local representative, who should have a plurality in the district. The second vote goes to a party list. Half of the seats are local, and the other half are by party list, chosen to make the whole thing as proportional as possible. (Subject to a 5% minimum threshold designed to prevent the proliferation of 1 and 2% parties that helped bring down the Weimar Republic.)

Should work, in theory. In practise: the parties run everything. The leaders have safe seats high up on the party list, so it doesn't matter if they lose their districts. Your local rep doesn't listen to you, but does what the party says. There are occasional defectors.

The current proposal is trying to group ridings into regions, and you're supposed to get some proportionality at the regional level. But the vote transfers are too complicated. Here's the vote transfer pseudocode from localpr.ca:

Code: Select all

while (some seats remain unfilled) {
  if (a hopeful candidate, C, reaches quota) {
    C.state = Elected
    for all hopeful candidates, H, in Cs riding
        H.state = NotElected
    transfer any surplus votes to remaining hopeful candidates
  } else {
    let C be the unprotected candidate having the fewest votes
    if (C is the last candidate in their riding)
      C.protected = true
    else {
      C.state = NotElected
      transfer Cs votes at full value to remaining hopeful candidates
    }
  }
}


Rather than trying for proportionality at the regional level - at the cost of transparency - it might be interesting to see a two-house system where one house is winner-take-all and the other is proportional. For example: taking the useless US Senate and make it truly proportional at a national level, while leaving the US House as winner-take-all districts.
Last edited by reval on Fri Mar 24, 2017 4:07 pm UTC, edited 1 time in total.

User avatar
aldonius
Posts: 105
Joined: Sun Jan 17, 2010 2:33 am UTC

Re: Stress Test for a Proposed Electoral System

Postby aldonius » Fri Mar 24, 2017 3:53 pm UTC

reval wrote:The leaders have safe seats high up on the party list, so it doesn't matter if they lose their districts.


If you're willing to commit to a preferential ballot, then mixed-member proportional is possible with a single ballot:
  • Run a preferential-voting local-district election as per normal (be that IRV or something more sophisticated).
  • There is no party list. All candidates must run in a district, and of course parties must run a candidate in a district to get any votes there.
  • Impose some electoral threshold (I suggest the reciprocal of the number of local seats); all votes for independents and groups getting less than the threshold are distributed to more successful groups.
  • Allocate additional seats for proportionality, possibly taking account of preferences as listed to handle fractional remainders of seats.
  • I suggest a variable-size parliament with a maximum number of additional seats. Use the smallest number of additional seats such that all groups are allocated at least as many seats proportionally as they won locally. (If that isn't possible, have the maximum-size parliament.)
  • Select the additional MPs from the best losers. If using IRV, we'll do it not by primary vote but by maximum proportion of voteshare after preferences (i.e. at point of elimination). For a direct Condorcet system the correct metric would probably be aggregate for and against.
  • As such, candidates are always incentivised to do as well as possible locally - a 47:53 loss might well be enough to get in as a 'best loser', especially if their party has enough close losses to be 'under-represented'.

reval wrote:Rather than trying for proportionality at the regional level - at the cost of transparency - it might be interesting to see a two-house system where one house is winner-take-all and the other is proportional. For example: taking the useless US Senate and make it truly proportional at a national level, while leaving the US House as winner-take-all districts.


This is how we do it in Australia, both federally and in four of our six states.

reval
Posts: 59
Joined: Fri Sep 23, 2016 2:56 pm UTC

Re: Stress Test for a Proposed Electoral System

Postby reval » Fri Mar 24, 2017 4:17 pm UTC

I like the variable-sized parliament - but with only enough extra seats to get up to some minimal standard of proportionality. The extra seats would be evanescent, weakening party control, and the "best losers" would have to stay focused on their home districts. On the other hand, the "closest" districts would effectively get additional representation, while more monolithic districts would lose out. Hmm.

...a two-house system where one house is winner-take-all and the other is proportional.

aldonius wrote:This is how we do it in Australia, both federally and in four of our six states.

How's that going? Is it doing anything thing to weaken the control of the parties over individual representatives? Any signs of life from the individual reps?

User avatar
Thesh
Made to Fuck Dinosaurs
Posts: 5247
Joined: Tue Jan 12, 2010 1:55 am UTC
Location: Colorado

Re: Stress Test for a Proposed Electoral System

Postby Thesh » Fri Mar 24, 2017 5:01 pm UTC

reval wrote:Is it doing anything thing to weaken the control of the parties over individual representatives?


Why would you expect it to? Hell, just look at how partisan things are in the US... How you elect isn't the problem (although it makes it worse if you don't have PR), it's that the houses of government are designed around the idea of having a ruling party within those houses. They really only need procedures such as using petitions to bring bills to a vote, but instead we have artificially added hierarchy. The majority of the majority party nominate their candidate for positions of power within parliament, and the majority party ensures they win. Thus, the entire parliament becomes run by the majority of the majority. Not to mention that everything has to go through committees in which a handful of people are granted disproportionate influence over the legislative process, allowing them to make massive demands for pork.

If you don't have proportional representation, it's considerably worse because it becomes the majority of the majority of the plurality who elected candidates from a short list that's been narrowed down in a fashion that further limits democracy, and amplifies the influence of money and the limits candidates to those who are able to take a good amount of time off of work to campaign and fund raise (esp. in competitive districts).
Honesty replaced by greed, they gave us the reason to fight and bleed
They try to torch our faith and hope, spit at our presence and detest our goals

djangochained
Posts: 15
Joined: Mon Feb 27, 2017 1:01 pm UTC

Re: Stress Test for a Proposed Electoral System

Postby djangochained » Tue Mar 28, 2017 1:19 pm UTC

Seems interesting, will take a look after

Flintstone
Posts: 17
Joined: Wed Mar 29, 2017 12:08 am UTC

Re: Stress Test for a Proposed Electoral System

Postby Flintstone » Wed Apr 19, 2017 7:13 am UTC

If i could post links, I would link to an interesting youtube series where they go through several different voting systems and their benefits and problems. I think it was... Grey CCP or something? No, it was CGP Grey who did the animal series. It's easy to find if you're interested.

I'm afraid I haven't understood the proposed benefits of this system described. Often, even a county can have different political wants from the next county over (there bright red counties even in California). What is the benefit to combining such groups?


Return to “Serious Business”

Who is online

Users browsing this forum: No registered users and 4 guests