Voting System Reform Commission: Part 2

[Peter]

This thread will contain the discussions of Part 2 of the VSRC.

Our mandate in this is as follows:

"The Commission will discuss any and all systems of voting that they feel would be appropriate for use in Atlasian federal elections, including the current system of preferential voting."

First of all we will establish a list of voting systems that we will discuss. I can think of the following off hand:

FPTP
FPTP with runoff (the French/Louisiana system)
Preferential Voting (the present system)
Approval Voting (Southeast System)
Approval Voting with Runoff
Condorcet Voting
Borda Count

Any other suggestions will be considered.

[Emsworth]

That seems to be an exhaustive list of the feasible voting systems. I would just add that Condorcet voting is not actually one system; rather, the term covers several different systems that satisfy the Condorcet criterion (see here).

[ilikeverin]

That seems to be an exhaustive list of the feasible voting systems. I would just add that Condorcet voting is not actually one system; rather, the term covers several different systems that satisfy the Condorcet criterion (see here).

I think to start out we should just have 'some sort of Condorcet voting'

That seems to be an exhaustive list of the feasible voting systems. I would just add that Condorcet voting is not actually one system; rather, the term covers several different systems that satisfy the Condorcet criterion (see here).

We should probably lump them all together so it's less complicated.  If we choose to do Condorcet (yay!), then we can decide which kind of Condorcet voting we want to do.

[Joe Republic]

I hope the Secretary of Forum Affairs will be open to considering other voting systems, as well as the ridiculously complicated Condorcet system.

[ilikeverin]

I hope the Secretary of Forum Affairs will be open to considering other voting systems, as well as the ridiculously complicated Condorcet system.

Yes, but I prefer making my life a living hell

[Bono]

I'm not a comission member, but may I sugest adding Range Voting to your shortlist?

[Bono]

Also Coombs' Method, Bucklin Voting and Supplementary Vote.

[Peter]

We will consider all of these, though I have to say that Bucklin Voting looks quite a bizarre way to run an election.

[ilikeverin]

Here's a fun idea: get an idea of which system is the best (it wouldn't be binding, just give everyone an idea of what all the systems are) by running the election for which system is best using all the methods

[Colin]

You know it would have been alot easier if you had just refered everyone to this page first:

http://en.wikipedia.org/wiki/Voting_systems

[Bono]

You know it would have been alot easier if you had just refered everyone to this page first:

http://en.wikipedia.org/wiki/Voting_systems

Yes it would have been.
Yep, you're right.

Sure.

[Peter]

We will consider all of those options presented on the Wikipedia page under the Single Winner Voting Systems list.

I motion to automatically reject: Exhaustive Ballot (Papal conclave system), Motion and Amendment, and Random Ballot. This motion is considered to be carried unless there is an objection.

Our method of consideration will be as follows:

Basically, I will say what voting system we are presently considering, possibly with a brief synopsis of what happens in it (which due to my gross laziness will probably be copied from Wikipedia). We will then proceed to discuss the voting system, and hopefully develop a consensus as to whether the voting system is appropriate for Atlasia or not.

When we have finished all on our list, we will then return to consider in greater detail all of those we considered to be appropriate.

We will start with First Past The Post (FPTP). The system needs no explaination as it is the one that most of us are used to in the real world.

[Colin]

Well I would like to rule out FPTP automatically. First of all it breeds a two party system which I feel would be detrimental to Atlasia. Secondly in our current political automosphere where we have 5 or 6 parties running for the Presidency we can end up with the winner recieving only around 20% and still being able to win. This, of course, would not show the true intentions of the voters and would, generally, lead to more coalition building and more fusion tickets. It would generally reduce the amount of voices in our political atmosphere as well as re-create a two or three party system. I, for this reason, feel that FPTP, even with a runoff vote though I'll explain my reasons later, would be unacceptable for Atlasia.

And on the motion to table those three systems I fully agree with the chair.

[Emsworth]

I motion to automatically reject: Exhaustive Ballot (Papal conclave system), Motion and Amendment, and Random Ballot. This motion is considered to be carried unless there is an objection.

I have no objection.

I agree with the sentiments expressed by the President pro tempore, and concur that pure FPTP is not appropriate in our case.

[Sam Spade]

I don't like FPTP either.

[Peter]

Given that nobody has spoken up in its favour, I assume we have consensus to reject FPTP as inappropriate to Atlasia.

Next we have: Two-round runoff voting. This is better known to some of us as Second Ballot and is used in French Lower House and Presidential elections, as well as most elections in Louisiana. Effectively it consists of a first round of pure FPTP, with a second round runoff between the two highest finishers of the first round.

[KEmperor]

I strongly support the FPTP with Runoff system.  I feel it has the best of both worlds.  In the first round, you can vote for whoever you like, and if over 50% of the people agree with you, that person wins.  But if no one is popular enough to win outright, you have a runoff with the top two vote getters. I think this leads to greater legitimacy when people win the election, because they still would have gotten over 50% of the second round, whereas in our current system people often win with 20% of the vote.

Also, this eliminates the problem of a high preference for someone actually hurting then.

[Emsworth]

I have no objection to FPTP with Runoff, and agree with the comments of the Chief Justice.

[Colin]

I disapprove of the FPTP with a runoff system although not to the extent that I disapprove of the pure FPTP system. Whilst it takes care of some of the faulty point about FPTP it also maintains many of the worst parts about FPTP. In the last election if we had a run off between first round winners it would have been an election of Harry vs. Phil. I don't know about you but I would have a hard time even getting myself to vote in that runoff if it was those two candidates. With FPTP with a runoff it would lead to the building of two or three large coalitions. While we still have a multi-party system, especially on the Regional and Sentorial level, in Presidential elections we would be at the whims of endorsements, coalitions, and back door politicing in order to get enough of a coalition to get first or second place in the first round vote. As it stands now parties like the CDP, the SDP, and even the UP and the ILP would have problems getting to the first or second place threshold unless they have very strong candidates. This leads to smaller parties coalitioning with other parties and leading back to a two or three party system.

Also FPTP with a runoff brings up the possibility of a result like the 2002 French presidential elections would occur. We could have a very strong extremist candidate get enough of a percentage of the vote, especially if the left, right or centre are devided, to get into the second round. As this showed in the 2002 election in France this lead to an overwhelming, over 80%, win for Jacques Chirac, an election that he didn't deserve to win with such large numbers, because of a split in the left.

Overall whilst I feel that this is a better system than FPTP I do not think it should be implemented in Atlasia.

[Peter]

I have mixed feelings regarding this electoral system. It has advantages and disadvantages like all of the systems we will consider. That said I think it could be appropriate in an Atlasian context, and thus I feel it is worthy of greater consideration later, though I do feel that other systems may be more appropriate.

Also, this eliminates the problem of a high preference for someone actually hurting then.

Whilst voting for somebody may not hurt them, it may not be the best thing you can do for your preferred candidate.

Consider the following election line-up and suppose for sake of argument that I am a Joe Republic supporter and that I am to be the last voter

Joe Republic - 30 votes
Ernest - 20 votes
AuH2O - 20 votes

Clearly Joe will go through to the last round facing either Ernest or AuH2O or perhaps both if they tie. The best thing I can do as the last voter is to vote for AuH2O to ensure that Joe has him as opposition. Tactical voting will remain just as prevalent as it is now.

[ilikeverin]

Right, what the two people above me said

[Sam Spade]

I am actually rather neutral to FPTP with runoff voting.  It is probably second-best to the preferential system presently in place (third would be approval voting), so I wouldn't shed any tears were it enacted.

The comments above basically echo my support and objections to FPTP with runoff voting.

[Peter]

There seems to be at least some interest in FPTP with Runoff, so we shall certainly consider that later in our more in depth considerations.

Next on the list is Elimination Runoff, one which I should have rejected earlier for much the same reasons as Exhaustive Runoff. It is considered rejected unless there is an objection.

Next on the list is Approval Voting (without Runoff). It is described rather aptly by Wikipedia as: Voters may vote for as many candidates as they like. Candidate with most votes wins.

This system is presently used by the Southeast Region for Gubernatorial elections.

[Joe Republic]

Would I be allowed to recommend that this Commission discuss another system of voting?  I'm not sure if this already exists by another name, but I've decided to call it 'scaled approval voting'.

The basic premise is a cross between preferential voting and simple approval voting.  Voters simply vote for the candidates they approve of, while also giving them a rating according to an arbitrary scale.

For example, let's say that an election has five candidates: A, B, C, D and E.  The rating scale would be out of 10, for argument's sake.  As a hypothetical voter, I consider Candidate C to be my favorite candidate, while Candidate D is my least favorite.  I consider Candidates B and E to be equally as good as each other, but only about half as good as C.  Candidate A is not considered a good candidate, but would still be preferred over Candidate D.  On my ballot, I list every candidate (including those I disapprove of), and give them a rating out of 10 (approve = 10; disapprove = 0).  So my hypothetical ballot would be as follows:

C: 10
B: 5
E: 5
A: 1
D: 0

These scores are then tallied up with other ballots, and we take it from there.  I haven't decided how we decide the result from those tallied scores yet, so I guess I'll leave that up to you guys.

[Bono]

Would I be allowed to recommend that this Commission discuss another system of voting?  I'm not sure if this already exists by another name, but I've decided to call it 'scaled approval voting'.

The basic premise is a cross between preferential voting and simple approval voting.  Voters simply vote for the candidates they approve of, while also giving them a rating according to an arbitrary scale.

For example, let's say that an election has five candidates: A, B, C, D and E.  The rating scale would be out of 10, for argument's sake.  As a hypothetical voter, I consider Candidate C to be my favorite candidate, while Candidate D is my least favorite.  I consider Candidates B and E to be equally as good as each other, but only about half as good as C.  Candidate A is not considered a good candidate, but would still be preferred over Candidate D.  On my ballot, I list every candidate (including those I disapprove of), and give them a rating out of 10 (approve = 10; disapprove = 0).  So my hypothetical ballot would be as follows:

C: 10
B: 5
E: 5
A: 1
D: 0

These scores are then tallied up with other ballots, and we take it from there.  I haven't decided how we decide the result from those tallied scores yet, so I guess I'll leave that up to you guys.

That's range voting.