[EM] STV vs Party-list PR, could context matter?

[David L Wetzell]

I give a rebuttal to the Electoral Reform Society's assessment of
party-list PR for the case of 3-seat LR Hare.
http://anewkindofparty.blogspot.com/2011/05/electoral-reform-society-united-kingdom.html


dlw

On Fri, Feb 17, 2012 at 2:54 PM, David L Wetzell <wetzelld at gmail.com> wrote:

>
>
> From: Richard Fobes <ElectionMethods at VoteFair.org>
>> To: election-methods at electorama.com
>> Cc:
>> Date: Fri, 17 Feb 2012 12:01:16 -0800
>>
>> Subject: Re: [EM] STV vs Party-list PR, could context matter?
>> On 2/17/2012 6:49 AM, David L Wetzell wrote:
>>
>>> ...
>>>
>>> It seems to me that most folks think the choice is between ranked
>>> choices or party-list PR.  ...
>>>
>>>
>>> So what do you think?
>>>
>>
>> I don't see this as an either/or choice,
>
>
> dlw: U2 apparently are not among most folks...
>
>
>> nor do I see a viable "both" option being suggested.
>>
>
> dlw: viability is a low-blow at this stage, but I guess it's a blow I use
> quite often.
>
>>
>> So I'll again suggest VoteFair ranking:
>>
>> VoteFair ranking uses "ranked choices" (1-2-3 ballots and pairwise
>> counting...) for identifying the most popular candidate -- for filling the
>> first seat in a legislative district.
>>
>> VoteFair ranking fills the second district-based seat with the
>> "second-most representative" candidate.  In the U.S., even without asking
>> voters to indicate a party preference, that would usually be the most
>> popular candidate from the opposite party (i.e. the opposite party compared
>> to the first-seat winner).
>>
>> To further increase proportionality, VoteFair ranking fills some
>> proportional seats based on the favorite party of the voters. (Whichever
>> party has the biggest gap between voter proportion and filled-seat
>> proportion wins the next seat.)
>>
>> We don't have to choose between proportionality (PR) and ranked methods.
>>  We can get both.  And in a U.S.-compatible way.
>>
>> If election-method reform is to happen in the U.S., it has to merge with
>> the reality of the two-party system.  And I believe it should accommodate
>> third parties only to the extent that voters are unable to regain control
>> of the two main parties.
>>
>
> dlw: I agree with the reality of the 2-party system.  I also believe that
> we need to make the case that our 2-party system will work much, much
> better if we give 3rd parties a constructive role to play in it.  Giving
> them access to one-third of the seats in the state assembly so they get to
> determine which major party is in power in that body every two years is
> such a constructive role.  It will give folks more exit threat from the two
> major parties, thereby making both of them more responsive to the moving
> center.
>
>>
>> As for STV, going beyond two seats easily produces unfair results.  And
>> in the U.S. the results also would be quite unstable (i.e. not mesh well
>> with the current two-party system).
>>
>
> Can you elaborate?
> I don't see why 3-5 seat STV with a droop quota wouldn't have results like
> what you described that would maintain yet transform the US's 2-party
> system.
>
> dlw
>
>>
>> Richard Fobes
>>
>>
>>
>>
>> ---------- Forwarded message ----------
>> From: Jameson Quinn <jameson.quinn at gmail.com>
>> To: David L Wetzell <wetzelld at gmail.com>
>> Cc: election-methods at lists.electorama.com
>> Date: Fri, 17 Feb 2012 14:35:58 -0600
>> Subject: Re: [EM] JQ wrt SODA
>> If first-mover is all that counts, then I'm afraid we're stuck with
>> plurality. Obviously, I hope and believe that's not true.
>>
>> Jameson
>>
>> 2012/2/17 David L Wetzell <wetzelld at gmail.com>
>>
>>> IRV's got a first mover advantage over SODA and to catch up you need to
>>> convince someone like Soros to help you market it.  It wouldn't matter if
>>> you got the whole EM list to agree with you that it was hunky-dory.
>>>
>>> But in the context of a 2-party dominated system, there aren't as many
>>> serious candidates and so what relative advantages there are of SODA over
>>> IRV will be less, which then makes the first-mover marketing problem more
>>> significant, especially if IRV can be souped up with the seemingly slight
>>> modification of the use of a limited form of approval voting in the first
>>> stage.
>>>
>>> dlw
>>>
>>> On Fri, Feb 17, 2012 at 12:27 PM, <
>>> election-methods-request at lists.electorama.com> wrote:
>>>
>>>> Send Election-Methods mailing list submissions to
>>>>        election-methods at lists.electorama.com
>>>>
>>>> To subscribe or unsubscribe via the World Wide Web, visit
>>>>
>>>> http://lists.electorama.com/listinfo.cgi/election-methods-electorama.com
>>>>
>>>> or, via email, send a message with subject or body 'help' to
>>>>        election-methods-request at lists.electorama.com
>>>>
>>>> You can reach the person managing the list at
>>>>        election-methods-owner at lists.electorama.com
>>>>
>>>> When replying, please edit your Subject line so it is more specific
>>>> than "Re: Contents of Election-Methods digest..."
>>>>
>>>> Today's Topics:
>>>>
>>>>   1. Re: Real-world examples of chicken dilemma?
>>>>      (Kristofer Munsterhjelm)
>>>>   2. STV vs Party-list PR, could context matter? (David L Wetzell)
>>>>   3. SODA arguments (Jameson Quinn)
>>>>   4. Re: STV vs Party-list PR, could context matter? (Jameson Quinn)
>>>>   5. Re: STV vs Party-list PR, could context matter? (James Gilmour)
>>>>   6. Re: Question about Schulze beatpath method (Markus Schulze)
>>>>
>>>>
>>>> ---------- Forwarded message ----------
>>>> From: Kristofer Munsterhjelm <km_elmet at lavabit.com>
>>>> To: Jameson Quinn <jameson.quinn at gmail.com>
>>>> Cc: EM <election-methods at lists.electorama.com>
>>>> Date: Fri, 17 Feb 2012 07:51:10 +0100
>>>> Subject: Re: [EM] Real-world examples of chicken dilemma?
>>>> On 02/15/2012 08:46 PM, Jameson Quinn wrote:
>>>>
>>>>> As I've said before, I'm writing a paper on SODA and the chicken
>>>>> dilemma. I'd appreciate any real-world examples of the dilemma.
>>>>> Obviously, since a true chicken dilemma is not possible with either
>>>>> plurality, runoffs, or IRV, I'm looking for cases that arguably would
>>>>> have been a chicken dilemma under approval. That means that the two
>>>>> "vote splitting" factions would almost certainly have clearly preferred
>>>>> each other to the opposing faction, but there was still enough bad
>>>>> blood
>>>>> and a close enough balance that they could easily have failed to
>>>>> cooperate. I'd say HI-01-2010 qualifies as a good example; US-Pres-2000
>>>>> doesn't, because many of the Nader voters affirmed that they would not
>>>>> have voted for Gore, and anyway, Gore won both the popular vote and the
>>>>> most self-consistent counts of Florida.
>>>>>
>>>>
>>>> Wouldn't the Burr dilemma count? That *was* Approval. Granted, it was
>>>> used to elect more than one candidate, but you could argue the property
>>>> would remain in a singlewinner context.
>>>>
>>>>
>>>>
>>>>
>>>>
>>>> ---------- Forwarded message ----------
>>>> From: David L Wetzell <wetzelld at gmail.com>
>>>> To: EM <election-methods at lists.electorama.com>
>>>> Cc:
>>>> Date: Fri, 17 Feb 2012 08:49:08 -0600
>>>> Subject: [EM] STV vs Party-list PR, could context matter?
>>>> It seems to me that a common sense solution would be to base which gets
>>>> used on the propensity for voters to be informed about the elections.
>>>>
>>>> Also, the two types seem to be bundled with different types of quotas.
>>>>  STV gets marketed with the droop quota here in the US.  I'm not
>>>> complaining because it's good to simplify things.  But if STV were bundled
>>>> with Droop then 3-seat LR Hare might prove handy to make sure that 3rd
>>>> parties get a constructive role to play in US politics.
>>>>
>>>> So I propose that 3-5 seat STV with a droop quota, perhaps using AV in
>>>> a first step to simplify and shorten the vote-counting and transferring
>>>> process, for US congressional elections or city council elections and
>>>> 3-seat LR Hare for state representative and aldermen elections.  The latter
>>>> two elections are less important and get less media coverage and voter
>>>> attention.  Is it reasonable to expect voters to rank multiple candidates
>>>> in an election where they often simply vote their party line?  Why not keep
>>>> it simple and use the mix of Droop and Hare quotas to both keep the
>>>> system's duopolistic tendencies and to make the duopoly contested?
>>>>
>>>> It seems to me that most folks think the choice is between ranked
>>>> choices or party-list PR.  I think it is a matter of context and that both
>>>> can be useful, especially when no explicit party-list is required for a
>>>> 3-seat LR Hare election.  The vice-candidates who would hold the extra
>>>> seats a party wins could either be selected after the victory or specified
>>>> before hand.
>>>>
>>>> So what do you think?
>>>>
>>>> I'm keeping the seat numbers down because I accept that those in power
>>>> aren't going to want an EU multi-party system and I'm not sure they're
>>>> wrong about that, plus the US is used to voting the candidate and having
>>>> their representative and they could keep that if there are relatively few
>>>> seats per election.
>>>>
>>>> dlw
>>>>
>>>>
>>>> ---------- Forwarded message ----------
>>>> From: Jameson Quinn <jameson.quinn at gmail.com>
>>>> To: EM <election-methods at lists.electorama.com>,
>>>> electionsciencefoundation <electionscience at googlegroups.com>
>>>> Cc:
>>>> Date: Fri, 17 Feb 2012 09:20:20 -0600
>>>> Subject: [EM] SODA arguments
>>>> For those who feel that Bayesian Regret is the be-all-and-end-all
>>>> measure of voting system quality, that SODA's BR for 100% strategic voters
>>>> will beat all other systems, including Range/Approval.
>>>>
>>>> For those who feel that Condorcet compliance is the be-all-and-end-all,
>>>> a majority Condorcet winner, or any Condorcet winner with 3 candidates and
>>>> full candidate preferences, is not just the winner with honest votes, but
>>>> in all cases the strategically-forced winner; this contrasts with Condorcet
>>>> systems, in which strategy can cause even majority- or 3-candidate- CWs to
>>>> lose.
>>>>
>>>> For those who feel that strategic resistance is the most important,
>>>> SODA is unmatched. It meets FBC, solves the chicken dilemma, has no burial
>>>> incentive (ie, meets later-no-help), and even meets later-no-harm for the
>>>> two most-approved candidates (where it matters most). It's monotonic, and I
>>>> believe (haven't proven) that it meets consistency. It meets participation,
>>>> cloneproofness, and IIA for up to 4 candidates.
>>>>
>>>> For those "middlebrows" who most value a system's acceptability to
>>>> current incumbents, SODA is top-of-the-line. It allows voters to vote
>>>> plurality-style and, if two parties are clearly favored by voters, allows
>>>> those two parties to prevent a weak centrist from winning, even if
>>>> polarization is so high that the centrist is an apparent Condorcet winner.
>>>>
>>>> For those who want simplicity: while it's true that the SODA counting
>>>> process is more complicated than approval, the process of voting is
>>>> actually simpler than any other system, because you can just vote for your
>>>> favorite candidate. For the majority who agrees with their favorite
>>>> candidate's preferences, there is no strategic need to watch the polls and
>>>> figure out who the frontrunners are, and no nail-biting dilemma of whether
>>>> to rank others as equal to your favorite.
>>>>
>>>> And for those who balk at delegation, SODA allows any voter to cast a
>>>> direct, undelegated ballot; and allows those voters who do delegate to know
>>>> how their vote will be used. Refusing to consider SODA because you don't
>>>> want to delegate, is like refusing to walk into a candy store because you
>>>> don't like chocolate; SODA allows, not requires, delegation.
>>>>
>>>> I think pretty much everybody on this list falls into one or more of
>>>> the above categories. So, what's not to like about SODA?
>>>>
>>>> Jameson
>>>>
>>>> ps. I clarified the SODA procedure<http://wiki.electorama.com/wiki/SODA_voting_(Simple_Optionally-Delegated_Approval)#Full.2C_step-by-step_rules> on
>>>> the wiki, though there were no substantive changes. I improved the
>>>> formatting, marked the steps which are optional, and better explained that
>>>> winning candidates use their delegated votes first because precisely
>>>> because they will probably choose not to approve others.
>>>>
>>>> Comments are welcome.
>>>>
>>>>
>>>> ---------- Forwarded message ----------
>>>> From: Jameson Quinn <jameson.quinn at gmail.com>
>>>> To: David L Wetzell <wetzelld at gmail.com>
>>>> Cc: EM <election-methods at lists.electorama.com>
>>>> Date: Fri, 17 Feb 2012 09:26:33 -0600
>>>> Subject: Re: [EM] STV vs Party-list PR, could context matter?
>>>>
>>>>>
>>>>> It seems to me that most folks think the choice is between ranked
>>>>> choices or party-list PR.
>>>>>
>>>>
>>>> I don't. I think that party-list removes voter freedom, and ranked
>>>> choices is too much of a burden on the voter. While either would be better
>>>> than what we have, I prefer to use delegation a la SODA.
>>>>
>>>> Thus my favored system is PAL representation<http://wiki.electorama.com/wiki/PAL_representation>.
>>>> It's true that PAL still has some (very attenuated) party-list-like
>>>> aspects, because party affiliation is used to match candidates to districts
>>>> at the end; but if you were willing to give up this (overlapping)
>>>> geographical representation aspect of PAL, you could make a similar
>>>> delegated PR system in which parties played no explicit role.
>>>>
>>>> Jameson
>>>>
>>>>
>>>>
>>>> ---------- Forwarded message ----------
>>>> From: "James Gilmour" <jgilmour at globalnet.co.uk>
>>>> To: "'David L Wetzell'" <wetzelld at gmail.com>, "'EM'" <
>>>> election-methods at lists.electorama.com>
>>>> Cc:
>>>> Date: Fri, 17 Feb 2012 18:01:55 -0000
>>>> Subject: Re: [EM] STV vs Party-list PR, could context matter?
>>>> But why would you want all these differences and complications?
>>>>
>>>> If you are going to use STV-PR for some of these elections, why not use
>>>> STV-PR for all of these elections to the various
>>>> "representative assemblies" (councils, state legislatures, US House of
>>>> Representatives, US Senate).  STV-PR works OK in both
>>>> partisan and non-partisan elections, so it should give fair and proper
>>>> representation of the VOTERS in all these different
>>>> elections.
>>>>
>>>> Of course, with districts returning only 3 to 5 members, the
>>>> proportionality and direct representation MAY be a little limited, but
>>>> if small numbers are needed to make the system acceptable to the vested
>>>> interests, then so be it.  STV-PR with 3, 4 or 5 member
>>>> districts is greatly to be preferred to plurality in single-member
>>>> districts and to plurality at large.  We had to accept local
>>>> government wards electing only 3 or 4 councillors as part of our STV-PR
>>>> package  -  that's practical politics.  But that reform has
>>>> transformed our local government  -  no more "one-party states".
>>>>
>>>> James Gilmour
>>>>
>>>>
>>>> > -----Original Message-----
>>>> > From: election-methods-bounces at lists.electorama.com
>>>> > [mailto:election-methods-bounces at lists.electorama.com] On
>>>> > Behalf Of David L Wetzell
>>>> > Sent: Friday, February 17, 2012 2:49 PM
>>>> > To: EM
>>>> > Subject: [EM] STV vs Party-list PR, could context matter?
>>>> >
>>>> >
>>>> > It seems to me that a common sense solution would be to base
>>>> > which gets used on the propensity for voters to be informed
>>>> > about the elections.
>>>> >
>>>> > Also, the two types seem to be bundled with different types
>>>> > of quotas.  STV gets marketed with the droop quota here in
>>>> > the US.  I'm not complaining because it's good to simplify
>>>> > things.  But if STV were bundled with Droop then 3-seat LR
>>>> > Hare might prove handy to make sure that 3rd parties get a
>>>> > constructive role to play in US politics.
>>>> >
>>>> > So I propose that 3-5 seat STV with a droop quota, perhaps
>>>> > using AV in a first step to simplify and shorten the
>>>> > vote-counting and transferring process, for US congressional
>>>> > elections or city council elections and 3-seat LR Hare for
>>>> > state representative and aldermen elections.  The latter two
>>>> > elections are less important and get less media coverage and
>>>> > voter attention.  Is it reasonable to expect voters to rank
>>>> > multiple candidates in an election where they often simply
>>>> > vote their party line?  Why not keep it simple and use the
>>>> > mix of Droop and Hare quotas to both keep the system's
>>>> > duopolistic tendencies and to make the duopoly contested?
>>>> >
>>>> > It seems to me that most folks think the choice is between
>>>> > ranked choices or party-list PR.  I think it is a matter of
>>>> > context and that both can be useful, especially when no
>>>> > explicit party-list is required for a 3-seat LR Hare
>>>> > election.  The vice-candidates who would hold the extra seats
>>>> > a party wins could either be selected after the victory or
>>>> > specified before hand.
>>>> >
>>>> > So what do you think?
>>>> >
>>>> > I'm keeping the seat numbers down because I accept that those
>>>> > in power aren't going to want an EU multi-party system and
>>>> > I'm not sure they're wrong about that, plus the US is used to
>>>> > voting the candidate and having their representative and they
>>>> > could keep that if there are relatively few seats per election.
>>>> >
>>>> > dlw
>>>> >
>>>>
>>>>
>>>>
>>>>
>>>> ---------- Forwarded message ----------
>>>> From: "Markus Schulze" <Markus.Schulze at alumni.TU-Berlin.DE>
>>>> To: election-methods at lists.electorama.com
>>>> Cc:
>>>> Date: Fri, 17 Feb 2012 19:27:05 +0100
>>>> Subject: Re: [EM] Question about Schulze beatpath method
>>>> Hallo,
>>>>
>>>> it can happen that the weakest link in the strongest path
>>>> from candidate A to candidate B and the weakest link in the
>>>> strongest path from candidate B to candidate A is the same link,
>>>> say CD.
>>>>
>>>> I recommend that, in this case, the link CD should be declared
>>>> "forbidden" and the strongest path from candidate A to candidate B
>>>> and the strongest path from candidate B to candidate A, that does
>>>> not contain a "forbidden" link, should be calculated. If again the
>>>> weakest link in the strongest path from candidate A to candidate B
>>>> and the weakest link in the strongest path from candidate B to
>>>> candidate A is the same link (say EF), then also this link should
>>>> be declared "forbidden" and the paths from A to B and from B to A
>>>> should be calculated. This should be repeated until the weakest
>>>> link in the strongest path from A to B and the weakest link in the
>>>> strongest path from B to A are different links.
>>>>
>>>> Markus Schulze
>>>>
>>>>
>>>>
>>>> _______________________________________________
>>>> Election-Methods mailing list
>>>> Election-Methods at lists.electorama.com
>>>> http://lists.electorama.com/listinfo.cgi/election-methods-electorama.com
>>>>
>>>>
>>>
>>> ----
>>> Election-Methods mailing list - see http://electorama.com/em for list
>>> info
>>>
>>>
>>
>> _______________________________________________
>> Election-Methods mailing list
>> Election-Methods at lists.electorama.com
>> http://lists.electorama.com/listinfo.cgi/election-methods-electorama.com
>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.electorama.com/pipermail/election-methods-electorama.com/attachments/20120217/66992536/attachment-0004.htm>