[EM] STV+AV

[Bryan Mills]

On Sat, Feb 4, 2012 at 3:21 PM, Jameson Quinn <jameson.quinn at gmail.com>wrote:

> 2012/2/4 Bryan Mills <bmills at alumni.cmu.edu>
>
>> > From: Bryan Mills <bmills at alumni.cmu.edu>
>>> > To: David L Wetzell <wetzelld at gmail.com>
>>> > > If there are 3-5 seats STV then the number of candidates won't
>>> > proliferate
>>> > > too much and there'd be 5-7 places to vote.  This would keep things
>>> > > reasonable.
>>> >
>>> > To get reasonable proportionality with only 3-5 seats per district
>>> > you'd probably need to go to an MMP system, with all its added
>>> > complexity.  Otherwise Droop proportionality doesn't buy you much over
>>> > FPTP; with 5 seats the Droop quota measures to a precision of ~17%,
>>> > and the remaining 17% in each district is still susceptible to
>>> > gerrymandering.
>>> >
>>>
>>> Not much?
>>> The goal here is not perfectionism wrt proportionality.
>>> The goal is to increase proportionality and to increase the number of
>>> competitive seats
>>> and to reduce the cut-throat competitive nature of US political rivalry
>>> between its two biggest parties
>>> so they can't dominate the other and have more incentives thereby to work
>>> together on the many issues that need work.
>>>
>>
>> I'm doubtful that 3-5 candidate districts actually would "increase the
>> number of competitive seats".  Each major party ends up with 1-2 safe
>> seats, and at that level of granularity gerrymandering and geographical
>> polarization are still significant enough to render the last seat
>> non-competitive in most districts.  (It would increase proportionality
>> somewhat - by transforming some of the safe-by-gerrymandering seats into
>> safe-by-Droop-proportionality seats - but you seem to be arguing that
>> proportionality isn't as important as competition.)
>>
>> Suppose we have two parties with a 50/50 split and 5 seats per district,
>> with one party more popular in urban areas and one more popular in rural
>> areas.  And suppose that the district lines are drawn such that 4/5 of
>> districts are slightly more rural than average and 1/5 of districts are
>> more urban than average, so that the 5th seat in each district becomes
>> relatively safe as well.  (We can do this fairly easily using geographical
>> boundaries by centering 1/5 of the districts around cities.)
>>
>> Scale that up to 400 legislators (80 districts).  What do we end up with?
>> 320 "natural" safe seats guaranteed by Droop proportionality (160 for
>> each party)
>> 80 gerrymandered-safe seats for the rural party
>> 20 gerrymandered-safe seats for the urban party
>>
>
> First, your numbers add up to 420. I think you meant 64/16 for the safe
> seats, which is only a 56/44 advantage, not 60/40.
>

Oops!  I had started calculating with 500, then switches to 400 and forgot
to update some of the numbers.  You're right, I should have had the
gerrymandered seats at 64/16, and that does give a 56:44 advantage rather
than 60:40.  (Of course, if you throw in some third-parties the whole
analysis changes too; the 50/50 example is meant to be representative of an
idealized two-party world.)

Note that the "safe" seats would still swing if there were a swing in
> national mood of around something less than 8%, not something less than 25%
> as in single-member districts. And the more highly-gerrymandered the map
> is, the tighter that margin, and so the greater the chances of it
> backfiring against the gerrymandering party. Gerrymandering is a fine art,
> but 8% doesn't leave a whole lot of room to play with. Considering safety
> margins and misfires, I doubt that the gerrymandering party could get
> anything close to the 6% representation advantage your (corrected) numbers
> suggest. So, while 2-3% unfairness is still a problem, I think it's a big
> step up from where we are.
>

Hmm, interesting.  So maybe that's not as much of a concern as I thought;
I'll have to give it some more consideration. I don't think it solves the
problem of multiple axes of policy preference, though, and 16% is a lot of
voters to leave unrepresented in the multiparty case.


Now, despite a 50/50 natural split, the rural party has a 60%
>> supermajority.  And, of course, if you draw the district lines differently
>> you can do the same thing for the urban party.
>>
>> So there's still relatively little hope that a system with such small
>> districts would produce a party-proportional legislature.  As you point out
>> elsewhere, it might still be possible to get an ideologically-proportional
>> legislature if you can get the parties themselves to shift ideologies.
>>
>>
>> > If you assume two major parties with ~40% of the electorate each, that
>>> > means that the 5th seat in each district is noisy -- but it's not
>>> > random noise, it's systematically biased by the parties' voting
>>> > strategies and the choice of district boundaries.  Larger districts
>>> > allow finer-grained Droop quotas and thereby reduce that noise.
>>>
>>> dlw: Smaller districts engender less opposition from those in power.
>>> They keep the constituent-legislator relationship more so.
>>>
>>
>> Absolutely agreed that smaller districts engender less opposition from
>> those in power.  That's because smaller districts don't fix the biases that
>> keep them in power.
>>
>> They do maintain the constituent-legislator relationship, *for the subset
>> of voters who voted in favor of the legislator*.  For the remaining Droop
>> quota of un- or under-represented constituents the nonexistence of the
>> constituent-legislator relationship is also maintained.
>>
>
> Here's my chance to plug PAL representation<http://wiki.electorama.com/wiki/PAL_representation>,
> which does PR but uses existing-sized districts and preserves a specific
> constituent-legislator relationship for all but (up to) one Droop quota of
> voters.
>

Hmm..   Interesting, but it seems too complicated to me.  It's got all the
complication of delegation, approval, and STV - with a bit less voter
burden on the approval side since only the candidates have to pick approval
thresholds - plus the variable quota and elimination procedure, which
doesn't have a strong intuitive interpretation to me.  Maybe if you can
find a way to simplify the counting algorithm - or a way to explain it with
a more intuitive connection - but otherwise I think it would be too
difficult to get adopted.


 > >> But if we assume that partial rankings are effective, there's still
>>> the
>>> > >> strategy/computation tradeoff to deal with: allowing truncated
>>> ballots
>>> > >> still doesn't help with favorite-betrayal, and STV variants less
>>> > >> susceptible to favorite-betrayal are also less susceptible to
>>> efficient
>>> > >> counting.
>>> > >>
>>> > >
>>> > > dlw: Truncated ballots may not end favorite betrayal, but it'll help
>>> with
>>> > > it.
>>> >
>>> > I don't see how; please elaborate.
>>> >
>>>
>>> This is essentially the same arg that IRV does not end the fact that some
>>> will still on occasion be pressured to betray their favorite.
>>> But it'll be of less consequence when it happens.  It won't be 3rd party
>>> dissenters, it'll be the supporters of a major party that does
>>> not position itself near the true political center who get pressured to
>>> betray their favorite and that in turn will pressure the major party
>>> to adapt or die.
>>>
>>
>> Are you saying that favorite-betrayal isn't a problem when those forced
>> to do it belong to a major party?  I hope I'm just misunderstanding your
>> point, but it sounds to me like you're describing a system like FPTP but
>> with major-party spoilers substituted for minor-party spoilers.
>>
>>
>> > >> With an implicit "first-preference" approval, it has the same
>>> problem as
>>> > >> traditional STV (i.e. IRV), namely of unduly rewarding
>>> > favorite-betrayal.
>>> > >> With an implicit "all-ranked" approval, the overall system would
>>> likely
>>> > >> violate later-no-harm with much higher frequency; by expressing a
>>> > >> preference between two dispreferred candidates one might
>>> unintentionally
>>> > >> put the higher of the two in contention.
>>> > >>
>>> > >
>>> > > dlw: I'd say empirically we'd see just how high of a frequency LNH
>>> would
>>> > be
>>> > > violated.  Jameson Quinn had a hard time coming up with a
>>> pathological
>>> > > example for IRV3/AV3 and I imagine it'd be similar for the above.
>>>  The
>>> > 1st
>>> > > stage would reduce the number of candidates to N+2 and it seems
>>> likely
>>> > that
>>> > > the N+2nd and N+3rd candidates in terms of "all-ranked" approval are
>>> less
>>> > > likely to be among the N winners.
>>> >
>>> > Hmm, ok.  I'm operating on the assumption that voters will vote
>>> > strategically if doing so is easy, and will vote approximately
>>> > honestly if strategic voting is difficult.
>>> >
>>>
>>> okay.
>>>
>>> >
>>> > We're taking the top S+k winners and running some ideal STV method on
>>> > them; let's try to find an "easy" strategy.  Here's my idea:
>>> > 1) Gather a set of related parties to form a majority-coalition.
>>> > 2) Have the coalition propose exactly S+k candidates.
>>> >
>>>
>>> good luck coordinating that..
>>>
>>> 3) Ask coalition voters to vote for all of the coalition candidates in
>>> > any order they choose.
>>> >
>>> > Since a majority of candidates approve of every coalition candidate
>>> > and disapprove of every competing candidate, the coalition candidates
>>> > win the approval vote.
>>> > By adding the "approval" phase to the STV election, I'm able to turn a
>>> > simple majority into a 100% supermajority.
>>>
>>>
>>> > Is there a flaw in my strategy?  (I don't think there is, but I may be
>>> > missing something.)  If not, we'll either need to abandon a fixed
>>> > limit on the number of candidates or we'll need something more
>>> > sophisticated than a simple approval-vote to filter them.
>>> >
>>>
>>> dlw: It's not realistic.
>>>
>>> You'd need to have serious intra-party discipline to keep the no. of
>>> candidates down to S+2
>>> and to get a majority of voters all to vote for all of that S+2
>>> candidates.
>>>  That is a serious coordination problem.
>>>
>>> But if it did happen then it'd "work" in terms of making the leading
>>> coalition of parties cast a broad net that strongly met the needs of most
>>> people.  This would be much better than a bunch of non-competitive
>>> single-winner elections.  In that case, we're in DINO land.
>>>
>>
>> By "strongly met the needs of most people" you appear to mean "met the
>> needs of a bare majority of people marginally better than the
>> alternatives".  My concern is that in this scenario 25% of the electorate
>> would benefit substantially, 25% would benefit marginally, and the
>> remaining 50% would be arbitrarily worse off.  That's essentially the same
>> worst-case behavior as the current majority-of-majorities setup, but with a
>> simpler strategy required to implement it.
>>
>> That being the case, I think we'd be better off with small-district STV
>> than with large-district STV with this sort of approval-based filtering.
>>
>>
>> > >> It may well be that these issues are all less severe than in the
>>> > >> deterministic alternatives to STV, but I still think they're enough
>>> to
>>> > >> merit consideration of nondeterministic alternatives.
>>> > >>
>>> > >
>>> > > In terms of the US's political culture, nondeterministic
>>> alternatives are
>>> > > not going to happen anytime in the near future and we need electoral
>>> > reform
>>> > > ASAP!!!!
>>> >
>>> > Sadly, I think both nondeterminism and STV share the "not going to
>>> > happen in the near future given political culture in the US"
>>> > classification, given that US law requires single-winner FPTP
>>>
>>> > elections for federal representation and the major parties (who
>>> > control the legislature and benefit greatly from FPTP) have no
>>> > incentive to change that law.
>>>
>>> dlw: STV need not end 2-party domination.  Reforms that do not end
>>> 2-party
>>> domination are more fit in the US and should be the only ones pushed.
>>> And, as I've shown, it's implementation can be simplified.
>>> Thus, it can become a  political jujitsu issue, whereby it is more
>>> rational
>>> for those in power to accommodate than to resist the proposed change.
>>>
>>
>> The belief that the 2-party system can accurately reflect voter consensus
>> relies heavily on the assumption that voters' differences of opinions
>> correlate sufficiently well with a single dimension of variability, so that
>> tending toward the center along a single axis produces centrist results on
>> all issues.  I do not accept that assumption: in my experience, Americans
>> disagree along at least two axes that do not correlate perfectly (fiscal
>> policy and social policy).
>>
>>
>> > So as far as I can tell the only option for meaningful reform is a
>>> > constitutional amendment, and that means reforming 75% of the states
>>> > as a first step.  This is not a short-term process.
>>
>>
>>> I think one could argue that the current law requiring single-winner
>>> elections is discriminatory twds minorities, and adopted under bad
>>> circumstances, and thereby unconstitutional.  This would not require a
>>> constitutional amendment.
>>>
>>
>> I think you're perhaps overly optimistic about the willingness of courts
>> to overturn election law.  But we'll see - I'd be thrilled to be proven
>> wrong about this one.
>>
>> ----
>> Election-Methods mailing list - see http://electorama.com/em for list
>> info
>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.electorama.com/pipermail/election-methods-electorama.com/attachments/20120204/9f4bfab2/attachment-0004.htm>