[EM] Approval Voting and Long-term effects of voting systems

[Daniel LaLiberte]

Thanks very much for your extensive responses.  Much for me to dig into and
think more about.

First I'd like to reiterate a couple points, hoping to get more feedback on
them.

One of the things I want to see more discussion about, and I have failed to
receive much feedback on this anywhere I bring it up, perhaps because it is
just too dang hard, is that we need to investigate and understand the
long-term effects of applying each voting system.  Let me broaden this a
bit to understanding how a voting system works within the context of the
rest of the constitution and the society as a whole as they co-evolve.  And
to do this, starting from the relatively simple criteria for evaluating
voting systems that we still seem to have lots of disagreements about, can
we somehow leap beyond that narrow view to say much at all about how the
voters, candidates, and society as a whole will change in response to the
constraints imposed by the voting system?  We need to try to do this,
right?

Maybe consider an analogy with weather and climate.  Weather is what
happens in each election.  Climate is what happens over many elections.

I imagine even plurality was expected to be a good enough voting system,
but despite the intention by at least some of the framers of the US
constitution to avoid political parties, they emerged as a dominant force
anyway.  Should they have known this was likely to happen?  Could they have
known with a little study?  Can we do better now?

I'd like to claim, and help other people understand, that Approval Voting
is not just one of the best voting systems but unquestionably far and away
the best, at least for current society and technology, both of which are
still surprisingly, um, easily confused.  But I say this not just because
of the characteristics voting systems in each single election, but also
because I am projecting how the repeated application of voting systems
changes things.  And I like where Approval Voting appears likely take us.

And this leads to the second main point that I believe needs some more
discussion, which is the context for deciding on which voting system we
want to use based on where it is likely to take us and, moreover, where we
*want* to go.  I believe the voting system is not a small matter, not just
a fine detail that can be addressed later when considering how a
constitution may play out, but it is a fundamental building block, at the
level of an atom, and what we can build out of it depends critically on
which atoms we have available to us.  So we need to consider both extremes
together, both the goal of what we want to build, and given that, how do we
expect to get there and what can we start with that will allow us to get
there.

Perhaps this is too much to discuss in this list that is focused on
election methods.  Maybe we don't need to debate where we want to go, which
admittedly is a huge subject, but I think we should want to understand
where we are likely to go based on the voting system(s) we choose.

And maybe not right now, but when it comes up, when it is relevant in
future discussions.

A few more comments are embedded below.


On Mon, Nov 21, 2016 at 2:51 AM, Michael Ossipoff <email9648742 at gmail.com>
wrote:

>
> Daniel--
>
> Well-said.
>
> Short answer: You're right. There are a lot of reasons why Approval is the
> best.
>
> As you suggested, people in this country are subject to the
> "lesser-of-2-evils problem". With Approval, even if someone feels a need to
> fully support a bad compromise like Hillary Clinton, at least s/he has no
> reason to not also approve hir favorite, Jill.Stein.
>

Maybe compromise is always an essential ingredient.  That is, maybe it can
never enough to just order your preferences in isolation from everyone
else.  Instead, we have to be at least a little bit strategic, guessing
where your favorite candidate is relative to other candidates in the view
of other voters.   And this compromise should not be viewed as a bad thing
but a good thing.


> Some people complain that Approval prevents that person from expressing
> their preference for Jill over Hillary.
>
> What they're missing is that there are some people who, with any method
> that fails FBC, would feel forced to rank Hillary alone in 1st place, just
> as you said.
>
> FBC (Favorite Burial Criterion) is a criterion that says:
>

aka Favorite-Betrayal - I had to look it up :)


> To top-vote a candidate means to not vote anyone over hir, and to vote hir
> over someone.
>
> Say you top-vote some candidates, and one of them would win. But then you
> move another candidate up to top, and the result is that now _none_ of your
> top-voted candidates wins (including the new one), and now the winner is
> someone you didn't top-vote.
>
> FBC says that shouldn't happen. Raising an additional candidate to top
> shouldn't change the winner from someone you top-voted to someone you
> didn't top-vote.
>
> In other words, it should be completely safe to top-vote Favorite, without
> fear that doing so could change the winner from (top-voted) Compromise to
> Worst.
>

FBC does seem to me to be a criteria we must not violate.  Without it, we
are essentially punishing people for expressing their favorite.

It's obvious that Approval meets FBC. Overcompromisers badly need a voting
> system that meets FBC.
>

>From the term, it sounds like "overcompromisers" are doing something
wrong.  Do you mean them to be just an element of society, one end of the
spectrum from over- to under- compromisers?  Acknowledging the reality that
voters are not ideal shoppers will help us move forward.


> Approval isn't the only method that meets FBC, but it's of course by far
> the most easily-implemented voting system that's any good.
>

And that is very important.  Given the difficulty of even getting people to
understand Approval, I will be overjoyed when we are finally are using it.


> There are other reasons why Approval is the best:
>
> If you're a progressive (and everyone in the 99% would benefit
> tremendously from an honest, progressive government), then there are two
> blatantly, sharply-distinct sets of candidates: Progressives and
> Republocrats. It's a matter of honest & progressive, vs corrupt, dishonest
> and corporate-owned.
>
> That's an instance of what I call a "strong top-set" and a "strong
> bottom-set", defined as follows:
>
> A strong top-set and a strong bottom-set are two sets of candidates such
> that the merit difference _between_ the sets is incomparably more important
> than the merit difference _within_ the sets.
>
> In other words the merit difference within the sets is entirely negligible
> in comparison to the merit difference between the sets.
>
> Your goal, then, is just to elect from your strong top-set.
>
> That's innomparably more important than choosing among your strong
> top-set, or trying to keep one strong bottom-set member from winning
> instead of another strong bottom-set member.
>
> ...and of course it's the strong top-set that you prefer to the strong
> bottom-set.
>
> Approval could be called "Set-Voting". It's a perfect match for what you
> want to do: Maximize the probability of electing from your strong top-set.
>
> Approval asks the right question.
>
> For the person who has a strong top-set and a strong bottom-set, it's
> possible to say what the objective best method is: It's Approval.
>

What about for people who do not have such a clear distinction between the
top and bottom sets?  Where they draw the line is a harder decision and
will probably factor in the likelihood that various candidates will win.


> Now, some people want to have it both ways. They want to elect from their
> strong top-set, but they also want the luxury of choosing _among_ that set.
>
> It doesn't work that way. You can't have it both ways.
>
> If a voting-system allows you a choice between casting  an effective
> approval-set vote, or ranking the candidates in order of preference , and
> you choose the latter, then you're increasing the probability of electing
> from your strong bottom-set
>

I assume you mean ranking candidates honestly.  So the choice with a
ranking system seems to come down to:

1. Rank honestly and risk electing from the strong bottom-set, or
2. Rank strategically by ranking one of the leading candidates in the
top-set higher than your favorite, and risk that your favorite might not
win.

But what happens if you can rank a leading candidate and your favorite the
same?  Then you avoid that dilemma, right?  This is what happens, in part,
with Approval Voting at least regarding those two candidates.  So let me
add this third alternative.

3. Rank one of your preferred leading candidates and your favorite
candidate the same, and avoid the above risks.

And if this is true, then my guess is that the same applies to all the rest
of the candidates you might otherwise rank lower.  In other words, giving
them all the same rank, as in Approval Voting, avoids the problems of
ranking them according to your honest preferences.


> So, rank methods don't improve on Approval. As I said, improvements on
> Approval are illusory.
>
> Well, the nearest thing to a "problem" that Approval has is the
> chicken-dilemma, and some rank methods can avoid the chicken-dilemma.
>
> That's the only genuine excuse to use a rank-method. There are methods
> that both avoid chicken-dilemma, and meet FBC. The best way to use such a
> method is to top-rank all of your strong top-set, and not rank any of your
> strong bottom-set.
>
> ...and, if there's a chicken-dilemma, then you can make use of that
> rank-method's way of deterring chicken-dilemma defection...as the only
> exception from voting as in Approval, by top-ranking your strong top-set.
>
> That's really the only justification for a rank method.
>

Maybe the way to consider the chicken dilemma is in the context of what
happens over time.  I would expect that repeated applications of Approval
Voting will result in having more candidates run who will appeal to many
more voters.  We're not used to thinking in terms of the broad appeal, such
as what Bernie has gotten, and he had a huge struggle just getting noticed,
at first.  I expect more like Bernie will come out of the woodwork if they
were rewarded for doing so.

Anyway, the result of having more candidates that most voters like is that
it won't matter so much that there are three candidates who are close
enough in popularity to possibly result in a surprising win by the slightly
less popular candidate.

Then again, the chicken dilemma might be a motivation for cooperating
between similar candidates to pick one of them rather than letting all
run.  I.e. parties.



> But, for some electorates, maybe ours here, there's a _psychological_ need
> for rankings. Some overcompromisers (as I said) would insist on approving
> Hillary, along with Jill.
>
> But some of those, if they had a rank-method, would be content to rank
> Hillary a little below Jill. For them, the rank method improved their
> voting. For that type of overcompromiser, a rank method softens their
> voting errors.
>
> Ranking can likewise soften the voting errors of rival parties who are so
> inimical (though close in policy proposals) that they'd refuse to approve
> eachother in Aproval.
>
>
Don't know if I care about such cases.  Maybe we should punish (or not
reward) all the rival parties who are so inimical even though they are
close in policy proposals.

Again, I think it is useful to think where Approval Voting, or other
systems, are likely to take us.  What we have currently is widely
recognized as an aberration, which is, I believe, a repeated result of the
application of Plurality Voting, which creates two dominant parties that
concentrate more and more over time, as they increasingly ignore the rest
of the people until they can't stand it any longer.

So we shouldn't judge Approval Voting, or other systems, strictly on how
they would respond to our current situation.



> So rankings can soften voting errors, for rivals and for some
> overcompromisers.
>
> ...but not for all overcompromisers. Some overcompromisers are so
> overcompromising that the only thing that can keep them from voting
> Compromise over Favorite is if they have an opportunity to rank them both
> at top, with the assurance that top-voting Favorite can't possibly hurt
> Compromise. ....in other words, for them, the method must allow equal
> top-voting, and must meet FBC.
>

As I suggested above.  Yeah!



> So, what kind of method would meet the needs of both of those 2 kinds of
> overcompromisers? It would be a method that allows equal top-voting, meets
> FBC, but also allows ranking.   ...and preferably it should give good
> protection of higher-ranked candidates against lower-ranked ones.
>


> There are a number of good methods that have those attributes. Bucklin is
> the familiar one, During the Progressive Era, Bucklin was used in at leasts
> 39 cities.
>
> There are others too.
>

I'll have to think about it some more, but I am not sure the complexity of
more than one level of ranking is worth the trouble and risk, especially
for the current level of society. But I wouldn't want to exclude it, if
there is value and it can be applied reliably.



> But I've only been talking about the _psychological_ need for ranking. But
> I mentioned earlier that, disregarding that psychological need, there's
> really only one genuine practical use for a rank-method: Avoiding chicken
> dilemma.
>
> Chicken dilemma protection is very costly in terms of problems that come
> with it, to the detriment of other properties. ...especially to the
> detriment of the method's general protection of higher middle-ranked
> candidates against lower-ranked ones and unranked ones.
>
> But, because the chicken dilemma is a genuine reason to use rankings, and
> the only genuine reason, I'd prefer to have chicken dilemma protection,even
> at the cost of general protection of middle-ranked candidates.
>

Again, I am not sure the chicken dilemma is necessarily as severe a problem
as you make it sound, once we stabilize around new political patterns that
result from years of Approval Voting, but I'll have to think about it more.


> We've been discussing a few methods that meet FBC, have a way of avoiding
> chicken dilemma, and give as good a general protection to mid-ranked
> candidates as can be gotten in a method that meets FBC & avoids chicken
> dilemma.
>
> They're Simmons' method ( MDDA(pt/2) ), and MMPO(pt/2).
>
> MMPO(pt/2)'s chicken dilemma deterrence is automatic, but its general
> protection of middle-ranked candidates is uniformly questionable, chancy,
> something of a crapshoot.
>
> Simmons' method does a top job of that general protection for the
> middle-ranked candidates against whom you aren't using the chicken-dilemma
> defection-deterence measure (...which consists of denying them approval)..
>
> With that, I'll conclude this note.
>
> Michael Ossipoff
>
>
>
>
>
>
>
>
> On Mon, Nov 21, 2016 at 1:08 AM, Daniel LaLiberte <
> daniel.laliberte at gmail.com> wrote:
>
>> This message is about two related subjects:
>>
>> 1. Factoring in the long-term emergent effects of each voting system.
>> 2. An example of how Approval Voting results in better long-term effects.
>>
>> Among the many criteria for evaluating voting systems (
>> https://en.wikipedia.org/wiki/Voting_system#Evaluating_voti
>> ng_systems_using_criteria) I don't see any that address the long-term
>> effects of using each voting system.  In other words, the effect on one
>> election is certainly important, regarding the satisfaction of the election
>> results by voters and candidates.  But I would argue that it is even more
>> important to consider the long-term effects that emerge when applying a
>> voting system repeatedly over many elections.  A small bias one way or
>> another may not be very apparent if you only look at one election, but over
>> may elections, they can add up and perhaps compound the bias exponentially.
>>
>> There are many long-term effects to consider, but in particular, I am
>> thinking of one pernicious problem: the tendency for two major parties to
>> emerge and dominate all politics which results from the repeated
>> application of plurality voting.  This problem is fairly easy for most
>> people to understand, although I am surprised to see that there seems to be
>> a lot of denial about this effect as well.  Some would even defend having
>> only two major parties, or very few parties.  That is an interesting
>> subject to discuss, but regardless, I believe we should be aware of how our
>> choice of a voting system will affect things over time, how society is
>> likely to evolve based on the rules we lay down, and in fact, how it is
>> actually very likely that the dominant forces in society will quickly and
>> vociferously defend whatever rules resulted in their rise to dominance.
>>
>> But back to this one question, studies and long-term experience have
>> shown that other voting systems besides plurality, in particular IRV, also
>> result in the dominance of two major parties.  This may be more surprising
>> to people, but looking at the underlying cause, it seems we can make a
>> rather important simplifying argument about most voting systems regarding
>> this problem. I would assert that the underlying cause of this problem of
>> two-party dominance in any voting system is that it gives voters the
>> ability to rank or order at least one candidate higher than the rest.
>>
>> The reason this ability to rank candidates becomes a problem is the
>> spoiler effect, where voters will have a strong motivation to give their
>> highest rank to one of the leading candidates. If they don't, then they
>> weaken the chance of that candidate winning and therefore strengthen the
>> chance of the less preferred leading candidate.  Because one of the leading
>> candidates is likely to win, all the rest of the rankings of non-leading
>> candidates hardly matter at all.
>>
>> In any election, there will be two candidates who are the strongest in
>> terms of popular support, and thus the most likely to win. Consequently (to
>> grossly over-simplify the process) with any voting system that permits
>> ranking, groups of voters will tend to coalesce around support for these
>> two leading candidates to encourage everyone to support their preferred
>> leading candidate. Eventually two major parties arise, and everyone who
>> doesn't join one of these two major parties is excluded.
>>
>> So once voters and candidates figure this out, any such voting system
>> ends up devolving into the dominance of two major parties that we get with
>> simple plurality voting.  In fact, one might argue that plurality voting is
>> better just because it is simpler.
>>
>> But Approval Voting avoids this problem. Equal-rank approval votes mean
>> voters don't get the option to express their preferred ranking, but because
>> of that, they aren't at all motivated to bias their ranking dishonestly.
>> They only have to decide which candidates to approve, or where the cut-off
>> is between approval and disapproval.
>>
>> Given that there are, as before, two leading candidates, how does
>> Approval Voting affect whether one of those two leading candidates will
>> win?  One of the two leading candidates is likely to win even with Approval
>> Voting, so it would appear there is no benefit, but that would be a
>> short-sighted way to judge a voting system.  In subsequent elections, it
>> would seem likely that more candidates will run who have broader appeal to
>> ALL voters, not just a majority or plurality. Because the winning
>> candidates will be those who are most approved of by the most voters, there
>> will be no value in parties that typically focus on appealing to no more
>> than half of the voters.
>>
>> So I suspect the long-term use of Approval Voting would be
>> self-correcting toward better and better representation of ALL voters, not
>> just half the voters, because in each election, almost all of the voters
>> contribute to choosing the winning candidates, and that only gets better
>> over time as the candidates who decide to run get closer to receiving the
>> approval of all voters.
>>
>> Can any other voting system claim a similar long-term effect?  Even a
>> voting system with three ranks, e.g. Approve, Neutral, Disapprove, would
>> encourage voters to approve one of the leading candidates, and give neutral
>> or disapprove votes to the rest.  I wonder if Approval Voting might be the
>> ONLY system that has this long-term effect.
>>
>> What I am aiming for is a voting system that self-corrects over time.  No
>> matter what voting system we choose, there is probably always going to be
>> at least some small bias, some inequities or incompleteness.  So we need to
>> understand this and deal with it.
>>
>> But one important question should not be overlooked: What do we want to
>> self-correct toward? That is, what is the long-term goal? I believe we
>> should want to move toward a closer or better representation of society as
>> a whole, but there are other ways to look at that.
>>
>> Looking forward to reading your feedback.
>>
>> --
>> Daniel LaLiberte
>> daniel.laliberte at gmail.com
>>
>> ----
>> Election-Methods mailing list - see http://electorama.com/em for list
>> info
>>
>>
>


-- 
Daniel LaLiberte
daniel.laliberte at gmail.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.electorama.com/pipermail/election-methods-electorama.com/attachments/20161122/cecdd0fa/attachment-0001.htm>