[EM] MultiGroup voting method

[Abd ul-Rahman Lomax]

At 05:25 PM 4/6/2007, Juho wrote:

>>Is Juho aware of the use of Asset Voting that I have proposed? It
>>seems far simpler and more flexible than what Juho has proposed,
>>with similar effect.
>
>Yes I'm aware of Asset Voting. There are similarities but also lots
>of differences. Maybe the biggest differences come directly from the
>tradition. The MultiGroup method fine-tunes the traditional party and
>region system and (as I mentioned, at least in the vanilla version)
>the votes are very simple votes to one candidate. When compared to
>that, Asset Voting seeks totally new tracks.

Consider this: if you have a system designed for one vote, what do 
you do if the voter marks more than one? Typically, it will be 
considered an error and the vote is discarded. In Approval, you 
simply count all the votes and the appearance that the voter has cast 
more than one vote is illusory, because the only vote which actually 
counts is one cast for the winner. All the others are moot.

But in Asset, no votes are moot. So if the voter casts more than one, 
the vote must be split. It's actually a simple consequence of 
disliking the discarding of votes.

But it is not an essential feature of the use of Asset that I've 
proposed. Asset itself wasn't designed for multiwinner, but it 
happens to work excellently for it.

Asset as I've proposed, without doing it explicitly and requiring the 
party machinery, does "fine-tune" it.

Once you are going to consider having overlapping districts, i.e., 
there can be more than one representative who represents a particular 
geographic location, with some representing the entire state, my 
question remains. Why be more complicated?

Complex reforms are pretty unlikely to be implemented. I don't think 
that the citation of "tradition" as a difference makes sense. Neither 
of these is traditional. If we really want to go outside of 
*political* tradition, we could go to delegable proxy (though this, 
in fact, simply brings in long-standing tradition in corporate 
governance); Asset, as I've been describing it here, remains with a 
peer legislature.

(The real difference between Asset and Delegable Proxy is that Asset 
uses something like delegable proxy to elect a peer legislature, 
whereas delegable proxy doesn't seek or need a peer legislature, it 
allows representatives to have variable voting power. DP would be, 
indeed, as mentioned, a departure from political tradition, and some 
worry about how a non-peer legislature would function. I personally 
prefer to see other applications of DP have some history under their 
belt before being applied in mission-critical applications. Asset, on 
the other hand, I have no problem with being implemented tomorrow.)

I will also note that I did look over Juho's proposal, and, beyond 
seeing similarities to Asset, I didn't understand how it works. That 
should be taken as a flaw. (Certainly how it works may have been 
explained, but that I didn't see it readily means that it has not 
been explained in a way to make it easy to follow. Part of the 
problem could be the complexity.)

>Another existing stream with connections to multiple interests is the
>possibility to give proxies to different persons on different topics.

What does this have to do with Multigroup?

I don't like multiple proxies even in open Delegable Proxy. It 
creates vast complications and dilutes responsibility. However, in 
the FA/DP world, "different topics" means different organizations, 
so, in fact, one does assign "different proxies on different topics."

The bills before a legislature are generally not "topics." There is 
not one legislature for, say, business law, and another for criminal. 
Or one for dealing with the environment and another for dealing with 
the budget. In fact, all these topics are inter-related. The closest 
that one might get to this is the use of proxies in the committee 
system, where a legislator *may* assign a proxy to vote for the 
legislator in committee. This is existing practice in some 
legislatures. Of course, a "committee" is set up specifically for a "topic."

>Asset Voting...
>>A political party that was small and spread thin, but with enough
>>loyal voters, state-wide, to gain a quota of votes, would gain
>>representation state-wide. A party with even less support than
>>that, could cooperate with other similar groups to create a seat
>>that represents more than one party, presumably with similar
>>agendas or interests.
>
>I tried to go also further, to allow even country wide small (roughly
>quota size, cross region, possibly cross party) groups to get one
>representative.

That's not "further." That's what I proposed and describe, and what I 
described went a little further. Asset essentially allows the voters 
and the candidates who receive votes to organize how groups are 
represented. The "seats" may respect regions, parties, or interests, 
or just may reflect an attempt to elect trustworthy and competent people.

I say "further" because I described what would happen if a group 
*doesn't* have a quota.

>  That is, if such groups are allowed by the system
>(and by the parties if these candidates come from existing parties).
>(I mean that MultiGroup is basically just a calculation method that
>can support all this, but on the other hand can be used in many
>different, also limiting ways in different countries, within
>different parties etc.)

If the system doesn't *require* formal groups, and Asset does not, it 
fully "allows" them.


>Asset Voting...
>>It takes no complex, top-down, imposed system. All it takes is
>>allowing votes to amalgamate intelligently, i.e., under the
>>direction of a trusted candidate.
>
>I wouldn't call political party strycture of a democratic political
>system an "imposed system".

No, you've missed the point. To apply the terminology of "imposed 
system," you would look at the election methods and procedures. 
Currently, political parties in the U.S. typically own the ballot. 
Yes, you can get on the ballot as an independent, but it can be an 
onerous process. Party designations are on the ballot. Given that 
people are being elected, not parties (a winner can say, I changed my 
mind, I'm a Republican now, or whatever), it is an imposed system 
that the party names are on the ballot at all.

If there is some structure for allocating votes, something, perhaps 
that Juho is designing now, that's a "top-down, imposed system." I 
mean that it is not created from the bottom, contrasted with how 
Asset creates seats or delegable proxy selects top-level proxies.

>  In many existing political systems there
>are characteristics that resemble that though. And the intention of
>MultiParty is to alleviate those. A "party-less bottom up" generation
>of the "leading class" and political groupings is another approach
>(with + and -).
>
>>What Juho is trying to do is to automate a process, using fixed
>>rules, instead of allowing human intelligence, which is far more
>>flexible and which can deal with unanticipated contingencies, etc.,
>>to allocate votes and seats.
>
>Also, to give more power directly to the voters, while maintaining an
>easy way to vote, easy understanding of what the candidates stand
>for, and with accountability.

Asset is about as simple a voting system as can be described. I'd 
think that most voters would simply, out of the universe of persons 
they would trust to serve as electors, pick one, the one the most 
trust. The split vote aspect of asset is simply an additional 
freedom, not a necessary burden. It also recovers useful information 
from overvotes, contrasted with no-overvote systems which discard all 
the information (and which thus elected Bush in 2000).


>>It wasn't clear to me how Juho's scheme really differs from Asset
>>Voting
>
>I have the feeling that the differences may be more numerous than the
>similarities, so I don't really know which properties to comment.
>Right now the topmost thing in my mind is the difference that in
>MultiParty candidates declare their affiliations/preferences/policy
>before the election,

Is that a difference? Can't candidates do that in Asset? Are 
candidates *required* to do this in Multiparty? What if a candidate 
declares "I will simply serve the public interest as I see best at 
the time." And thus indicates no specific party affiliation or 
preference or policy. Is this candidate disqualified in MultiParty? 
If so, it's an "imposed system." If not, then, what is the difference 
between this and Asset. Not this!

>  for the voters to make their pick, while in
>Asset Voting the system is based on bottom up votes/trust and
>negotiation process that eventually leads to formation of the groupings.

Yes, there is the negotiation process. However, that only applies, 
necessarily, to votes that would otherwise be wasted. I'd generally 
assume that most candidates who receive a quota of votes at the 
outset will simply take their seat. And it isn't necessarily 
"negotiation," though negotiations *may* be involved. A candidate may 
simply reassign excess votes. A candidate with enough votes doesn't 
have to negotiate with anybody. He or she can simply create another 
seat. It's the excess votes, the fraction-of-a-seat blocks of votes, 
not under the control of one person, that may involve negotiations.

The complexity of "voters make their pick" is vastly underestimated. 
It is this that actually disempowers the voters. Under the guise of 
allowing the voters to choose, special interests can manipulate 
elections by, through media and the manipulation of how questions are 
presented, leading voters who, quite simply, are too busy to do the 
research, to make choices contrary to their own interests. I've seen 
again and again how the knee-jerk responses of the uninformed are 
utilized to warp outcomes away from what the voters would actually 
choose if they had the time to gather the needed information. (To be 
sure, they would also need the motivation, but, in fact, even where 
they have the motivation, they typically don't have the time. It 
takes *time* to become adequately informed, if one is going to make 
issue-based decisions.)

So instead of deciding based on platforms -- which are simply 
promises and, often, platitudes, designed to create an appearance of 
"this is what we want" for voters -- I vastly prefer to see 
representation based on trust. Personal trust. There is no substitute 
for it except the deception, hypocrisy, and pretense that passes for 
politics today. Even the best and most trustworthy candidates are 
forced to play the game.

In an Asset system, I see elections evolving into choices of 
representative on smaller and smaller scales, where the voter, in 
general, personally knows the candidate the voter is voting for. 
That's impossible without some kind of vote reassignment, some 
gathering of votes cast for candidates who didn't gain a quota. And 
the obvious way to do this is to allow the candidates to redistribute 
the votes. Given the utter simplicity of this, I'd really want to 
know why we need something more complicated!

We don't really have true representation, in fact, until we have 
something like this. I don't have a representative if it is not the 
person I chose -- or someone chosen by this person, etc.

(I've pointed out that standard proxy, I think, does allow 
delegability. It's just not normally used. As I've mentioned, when we 
were buying a house, we were also in Ethiopia, so we gave our 
attorney a power of attorney. I'm pretty sure that if some critical 
action needed to be taken, and the attorney couldn't personally do 
it, he could and would designate someone else to do it for him. 
That's delegable proxy. However, what we are doing with DP is 
suggesting that this delegation be considered routine, and that it be 
used to automatically create a bottom-up hierarchy allowing the 
scalability of direct democracy.)

Asset essentially applies proxy concepts, which are very 
well-established, and which are clearly what people choose who have 
the power to get what they want, to elections. Corporations allow 
proxy voting because the wealthy people who formed them originally 
demanded it. Proxy voting *enhances* the power of the voter, it 
doesn't reduce it, as long as the proxy can be freely chosen.

(Small shareholders often don't have the power to choose a proxy whom 
they actually trust, based on knowledge of the person; we are 
suggesting delegable proxy as a way to empower small shareholders. 
Large shareholders don't need it.... They already hire firms which do 
nothing but represent shareholders by holding proxies. It isn't 
necessary for corporate structures or procedures to be changed to get 
DP for small shareholders, for such can organize independently and 
use DP to collect top-level proxies, who would then receive or 
recommend the formal registered proxy assignments, assigned, then, 
directly from the shareholders to the corporate process.)