Re: [CES #7272] Choice of Representation

[Markus Schulze]

Hallo,

> People wanting to represent others in the legislature would file their
> nomination papers with the appropriate clerk or office. They would be
> put on the ballot. Each voter can vote for one person to represent
> them. All candidates would then serve in the office for which they ran.
> The weight of their vote on legislative matters would equal the number
> of votes that they received in the election.

This election method has already been proposed zillions of times.
How do you solve the Banzhaf problem? See:

http://lists.electorama.com/pipermail/election-methods-electorama.com/1999-October/003189.html

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

Markus Schulze

[William Waugh]

Thank you, Markus Schulze.  I am taking the information you provided back to the forum where the proposal was made https://www.facebook.com/groups/195236817222399/permalink/405728609506551/?comment_id=433336486745763&offset=0&total_comments=26

On Sunday, February 10, 2013 1:57:33 AM UTC-5, Markus Schulze wrote:

Hallo,

[Markus Schulze]

Dear William,

the points are:

(1) When the proposed election method is being used, then the
real voting power (according to Banzhaf or according to Shapley
and Shubik) of a representative can differ significantly from his
nominal voting power.

(2) Voters are only interested in increasing the real voting
powers of their representatives.

(3) Suppose a vote is "wasted" when it increases only the nominal
voting power, but not the real voting power. Then, when the
proposed election method is being used, the number of wasted
votes can be very large.

Markus Schulze

[Jameson Quinn]

2013/2/11 Markus Schulze <Markus....@alumni.tu-berlin.de>

Markus Schulze

--
You received this message because you are subscribed to the Google Groups "The Center for Election Science" group.
To unsubscribe from this group and stop receiving emails from it, send an email to electionscien...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

Professor Schulze's points are all of course true. However, as the number of representatives goes to infinity and the voting power of each remains bounded, real voting power approaches nominal voting power and the fraction of wasted votes approaches zero. I believe that the number of reps would be high enough in practice that voting power discrepancies would not be a major concern.

Jameson

[William Waugh]

I'm not sure whether there are good counterarguments to what the conversants are saying on Facebook on this; I'll just reproduce it here.

William Waugh

I now think "Choice of Representation" may be fundamentally broken. Markus Schulze said "This election method has already been proposed zillions of times." He raises an example of the problem with it. "A has 37 votes, B has 30 votes, C has 21 votes, D has 12 votes. Then A, B, and C have the same voting power because every proposal needs the approval of only two of them. D has no voting power." In a scenario like that in a committee, the voters represented by D have no representation because their representative has no power.

Like ·  · Unfollow Post · Yesterday at 6:13pm

• 

  Jim Mueller A carefully constructed totally unrealistic example does not negate the value of the concept that everyone should have the Right to Representation of their choice.
  Even if this fact situation were to exist there would still be a lot of value in it.

  Yesterday at 6:47pm · Edited · Like
• 

  Jim Mueller What is the configuration if Direct Representation is not in place? One office holder A has all of the power? Then Direct Representation is making a big difference overall even if everyone doesn't have a representative who can change the outcome
  
  D is on the inside and can speak. He has the information to take to the media to expose inside deals and bad legislation and he has a base from which he can increase his number of votes in the next election.
  
  B,C and D have a lot more power than if they had been excluded by winner-takes-all.

  Yesterday at 6:49pm · Like
• 

  Jim Mueller William Waugh, The example is a 4 person committee with a carefully selected total of 100 and totals for the representatives so that one of them, D, in conjunction with any other one would not be able to pass anything.
  Well guess what without weighted votes no 2 people on a 4 person committee can take any action.
  
  A committee only recommends legislation. The whole body will vote on the bill. In the example what is the make up of the whole body? It is very likely that D could have an impact in the whole body whereas in winner-takes-all he would have been excluded and most other systems might have excluded him/her also.

  Yesterday at 7:10pm · Edited · Like
• 

  Jim Mueller With winner-takes-all elections all of the state legislatures are controlled by one of the other of the 2 dominant political parties in this country. When a committee is appointed, if it has an odd number of members the majority party will have the higher number of members. Most of the time the minority members have no power. When there is an even number, the majority party will have an advantage of two or more over the minority party. The minority party members will rarely if ever be able to influence the outcome of any vote.

  Yesterday at 7:28pm · Edited · Like
• 

  Jim Mueller The example given is similar to the Nassau County situation that caused Banzhaf to challenge the validity of the power system.
  http://en.wikipedia.org/wiki/Banzhaf_power_index
  
  In that case 9 of the 30 votes were in the hands of one person, another ...See More

  

  Banzhaf power index - Wikipedia, the free encyclopedia

  en.wikipedia.org

  The Banzhaf power index, named after John F. Banzhaf III (originally invented by...See More

  Yesterday at 8:04pm · Like · 1 · Remove Preview
• 

  Marcia Also with C.o.R. the votes are actually what the public really want, so if only 12 people want the bill why should it pass? or be blocked?

  7 hours ago · Like

[Bruce Gilson]

Markus is, I believe, under a misconception as to what the Banzhaf power index really measures. If you bel;ieve that a group with 35% of the citizenry behind it should win 35% of the time, then you might rely on the Banzhaf power index (actualy, this is something of an oversimplification, but very close to what it means). But the only way you could really accomplish this is to eschew voting and simply put everyone's name on a slip of paper, drawing one from a bin whenever a question is to be decided, and letting the named person make the decision. This is not a system which have heard of anyone advocating.

The Banzhaf power index in fact shows the power of *aggregating* votes. If we applied any of the standard proportional representation formulas to the Schulze scenario, where the votes for four parties divide 37:30:21:12, it will lead to a set of party delegations with 2:2:2:1 (if a 7-member assembly) or 3:3:3:2 (if an 11-member assembly). While the Banzhaf power index gives all representatives equal power (14% in the 7-member case, 9% in the 11-member case), this would not be an accurate measure of power if members of a party vote together all the time. If parties are disciplined, as in Great Britain, so a representative's vote is determined by his party, these 7- or 11-members would have exactly the same characteristics as the weighted assembly, with 37:30:21:12 weights, described: the smallest party would never be able to influence the result and the other three would all be able to win by coalescing with any one of the other two.

It is the aggregation that matters. Compare the House of Representatives in the USA with the Electoral College. The ratio of electoral votes between California and Wyoming is smaller than the ratio of representatives. Yet California is considered a bigger prize in the Electoral College than in the House, because its electors are chosen *en bloc.* No matter how badly gerrymandered California might be, there will be members of both parties in its delegation. Not so with its electoral college delegation. And that is why, using the Banzhaf index, the big states come out as powerful.

But the Banzhaf index does not take into account one important factor — how likely a state is to be a “swing” state. Candidates know better than to campaign more than perfunctorily in California (a reliably Democratic state) or Texas (an even more reliably Republican one). They will devote more energy to Virginia, though the Banzhaf index is much smaller for Virginia than either of the biggest two. They do this because, while the Banzhaf index measures the probability that a swing in California, Texas, or Virginia will change the final result, it does not include the probability that there *will be* a swing in California, Texas, or Virginia.

So I think Markus' focusing on the Banzhaf index is misplaced.

 

[Abd ul-Rahman Lomax]

The Banzhaf "problem" covers over the fact that a minority won't win.
It assumes fixed positions. I assumes that a member with too few
votes cannot influence decisions, missing that a major function of a
representative isn't in voting, it's in deliberation and the
presentation of evidence and arguments. Yes, if a member of a
committee has little voting power, and the rest of the committee
doesn't want to hear from them, they can be excluded. But that's just
ordinary common sense!

There is a real problem, however, with the proposal,that should be elaborated.

That is the problem of deliberation. It was noted that a legislature
could meet -- and vote -- on-line, but this entirely misses the
problem of scale in deliberation. As the number of participants
increases, the noise increases. That, in fact, is why legislatures do
most business in committee. It's impossible to do it on the floor.
The number of participants must be kept at a manageable level.

There is another way, that takes the proposal for direct
representation, and uses it to elect or empower what I call "seats."

This is Asset Voting, and it originated as a tweak on STV. I call the
candidates, as registered and who have received even one vote,
"electors." In a full implementation, these electors have two
functions: first of all, they may amalagamate their votes to create
seats, which will, by default, represent them in the legislature. The
electors elect seats by contributing votes to them, votes received in
the general election.

Because many people, seeing this proposal, make unwarranted
assumptions about it, I'll give a more details.

The electors may transfer unspecified votes, but may, instead,
transfer *specific* votes, that is, votes from a specific precinct.
That is, the votes may have a location attached. Electors may
transfer fractional votes. The basic rule is that a "seat" is elected
when an elector holds, directly, or as transferred, a quota of votes.
The seat then will be able to, by default, exercise a quota of votes
in the legislature. This would be "one legislative vote." Rules may
include procedures for electors to withdraw their votes from seats,
rules under which a seat is lost, or possibly continues with somewhat
reduced voting power.

I generally assume that the legislature has no fixed districts.
Because of location-specified vote transfers, and because people may
naturally vote for someone local, seats may defacto represent voters
in some district, but some seats may have no attached district, will
be "at large."

The legislature itself will make its own rules. It will optimize its
own size by determining the optimal number of seats, but, of course,
it will start with some number. It may be best to allow the number of
seats to vary to some degree, by varying the size of the quota. Some
votes will be unutilized, but *in principle* no votes are wasted.

The electors, then, in a full implementation, may vote directly on
any issue. If they vote on a measure the same as their seat, there is
no change in the reported vote. However, if they vote differently,
their fraction of a vote counts, and is deducted from the seat's vote.

However, given that they have a high degree of freedom in choosing
who will represent them, it can be expected that most issues will
simply be decided by seat votes. Some votes in a legislature,
particularly those concerning "questions of privilege," would be
decided solely by the votes of seats, without remote voting being
enabled. It is my sense that only rarely would remote voting reverse
a legislative decision, when it's close.

What has been done with this kind of device is to separate the
process of deliberation and that of amalgamation. Amalgamation *can*
be done on a large scale, but the problem of scale in democracy has
always been in the attempt to combine the two processes. Town
meetings break down when several hundred people are in attendance.
They can decide things, but only with little process of the
adjustment of proposals through amendment and debate. *Representation
is needed for deliberation.*

It is possible for the entire legislative process to be transparent,
as to deliberation on the floor. Electors, then, may observe this
process. If they have something to contribute, they may do so through
their elected seat, who presumably will be friendly to their ideas.

If delegable proxy is used, committees may *effectively* have the
entire legislature participating, but just through a few seats.
Committee decisions are *never* binding on the whole legislature. (I
recommend delegable proxy for non-binding consensus process.)

Most electors will not participate, that's my prediction, except on
issues of special interest to them. The seat will do the job, and
that's why the seat will be paid, and not electors.

I'd like to note that Asset, as described, *does* allow the variable
voting power of Waugh's proposal. It just sets up a default process
for representation in debate.

Consider this: if there are a million people who can enter a motion,
how is it decided what motions to handle and debate? There needs to
be some filtering. Asset sets up the seats as filters.

The seats represent all their voters (direct and indirect) in the
deliberation process. Electors directly represent their voters in
selecting and supporting seats and, for some electors, participating directly.

A seat might even routinely assign a "participating elector" to
function on a committee where the elector has expertise. (If seats
can assign proxies for special purpose, the elector might even be
able to vote. Note that if this is done, the elector would be voting
-- in committee, I assume -- on behalf of *all* the seat's electors,
not just himself or herself.

This is far more than a voting system. It's a method of constructing
an intelligent social process.

And it starts with Asset, which can very easily be implemented in any
organization. We tried it for our Steering Committee election some
years ago. It worked. What we got was a three-person steering
committee that, very clearly, represented the entire electorate. We
might not have done quite so well if there had been a hold-out. But
we still would have had something as good or better, in terms of full
representation, than any other system I've heard of.

It wasn't magic. Having the steering committee did not automagically
create activity. We had a committee that could make decisions on
behalf of the membership, but, at that point, membership wasn't
crisply defined, and it seemed that people assumed the committee
would *lead*. All Chiefs and no Indians.... No, we simply
*represented* the (diffuse) membership, which collectively did very
little, did not present us with proposals for decision. But that was
not a problem with the election method.

Asset does not require specially informed voters. It was invented by
Lewis Carroll specifically because STV -- he was a major theorist
behind STV in the late 19th century -- had a problem: most voters
only knew their favorite. To be able to rank a series of
representatives, sensibly, requires *political expertise* that most
voters don't have. STV, then, to function, requires a party system,
to advise voters. Many voters, if allowed to truncate, then, do, and
then, because of eliminations, many voters are unrepresented. If they
are not allowed to truncate, they donkey-vote. It's garbage information.

With Asset, voters need only choose based on a single criterion: whom
do they most trust to represent them in the legislative process?

They may choose a neighbor, or they may decide to choose someone
well-known. Whomever they choose, they are empowering that person to
exercise voting power on their behalf.

In considering Asset, people often assume the continuation of the
existing system, that there will be *relatively few candidates.* They
assume people vying for power, holding out to be personally elected,
etc. Someone who does that will become quite visible as someone
unable to compromise. If they stick to it, those who voted for them
will end up unrepresented by a seat. If it's direct/representative
asset (direct voting allowed). and if the elector participates, the
votes are not lost. But the elector has to do more work, for
practically no gain.

By transferring votes, the elector gains influence in floor
deliberation. They do not lose their right to vote, but they lose the
*obligation* to vote on every issue, if they are able to choose a
seat they trust.

Very simple to set up Asset. CES should develop a standard set of
rules for organizations.

The basic Asset concept itself is powerful. With our election, the
rules had been poorly specified; but following the concept, the
electors were able to work out what to do in a few days. We hadn't
specified the quota, for example. But because we, as electors, wanted
everyone to be represented, *it didn't matter.*

It was a three-person committee. Once we had two members elected with
the Hare quota, those two members, by agreement, could have made
whatever rules were necessary to complete the election. What they
would be doing in that case, is agreeing on who would break ties
between them. *This is standard arbitration process!* The two
members, by agreeing, were representing a supermajority of voters,
and supermajorities of two-thirds, can generally change any rule they
please (subject to certain procedural limitations).

Asset, strictly speaking, is not an "election method," but it can be
used as one. It's a method of making decisions by *freely chosen
representation.*

Society already functions this way, in many places, but it simply is
not formalized. (Or it is formalized only at one level, primitively,
as with stockholder proxy representation.)

[Abd ul-Rahman Lomax]

At 11:13 PM 2/11/2013, William Waugh wrote:
>I'm not sure whether there are good counterarguments to what the
>conversants are saying on Facebook on this; I'll just reproduce it here.
>
>
>

><https://www.facebook.com/william.waugh.33>William Waugh

>
>
>
>
>I now think "Choice of Representation" may be fundamentally broken.
>Markus Schulze said "This election method has already been proposed
>zillions of times." He raises an example of the problem with it. "A
>has 37 votes, B has 30 votes, C has 21 votes, D has 12 votes. Then
>A, B, and C have the same voting power because every proposal needs
>the approval of only two of them. D has no voting power." In a
>scenario like that in a committee, the voters represented by D have
>no representation because their representative has no power.
>

This is a committee representing 100 people. A and B collectively
represent 67 people, a supermajority. If this representation is
unconditional, yes, A and B have not only a majority, but a
two-thirds supermajority, which is sometimes considered a
"consensus." However, if A and B differ, then C can decide an issue,
because A and C represent 58 and B and C represent 51.

D has effectively no voting power, if all that is of concern is
"majority vote." However, most organizations, in fact, need to value
unity; if they don't, consider that D leaves the organization because
nobody listens to her. D and D's voters go somewhere else, do
something else. A, B, and C now represent 100% of the voters, but the
total organizational power has been reduced by 12%.

In the reformed committee, now with three members, each still has the
same votes, but A is now 42%, B is 34%, and C is 24%.

A and B still have a supermajority, but C may decide any issue on
which A and B do not agree.

Now, of course, I gave an extreme case: all the D voters leave. But
what is the relationship of D to A, B, and C?

Suppose proxy voting is allowed. D may decide to allow one of the
others to represent her, which would make sense. The committee is
still, then, a three person committee, with no individual having the
power to make a majority decision unilaterally. In the end, the
"variable voting" feature ends up with the same result as if what
were done were to be to elect any standard three-member committee by
a decent voting system. We'd only get a real difference if a member
had majority power, which is roughly equivalent to electing an
officer with two non-binding advisors.

My sense, however, is that for members, particularly of committees,
to have equal voting power, is probaby socially more functional than
routinely variable power. I allow, in my Asset proposals, variable
voting power *as a tweak.* The basic voting in an Asset Assembly is
of peers, with equal power, and the major variation in voting power
is kept at the elector level, where there are likely very many
electors, some with as little as one vote.

The Banzhaf "problem" boils down to a claim that a single vote has no
effective power. It *totally* neglects the power involved in the
right to participate in deliberation. If D is on that committee, D
may enter motions. As is routine, D can do nothing without the
consent of another member, a "second." D needs the support of either
A or B to avoid a supermajority dismissal.

What is often missed is that a supermajority dismissal can demolish
an organization's unity. A and B could lose a third of the
organization if they won't even consider proposals supported by C and
D. That's *expensive.* And yet we would not want D to be able to
*force* the committee to consider an issue where A, B, and C, don't
want to consider it. A and B have, with the voting power as
described, the right to dismiss a proposal immediately. There is
nothing wrong with that, as to process. As to wisdom, that's up to
them and Reality.

Generally speaking, political scientists have missed the problem with
assuming that deliberation and amalgamation must involve the same
group. There are reasons for this, but they have been inadequately
explored, and, because of this, political scientists have generally
considered the problem of scale in democracy as being irresolvable.

(The reason is that those who have not participated in deliberation
should not, it's thought, vote on the motions that result, and this
is the classic argument against proxy voting in democratic process.
However, "proxy voting" is often thought of as "remote voting," where
someone casts votes for someone else, distinct from their own votes.
Proxy *representation* does not have this problem. Yet we can go
further than that. If the deliberation is open and produces a record,
a member having the right to vote may review that. The problem boils
down to *who decides that a member is not adequately informed to
vote*? My conclusion is that the member decides, it voting is
*practical*, and this is exactly what happens with public elections,
people decide whether to participate or not. However, *may they
delegate their right of participation*? If these were property
rights, we'd have no question, they may. In spite of our thinking
that this is *our* country, that the people own it, we don't. That
level of democracy has never been accepted, and it has active opponents.)

[Abd ul-Rahman Lomax]

At 11:13 PM 2/11/2013, William Waugh wrote:

>I'm not sure whether there are good
>counterarguments to what the conversants are
>saying on Facebook on this; I'll just reproduce it here.

I will respond to the comments. This is
essentially a discussion I've been involved in for about a decade.



>
><https://www.facebook.com/william.waugh.33>William Waugh

>
>
>
>
>I now think "Choice of Representation" may be
>fundamentally broken. Markus Schulze said "This
>election method has already been proposed
>zillions of times." He raises an example of the
>problem with it. "A has 37 votes, B has 30
>votes, C has 21 votes, D has 12 votes. Then A,
>B, and C have the same voting power because
>every proposal needs the approval of only two of
>them. D has no voting power." In a scenario like
>that in a committee, the voters represented by D
>have no representation because their representative has no power.
>
>
>
>

><https://www.facebook.com/groups/152631684794770/#>Like
>· · Unfollow Post ·
><https://www.facebook.com/groups/152631684794770/permalink/483547471703188/>Yesterday
>at 6:13pm
><https://www.facebook.com/jim.mueller.5680>Jim

>Mueller A carefully constructed totally
>unrealistic example does not negate the value of
>the concept that everyone should have the Right
>to Representation of their choice. Even if this
>fact situation were to exist there would still
>be a lot of value in it.

><https://www.facebook.com/groups/152631684794770/permalink/483547471703188/?comment_id=483555571702378&offset=0&total_comments=6>Yesterday
>at 6:47pm ·
><https://www.facebook.com/groups/152631684794770/#>Edited
>· <https://www.facebook.com/groups/152631684794770/#>Like
Debate on this issue often assumes some kind of
natural right to representation. I don't think
natural rights directly exist; rather, we
*declare* them as a tool to facilitate a
functional society. Given that as being behind
"should," yes, we should have the right to chosen
representation. The alternative is that someone
else chooses for us. There is a way to accomplish
this called Asset Voting that keeps the concept
of representative democracy and that distributes
voting power among a set of chosen
representatives who each have, by default, equal voting power.

><https://www.facebook.com/jim.mueller.5680>Jim

>Mueller What is the configuration if Direct
>Representation is not in place? One office
>holder A has all of the power? Then Direct
>Representation is making a big difference
>overall even if everyone doesn't have a
>representative who can change the outcome
>D is on the inside and can speak. He has the
>information to take to the media to expose
>inside deals and bad legislation and he has a
>base from which he can increase his number of votes in the next election.
>B,C and D have a lot more power than if they had
>been excluded by winner-takes-all.

><https://www.facebook.com/groups/152631684794770/permalink/483547471703188/?comment_id=483557141702221&offset=0&total_comments=6>Yesterday
>at 6:49pm · <https://www.facebook.com/groups/152631684794770/#>Like
Asset Voting is a tweak on Single Transferable
Vote which, if voters were fully informed, would
get close to allowing full representation. In
practice, voters *cannot be fully informed,* and
votes are wasted, and a party system is
necessary, which begs the question (who controls the party?)

Asset Voting is essentially the system proposed,
but instead of electing seats in a legislature,
those receiving votes become "electors" who then
have the power to choose seats by assigning seats
a quota of votes. Extra votes may be assigned
elsewhere. And it is possible for electors to
maintain direct voting rights, but only the seats
have the right to enter motions, debate on the floor, etc.

Asset may be implemented simply, in small
organizations, to create representative
committees or boards or assemblies. It is not at
all complicated. This, then, suggests a reform
path: use an advanced system like Asset in a
voluntary organization. If it works as we expect,
the organization will be strengthened, and the
approach will be imitated. When people see it
work in small organizations, eventually, it will
be tried in larger ones, and, eventually, in public political systems.

I'll note that the last step may actually not be
necessary, because the political problem boils
down to how people communicate, cooperate, and
coordinate. The people, organized, can
essentially do whatever we want. It is that we
are not organized that it seems that special
interests groups have so much power.

><https://www.facebook.com/jim.mueller.5680>Jim
>Mueller
><https://www.facebook.com/william.waugh.33>William

>Waugh, The example is a 4 person committee with
>a carefully selected total of 100 and totals for
>the representatives so that one of them, D, in
>conjunction with any other one would not be able
>to pass anything. Well guess what without
>weighted votes no 2 people on a 4 person committee can take any action.
>A committee only recommends legislation. The
>whole body will vote on the bill. In the example
>what is the make up of the whole body? It is
>very likely that D could have an impact in the
>whole body whereas in winner-takes-all he would
>have been excluded and most other systems might
>have excluded him/her also.

><https://www.facebook.com/groups/152631684794770/permalink/483547471703188/?comment_id=483562668368335&offset=0&total_comments=6>Yesterday
>at 7:10pm ·
><https://www.facebook.com/groups/152631684794770/#>Edited
>· <https://www.facebook.com/groups/152631684794770/#>Like

Mueller is correct. The example, to use a
technical term popular in voting systems dicussions, sucks.

The power to participate in deliberation is
totally neglected. The *real* problem with the
proposal would show up when there are 12,527
people who want to sit on the committee. If there
is no means to boil this down to a few people (30
is high, actually), the committee will be
paralyzed, or will operate based on a few hardy
souls who are able to tolerate high noise levels,
and knee-jerk voting from the rest. Most people
won't participate, leaving decisions, if based on
a majority vote of those voting, and with a very
low quorum, being made by a handful of people
with most not paying attention. Ahem. Wikipedia.
*It works, but very poorly and unreliably.*

>J<https://www.facebook.com/jim.mueller.5680>im

>Mueller With winner-takes-all elections all of
>the state legislatures are controlled by one of
>the other of the 2 dominant political parties in
>this country. When a committee is appointed, if
>it has an odd number of members the majority
>party will have the higher number of members.
>Most of the time the minority members have no
>power. When there is an even number, the
>majority party will have an advantage of two or
>more over the minority party. The minority party
>members will rarely if ever be able to influence
>the outcome of any vote.

><https://www.facebook.com/groups/152631684794770/permalink/483547471703188/?comment_id=483567151701220&offset=0&total_comments=6>Yesterday
>at 7:28pm ·
><https://www.facebook.com/groups/152631684794770/#>Edited
>· <https://www.facebook.com/groups/152631684794770/#>Like

Again, Mueller is pointing out that the Banzhaf
problem already exists. It is not really a
problem, it's a political fact that if you are in
some small-minority faction, you have less power.
Banzhaf proposes "none," but that completely
neglects the power of participation, specifically
the power to enter motions. While, in the
designed committee, A and B may super-majority
suppress any motion, it's very close. C and D
have 33% of the voting power, just short of an
ability to resist a supermajority move. A and B,
by rights, may decide they don't want to consider
*any* motion, and in so doing they represent 67%
of the voters. It's certainly not unfair.
However, it would be discourteous of them to
refuse to hear D *if* C seconds it. Procedurally,
and routinely, D could enter a motion and C could
second. If A or B second, it can't be suppressed,
but that it could be suppressed by agreement of A
and B does not mean that it would be. More
commonly, it might be tabled (which only takes a
majority anyway). And then D and C could discuss
it informally and try to find some support from A
or B, to pick it up from the table.

The alternative for this committee is that D is
not on the committee at all. In which case, D
voters might support one of the others.

><https://www.facebook.com/jim.mueller.5680>Jim

>Mueller The example given is similar to the
>Nassau County situation that caused Banzhaf to
>challenge the validity of the power system.

><http://en.wikipedia.org/wiki/Banzhaf_power_index>http://en.wikipedia.org/wiki/Banzhaf_power_index

>In that case 9 of the 30 votes were in the hands
>of one person, another

>...<https://www.facebook.com/groups/152631684794770/#>See
>More<http://en.wikipedia.org/wiki/Banzhaf_power_index>
><http://en.wikipedia.org/wiki/Banzhaf_power_index>Banzhaf

>power index - Wikipedia, the free encyclopedia
>en.wikipedia.org The Banzhaf power index, named
>after John F. Banzhaf III (originally invented
>by...See More

><https://www.facebook.com/groups/152631684794770/permalink/483547471703188/?comment_id=483580035033265&offset=0&total_comments=6>Yesterday
>at 8:04pm ·
><https://www.facebook.com/groups/152631684794770/#>Like
>·
><https://www.facebook.com/browse/likes?id=483580035033265>1
>· <https://www.facebook.com/groups/152631684794770/#>Remove Preview
Banzhaf's analysis was defective. "Unfair,"
unless by comparison, implies some absolute
standard of fairness, which does not exist. The
alternative would presumably be a system where
each district elects one representative (the
Wikipedia article does not state what alternative
was proposed). Weighted voting is used in
corporations, where each shareholder exercises as
many votes as they own shares.

Now, I will agree that there is an unfairness in
the system. Each seat on the Board is elected by
majority vote in the district. This is very
different from what was proposed by Waugh on the
CES list. The unfairness arises in the
winner-take-all system, and that unfairness is
woven heavily through American political systems.
What's the alternative? Well, it could be
redistricting, where districts are amalgamated so
create roughly equal populations for each
district. However, different districts have different voter turnouts.

Asset is a system whereby all voters are
represented in the election process, allowing all
seats to be elected with equal numbers of votes.

In any case, election by majority in the
districts, as drawn, could leave as many as half
the voters -- or more, if plurality election is
allowed -- with *no* representation.

><https://www.facebook.com/marcia.everett.3>Marcia

>Also with C.o.R. the votes are actually what the
>public really want, so if only 12 people want
>the bill why should it pass? or be blocked?

><https://www.facebook.com/groups/152631684794770/permalink/483547471703188/?comment_id=483998244991444&offset=0&total_comments=6>7
>hours ago · <https://www.facebook.com/groups/152631684794770/#>Like

Right. However, functionally, arranging that
seats in a committee or assembly have equal
voting power fits better with how people actually
work, if it can be done. The representative
voting system proposed actually does not match
the Banzhaf problem, because there is no loss of
voting power, and because, very likely, there
would be a large number of eligible candidates.
The system has a very *different* problem, which
would become apparent as the number of voters and
candidates rises, as could be expected.

That could be fixed by requiring some minimum
number of votes to actually sit on the committee
or assembly. Those with less than those votes
could still participate remotely in actual
voting, but they would have no right to enter
motions or address the "committee of the whole"
without its permission. And, I'd suggest, they
could transfer their votes to others, to create more seats with the minimum.

In Asset, the "minimum" is the "quota." If
someone has more votes than the quota, or has
been given these votes with the right of
transfer, they can then help elect more than one
seat. I call those with votes "electors," and
electors might not necessarily choose to sit on
the committtee. After all, it's work.

Asset creates seats with equal default voting
power. If direct voting is allowed, it remains
equivalent in terms of actual voting power, to
the proposal. I don't know what C.o.R." means.

[Abd ul-Rahman Lomax]

At 11:13 PM 2/11/2013, William Waugh wrote:

><https://www.facebook.com/marcia.everett.3>Marcia

>Also with C.o.R. the votes are actually what the
>public really want, so if only 12 people want
>the bill why should it pass? or be blocked?

><https://www.facebook.com/groups/152631684794770/permalink/483547471703188/?comment_id=483998244991444&offset=0&total_comments=6>7
>hours ago · <https://www.facebook.com/groups/152631684794770/#>Like

Ah. Choice of Representation. It's more commonly
called proxy voting. Proxy voting with a secret
ballot doesn't have a special name, but Asset
Voting uses it. Asset was designed as a tweak on
STV, what to do with exhausted ballots? Dodgson
(Carroll) somewhat assumed that these ballots
have a single vote on them, and proposed that
this candidate be able to exercise the votes.

I've pointed out that if voters can simply vote
for one, there is no need for STV. Basically, I
conceive of an Assembly as a place where
representatives of the voters make choices, and I
simply extend that to an electoral college, where
chosen representatives of the voters cooperate to
choose members of the Assembly. And with direct
voting, but with elected seats voting by default
to represent those who chose them, we have a
hybrid that combines direct-representative and representative democracy.

Agitating for this in public elections is *way*
premature. The place to propose and use this is
in small private organizations, especially new
ones. You may find that with established
organizations, it's very difficult to implement a
reform like this. Essentially, it would establish
full equity, and that will necessarily remove
some measure of power from those who have power
in the existing structure. Who, by the terms of
the problem, have more power. This was called the
Lomax effect when I suggested it years ago.

They may naturally resist it. It is not that they
are greedy, or evil. It is that they believe,
typically, that they know best, and they might
even be right. However, it's possible that they
would understand that full representation would
increase the power of the organization, and thus
they might be willing to trust the general membership.

Assuming secret ballot, Asset can still create a
structure where every voter knows who,
specifically, represents them. (or at least
approximately so). It creates, around a central
committee, a penumbra of known public voters, and
every seat knows which of these elected him oir
her. This creates a much larger group that is,
essentially, a watchdog over or support for the
central committee. The individual voters,
particularly through the elector whom they voted
for -- they are probably friends or at least have
some natural rapport -- can influence central decisions, in an ongoing way.

I.e, individual voter sees some decision being
discussed and has an idea. The voter contacts his
or her elector, who contacts the seat. The seat
will listen to the elector, very likely, and the
elector will listen to the seat. But they will
not rubber-stamp, because if they routinely pass
on unripe or poor ideas, they will find
themselves unpopular and their opinions deprecated.

If this does not work in voluntary organizations,
it would not work in public structures.

If it works, it will be imitated.

[Marcia]

On Tuesday, February 12, 2013 at 11:38:41 AM UTC-5, Abd ul-Rahman Lomax wrote:

At 01:57 AM 2/10/2013, Markus Schulze wrote:

REPLY to Markus Schulze from  http://thirty-thousand.org

Sunday, October 11, 2009

Freedom and Legislative District Sizes

Is there a direct relationship between the total number of people per representative and freedom? Consider the tale of two states: California and New Hampshire. As explained later in this article, various freedom rankings give California a very low score with respect to the freedoms enjoyed by its citizens; in contrast, “Live Free or Die” New Hampshire scores high in the freedom indices. As of the 2000 Census, California has a total population of approximately 34 million. And New Hampshire’s total population is approximately 1.2 million (about the same as the city of San Diego). Now we get to the interesting part. California’s huge population is represented by 80 legislators in the lower house of their state legislature. Given that fact, how many legislators would you expect to find in the lower house of tiny New Hampshire’s state legislature? 30? 60? 120?

New Hampshire’s 1.2 million citizens are represented by 400 legislators in the lower house of their state legislature! Compare that to the 435 representatives who, ostensibly, represent 300 million Americans in the federal House of Representatives.

Thirty-Thousand.org (TTO) contends that reducing the population size of legislative districts forces the representatives to be more accountable to the citizens they are elected to represent and, as a result, smaller legislative districts foster greater individual liberty. Conversely, as legislative districts grow larger, so does the tendency for the government to become increasingly autocratic.

Though the relationship between district population size and freedom may not seem obvious, it is easy to understand by considering this thought experiment: imagine a continuum between two extremes of representation. On one end there is a highly representative legislature comprised of relatively small electoral districts with one representative for every 1,000 citizens. At the other end of the spectrum, there is a government ruled only by a single elected official (i.e., a governor). It is easy to imagine that the highly democratic government would afford more liberties to the citizenry, whereas the oligarchic scenario is likely to be highly statist.

A new TTO report, “State Freedom Indices and Legislative District Population Sizes”, establishes that there is indeed a significant correlative relationship between smaller district sizes and increased freedom.¹ This relationship can be shown using a simple analysis which relies upon a set of rankings known as “freedom indices”.

What is a “Freedom Index”?

A freedom index ranks all the states (which comprise the U.S.) according to the degree of freedom permitted by each of their respective governments. Identified in the TTO report are various freedom indices which focus on specific areas such as economic freedom, regulatory policy or personal freedom. For each category, the states are evaluated relative to a variety of pertinent criteria in order to determine their respective freedom scores. The evaluation criteria used to arrive at the freedom scores are fundamentally libertarian in nature; that is, greater personal and economic freedom is attributed to reduced governmental restrictions and taxation.

Legislative District Population Size

The average legislative district size for each of the fifty states is shown in the chart below.

A state’s district population size is simply its total population divided by the number of representatives in the lower house of its legislature. In addition to the lower house district size, the chart also provides a secondary representational ratio as a point of reference. (The secondary measure is equal to the state’s population divided by the total number of representatives in both houses of its legislature.)

Freedom and District Size

The TTO analysis examines each freedom index relative to the population size of the states’ legislative districts. Because the size of the legislative districts varies widely from state to state, the fifty states provide an ideal laboratory for evaluating the correlation between district population size and freedom. This simple analysis can be illustrated with the example already provided by California and New Hampshire.

In all of the freedom indices evaluated, California ranks near the bottom (least free) while New Hampshire is always ranked as a high-freedom state. California has a total population of approximately 34 million which are represented by only 80 legislators in the lower house of their state legislature. Therefore, the average size of their lower house district is 424,135. New Hampshire’s population of approximately 1.2 million is represented by 400 representatives. Thus, New Hampshire’s average district size is only 3,096. Consequently, even though California’s total population is 27 times larger than New Hampshire’s, California’s electoral district size is 137 times larger!

The fact that high-freedom New Hampshire has very small electoral districts while low-freedom California has very large districts begs this question: is that a random occurrence or is it indicative of a broader underlying relationship?

The Analysis

Using the various freedom indices cited within the report, all fifty states are ranked from most free to least. That list is then divided into thirds: the most-free group, the middle group and the least-free group. For each of those groups, the average size of the state legislative districts is calculated.

One of the freedom indices evaluated was “Freedom in the 50 States: An Index of Personal and Economic Freedom” which was produced by the Mercatus Center at George Mason University.²They report on five different indices as shown in the chart below.

In order to explain this analysis, let’s take a look at the “Overall Freedom” index, the fifth index in the chart. Note that the average district size of the “least free” group, 73,136, is substantially larger than that of the most free group (45,842). In fact, it is 60% larger!

For all of the indices shown in the chart above, the average size of the least-free group of states is significantly larger than that of the most-free group of states. As shown in the TTO report, the freedom indices provided by two additional reports (from two other think tanks) produced similar results with respect to the least-free districts being significantly larger than the most-free districts.

Conclusion

If there were no relationship between the states’ district population sizes and their freedom indices, then we would expect this analysis to produce fairly random results among the three groups (i.e., low freedom, medium, and high freedom). Clearly this is not a random relationship: 15 different freedom indices (from three different reports) were evaluated and, in every case, the average district size of the least-free group of states was significantly larger than that of the most-free group.

This analysis establishes that there is a significant correlative relationship between smaller district sizes and increased freedom. Moreover, Thirty-Thousand.org contends that this is a causal relationship; that is, as the legislative districts become larger, the government becomes increasingly oligarchic and statist.

© 2009 Thirty-Thousand.org, Inc.  

[Marcia]

First that reply above should be to Abu last post. Not good with this type reply system.  This is how I understand what Jim Mueller would like with Choice of Representation (CoR)  

Explaining Choice of Representation

By Marcia Everett on Tuesday, August 5, 2014 at 6:43pm

Districting is done to satisfy the "one man, one vote" concept. It makes the representatives equal but makes the citizens unequal-some have the representation of their choice and most do not. The theory is that because the population in each district is nearly equal then each person has the same proportionate share of a representative (a representative that they may have voted against and vehemently dislike).

Gerrymandering aside, great expense and effort goes into making legislative districts as close to equal in size of population as possible. The result is that many counties and municipalities are divided up into more than one, sometimes several districts. This creates more work and expense for the Election Officials in printing notices and ballots, setting up the machines, etc. The existence of districts makes campaigning more expensive by dividing up media markets into several districts or having a district span parts of several markets. 

So all of this work goes into making them equal and what is the first thing that they do? They give up a lot of their power by electing a leader who gets to decide a myriad of details including who is on what committees, which offices legislators are assigned to and what legislation will be scheduled for a vote and which bills will die without being voted on.

 In our relatively free market system people obtain the goods and services of their choice that they can afford to purchase or retain. Representation (Attorney, Personal Representative, Executor, Proxy, Agent, Guardian) involves the agent being legally bound to conduct themselves in the best interest of the client.

 In our election system, the government designates people as representatives even though the client (voter) is certain that the “representative” will not act in his or her best interest and will use force if necessary in order to prevent the other candidates from representing the people who voted for them.

That's Forced Representation.

By declaring that people have the Right to CHOICE OF REPRESENTATION, we establish a principle which people can aspire to, we expand the number of people who care about the outcome of elections and we show that this is not a partisan effort.

CHOICE OF REPRESENTATION makes the voters equal.  They all get the Representation of their Choice and make the representatives unequal by assigning them voting power equal to the number of people who voted for them.

In discussions of CHOICE OF REPRESENTATION, we use the term Representative Denied Office (RDO) for those people who have run for office, received votes and have been denied the opportunity to effectively represent the people who voted for them.  The RPO’s (Representatives Permitted Office) are sworn in.

Personally I think that having a representative of your choice is much more important than having the same number of people in each district. 

[Brian Langstraat]

Abd ul-Rahman Lomax,

In regards to your Posts about Asset voting:

I am investigating a similar concept in the Topics "Write-In Asset Voting" (https://groups.google.com/forum/#!topic/electionscience/JmJ54V9oQ6M) and "Write-In Delegated Full-List Transferable Voting" (https://groups.google.com/forum/#!topic/electionscience/jJ6fZuMyQH0).

Your concepts of assets instead of individual votes and Seats/Electors could easily be incorporated into these Write-In voting systems.

Overview of WIAV:

Voters would fill-in ballots by writing-in candidates.

The total votes for each candidate are their assets.

Candidates negotiate with other candidates and redistribute assets to elect the winner(s).

Overview of WIDFLV:

Any potential candidate would submit and publish their Full-List Transfers (FLT).

Voters would fill-in ballots by writing-in a candidate.

The total votes for each candidate are counted.

Votes for losing (or exceeding winning threshold) candidates are delegated per their FLT.

Candidate(s) that exceed the winning threshold are the winner(s).

Bruce R. Gilson,

In regards to your Post:

If you bel;ieve that a group with 35% of the citizenry behind it should win 35% of the time, then you might rely on the Banzhaf power index (actualy, this is something of an oversimplification, but very close to what it means). But the only way you could really accomplish this is to eschew voting and simply put everyone's name on a slip of paper, drawing one from a bin whenever a question is to be decided, and letting the named person make the decision. This is not a system which have heard of anyone advocating.

I am investigating a similar concept in the Topic "Write-In Lottery Voting" at https://groups.google.com/forum/#!topic/electionscience/ihbfOStZSMw.

Overview of WILV:

Voters would fill-in ballots by writing-in a list of candidate(s) up to the number of winners.

Ballot(s) are randomly selected in a lottery equal to the number of winners.

The top-listed, non-winner candidate on each selected ballot would become a winner.

Randomness has been used for selecting representatives in democratic societies (especially to avoid corruption), including by ancient Athenians and modern jury selection.

https://en.wikipedia.org/wiki/Sortition