[EM] Northern Ireland - more comments

[Donald E Davison]

 - - - - - - - - - - - - - - - - - - - - - - - - - October 06 1999
Greetings,

     Stephen Todd sent me some comments concerning the North Ireland STV rules.

Dear Steve,

     I gave your comments due consideration and decided that you are
failing to understand the math of the Northern Ireland rules. Flawed math
cannot be denied.

     There are two disputes. One is the truncating of the transfer value to
two decimal points. The other is of more importance because it will result
in a greater flaw. That other dispute is the basing the transfer value on
only the transferable votes when surplus votes are being transferred.
     I am going to edit out your comparisons with the Hare Clark rules,
because I feel the Northern Ireland(NI) rules should stand on their own
when faced with my charge of corruption. Besides, I do not know the rules
of Hare Clark that you allude to.

Steve: You have misinterpreted the effect of the NI rules.  By truncating
the transfer value to two decimal points, the rules ensure that every
candidate receiving a transfer of votes, is neither advantaged nor
disadvantaged by the actual number of transferable papers received.

Donald: What you are saying here is not correct. When you round up or down
to two decimal points you will be increasing or decreasing the transfer
value per ballot. The candidate that is to receive the most transfer
ballots will gain or lose the most whole votes - he will be advantaged or
disadvantaged to the possibility of winning or losing an election that he
should have won or not won. Either way, it is a flaw.

Steve: By discarding the remainder when the transfer value is calculated,
rather than after the papers have been transferred, the NI rules ensure
that all candidates receiving papers, receive them in exactly the same
proportion as all other candidates.

Donald: This is mathematically true, but it is nothing to be proud of. The
rules are taking exhausted votes and giving them to the remaining
candidates. Under what logic is it claimed that the remaining candidates
are entitled to the exhausted ballots when their names cannot be found
anywhere on those ballots.

Steve: In short, the NI rules treat all candidates scrupulously fairly in
relation to each other, the Hare-Clark rules do not.

Donald: Again, what you are saying here is not correct. The NI rules are
handing out extra votes, which were salvaged from the exhausted ballots.
The candidate that is to receive the most transfer ballots will be
receiving most of the extra votes, maybe enough to win an election that he
was not entitled to win. And it follows that another candidate will be
losing an election that he should have won. This is not treating candidates
scrupulosly fairly in relation to each other.

Steve: "Neither rules are 'corruption'.  They both treat non-transferable
papers in a way that the voters who cast them would not necessarily want
them to be treated, ..."

Donald: Do you not see the conflict in what you are saying here? If the
rules treat the ballots in ways that the voters do not want them treated,
then that is corruption.

Steve: "...but the NI rules do not treat any of the candidates unfairly in
relation to each other.

Donald: Again, I must say that you are not correct. The NI rules give more
extra votes to the candidate that is to receive the most transfer ballots.
In order to treat all the candidates fairly, the rules would have to give
each candidate the same number of unearned votes.

Steve: It is quite wrong for you to suggest that the NI election is not a
normal STV election.  In respect of which publicly-elected body, anywhere
in the world, is your idea of a normal STV election conducted?  There is
currently no such body.

Donald: If this is true, then this condems STV as a valid election method,
it is infected with too many Dirty Little Secrets.

     My idea of a normal and ideal STV election is as follows:
     1. The election is held in only one area - no districts.
     2. No artifical threshold
     3. Hare Quota
     4. Fractional Transfer
     5. Transfer Value based on all ballots of candidate with surplus.
     6. Number of demical places to be equal to: Two plus the number of
digits of the votes of the candidate with the surplus being transferred at
the time. Example: If a candidaate had 123,456 votes, we should use eight
decimal places. This is necessary because the total of all the separate
amounts of the transfer of surplus votes should equal exactly to the
surplus votes. It is not acceptable to merely write off any difference as
`fractional error' or some such thing. That error could elect one candidate
and unelect another.
     The results of an ideal STV election should be as follows:
     1. Every vote ends up on a winning candidate
     2. Every elected member receives the same number of votes.
     But even ideal STV has a flaw - a big flaw. The ideal STV election
depends on ideal voters - voters that are able to make enough informed
choices so that no ballot becomes exhausted. This is not going to happen.
Most voters could not make two well informed choices, but they could make
an informed choice of political party, and they should be allowed to do so.
While the ideal STV is a very good election method, it falls victim to the
rule of `Garbage In - Garbage Out'.
     For that reason, I only support the ideal STV for small, one area,
non-partisan elections. For the larger and partisan elections, I support a
combination of STV and Open Party List in some method in which the voter is
allowed to vote for both candidates and/or political parties.
     And, if districts are to be used, the method should also balance
up(Top UP) the party proportionality across all the districts.

Steve: As I said to you in my previous note, hand-counting rules developed
in the way they did to meet the challenge of manually counting many
thousands of ballot papers as expeditiously as possible.

Donald: I support rules that make it easier to hand count the ballots. In
that view, I support the policy of eliminating candidates before the
transfer of surplus votes because it will reduce math, but that is a policy
that is honest, for it treats every candidate and voter with equality. Any
rule on hand counting can and should be honest, that is, the rule is to
treat every voter and every candidate and every party with equality. The
Transfer Value rule of the Northern Ireland elections does not do this.

Steve:  In fact, the 'Clark' in Hare-Clark is the name of the Tasmanian
judge in the 1890s who used judicial activism to amend the rules to make
them fairer, i.e. by getting rid of random selections of ballot papers for
transfer.

Donald: This was a good change, an improvement. Randon selection introduces
a flaw into the election, but the flaw is about the same size, or less,
than the flaw of rounding off the transfer value to two decimal points. In
other words, fractional tranfer value with two demical points is no
improvement over random selection. The hand counters may as well use random
selection if it will be easier for them.

Steve: It was perfectly reasonable for those who developed STV more than a
hundred years ago, to regard non-transferable papers as just that,
non-transferable, and to calculate transfer values in respect of
transferable papers only.

Donald: It is perfectly reasonable for us to believe that those who
developed STV rules more than a hundred years ago could be corrupt. What is
so special about them that they would not be corrupt. A hundred years does
not make one a saint, nor correct flawed math.

Steve:  As I said last time, you are being too hard on these rules (which
still give better and more satisfying results than any other system).

Donald: No, I am not being too hard on the rules - corruption is corruption.
     There are people who know this rule is corruption, and most of those
people are people who are in favor of this type of corruption, they are the
ones that benefit from it - the members of the large factions.

Steve: As I said last time also, the solution is Meek-style STV.  Even
under your rules, you would still be forced to calculate a transfer value
for a consequential (secondary) surplus by reference only to those papers
which gave immediate rise to that surplus.  But the other papers received
previously by the now-elected candidate, contributed to the surplus, too!
But they cannot be transferred.  To include them in the calculation of a
transfer value would be mathematically unsound.  Some ballot papers would
end up with an effective value in excess of unity, and some would end up
having an effective value less than unity. Meek-style STV overcomes this
problem as well.

Donald: I cannot comment on the rules of the Meek-Style STV, because I do
not know those rules, but I would hope those rules are better than the
Northern Ireland rules.

Steve: Please have a look at the meek.doc attached.  (I hope you can open
them.)  I have a computer program that implements the Meek rules - and it
is just brilliant.

Donald: I am unable to open attachments.

Steve: The concept of transfer values is turned around. Instead, a 'keep
value' is calculated in respect of each candidate.  In effect, the
'remaining value' is the old transfer value.
     If, for the purposes of this exercise, you can put aside your advocacy
of the Hare quota, I think you will agree that these [Meek] rules are the
very best yet devised.

Donald: Why should it be necessary for me to put aside my advocary of the
Hare Quota?  If your Meek rules are, as you say `the very best yet
devised', then they should work better with the Hare Quota, or at least,
equally well.

Steve: (I can send more information if you become interested.)

Donald: I have just become intersested. Please send me the basic rules of
your Meek-Style STV via email. One page should do it.

Steve: Please let me have your thoughts in due course.

Regards, Steve

Donald: Yes, I will let you have my thoughts about your Meek-Style STV, if
and when I understand its rules.

Regards, Donald


   +=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+
   |                         Q U O T A T I O N                         |
   |  "Democracy is a beautiful thing,                                 |
   |       except that part about letting just any old yokel vote."    |
   |                            - Age 10                               |
   +=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+

                            N E W S L E T T E R

                    Worldwide Direct Democracy Newsletter
                     Four Issues per Year by Postal Mail
             Cost per year: Czech Republic 200 Kc,  Europe 12 DM
                          Outside of Europe  $10

              Make check payable to: Mr. Bohuslav Binka
              Mail to:  Mr. Bohuslav Binka
                        Bellova 15
                        Brno 623 00
                        Czech Republic
          - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
                         N E W    D E M O C R A C Y
              A Source of Study Material for Political Change

                        http://www.mich.com/~donald
          - - - - - - - - - - - - - - - - - - - - - - - - - - - - -