Sorry for some delay is replying.
(And thanks to all for the links. Those pages seem to include plenty
of good information. I may comment them later when I understand more.)
On Aug 18, 2008, at 2:49 , James Gilmour wrote:
Juho > Sent: Sunday, August 17, 2008 6:08 AM
To: Election Methods Mailing List
On Aug 16, 2008, at 0:51 , James Gilmour wrote:
Lists of
any kind will always be constraining. And they are unnecessary (as
well as, in my view, undesirable).
Constraining in the sense of not being most flexible, yes.
Why do you see lists as undesirable?
Because they shift the balance of power and accountability away
from the voters in favour of the party machines that make and
register the lists.
Yes, in many cases the level of partly control exceeds what is good
for the society. But this question is not black and white with
respect to the election methods. Open lists may typically have lots
of candidates to choose from, including also candidates that are
popular but only weakly tied to the party. This is because parties
try to collect all possible votes they can (by using many different
kind of lures that appeal to different kind of fish). On the other
hand it seems that at least in some STV-PR implementations the number
of candidates per party is small, allowing the party insiders to
efficiently control who is on the "list" and who will be elected.
Rules for registering candidates may be different in different
countries and may also be method independent in many cases. Parties
may often have a formal role, but I don't know what the typical rules
in STV-PR countries are.
To move from closed lists to open lists would certainly be
progress, and here in the UK we are campaigning for that, but only
as a
very much second-best reform. If we know, as we do, that STV-PR
will deliver what is really needed, why promote something less? It
is amazing how often the opponents of voting reform in the UK (and
elsewhere) play the "simplicity" card to justify FPTP in
single-member districts!
Open lists and many others would indeed be quite as simple for the
voters. STV is not too difficult for the voters either (specially if
the number of candidates is low).
Btw, one thing I like in trees is that they can express clearly that
the voters want party X to be more Y. If branches promoting Y get
many more votes than in the previous election it is hard for the
party to avoid making conclusions out of that. Also the
representatives of the "Y branches" are morally bound to promoting Y.
In politics it is too common that politicians talk nice things about
Y without committing too deeply, and when the time of action comes
they are free to find new argumentation to support whatever decisions
they make (still "in principle having positive feelings about Y
but..."). Having majority of the candidates behind Y would make
things clear.
Open lists do not support party internal proportionality (between
segments/branches). STV in principle does but it may be too complex
to follow what the opinions of the voters really were at the election
day and it doesn't bind the candidates to Y (in the sense that
positive words before the election may be all that the voter gets).
I just note here that it may be useful to have districts that are
about equal in size to keep the "party cutoff levels" (=
proportionality level with respect to minority opinions) roughly at
the same level everywhere (or alternatively use some additional
balancing mechanisms).
This concern with "equalness" can become something of an obsession,
especially where there are differences in support for parties
between rural and urban areas and the proposal is to have larger
districts (more members) in the urban areas and smaller districts
(fewer members) in the rural areas. It has been an issue here in
Scotland. But there is more to equality of representation than
having exactly equal numbers of electors per elected member and
having equal numbers of elected members in every electoral district.
The realities and practicalities of effective representation in the
geographically different areas should not be ignored. No matter
how hard you strive for "equalness" in every parameter, the voters
will screw it up for you because you cannot ensure equal turnouts
in all of the "equal" districts (unless you have compulsory voting,
perhaps!).
Large differences in size between the districts may lead to various
imbalanced situations. It is e.g. possible that the Greens get seats
in the cities but votes to them are lost votes in the rural areas.
This is the key reason behind current reform plans in Finland. It is
also not quite fair that a new "pro city" party would get many
representatives but a new "pro countryside" party would get none
(with same level of support).
There are however ways to fix this in the list based methods even if
the districts sizes differ, but in STV this is probably trickier to
do (having roughly equal size districts would be easy).
I tend to favour counting exact proportionalities at national (=whole
election) level ((if one wants PR in the first place)). I also tend
to think that most old stable democracies do not need explicit or
implicit cutoffs "to maintain the stability of the system" since most
of them seem to have more problems with having the same old boring
parties in power continuously rather than having problems with
fighting against too many diverse viewpoints. I also tend to favour
more fine-grained expression of opinions, as in STV or with trees, as
a way to allow the voters to better influence the direction the
system takes (reduces the risk of stagnation and alienation of the
voters from the "parties and politics that continue as before no
matter how we vote").
I don't know what "exact proportionalities" might be when a
preferential voting system is used.
I think it should be pretty much the same thing. Roughly, if n% of
the voters rank "X style" candidates first then "X style" candidates
should get about n% of the seats.
And there are certainly major
problems in trying to measure such proportionalities, exact or
otherwise.
STV is in principle good at this (but multiple small size districts
may favour large parties).
Malta has got itself into some very serious political
problems by taking the first preference votes as the "correct"
expression of party proportionality that should be reflected in the
seats won. But that ignores completely the effect of the transfers
of votes according to the voters' preferences, the very essence
of STV-PR. We all tend to use the first preferences in STV-PR
elections as the "best" indicator of relative party support, but that
assumption must have so many caveats attached to it that no-one
would think of using it to calculate the "exact" proportionalities
at national level (except in Malta!).
That may be a reasonably close approximation of party support but
does not handle properly horizontal votes that are like
MyCommunistFriend>MyRightWingPartyCandidate1>... . I'm not sure how
common such votes are (and if there are many of them they might
cancel the impact of each others).
What is a better way to handle the bias that e.g. small districts may
cause in STV? Maybe an explicit party vote?? If one does not fix the
balance one maybe just accepts some randomness in the balance between
parties of same size and some systematic bias in favour of the large
parties (or whatever groupings/ideologies).
Juho
___________________________________________________________
All new Yahoo! Mail "The new Interface is stunning in its simplicity and ease of use." - PC Magazine
http://uk.docs.yahoo.com/nowyoucan.html
----
Election-Methods mailing list - see http://electorama.com/em for list info
