On Jul 16, 2008, at 16:53 , James Gilmour wrote:
raphfrk > Sent: Wednesday, July 16, 2008 1:48 PM
From: James Gilmour
There is always a trade-off between guaranteed local representation
(small districts) and proportionality (large districts),
whatever the voting system.
Local representation isn't that important. The benefit that
is called 'local representation' is the ability of the voters
to directly control their representative.
My 45 years of campaigning for practical reform of voting systems
lead me to disagree with you, at least so far as the UK is
concerned. I suspect much the same applies in other countries that
have had many decades of the appalling British legacy of
single-member districts - hence the attraction of MMP, that APPEARS
to offer "the best of both worlds". It is not just direct
control and accountability - there is definitely a geographical
localism in what electors say they want.
However, ignoring logistics, PR-STV with a single
constituency would not have any tradeoff. There would be
(near) perfect PR and each voter would be directly
represented by someone they choose.
For the reasons given above, I cannot agree with this statement.
There is always a trade-off between overall proportionality and
local representation. In the 1890s some UK advocates of STV-PR
wanted the whole UK to be one electoral district, so the voter in
Caithness (extreme north) and the voter in Cornwall (extreme south)
could both vote for and be represented by the candidate from
Cambridge (middle England) if they thought that candidate would
best represent their views. Of course, practical reformers never
promoted such "perfect PR" ideas because they realised the
importance of geographically local representation to the real
electors in
the real world - to say nothing of the opposition from the
politicians (who cannot be ignored if you want to achieve reform).
These ideas are all non-starters, at least in the UK. Our electors
like candidates tied to specified geographical areas (even if
these areas are multi-member electoral districts).
I can see three different local/personal representation concepts
here. (just to clarify my thoughts, and maybe help some others too)
1) Single-member district related
- the basic philosophy/benefit is to keep the distance between
representatives and their voters/citizens short
- the voters know who their representative is (but the representative
may represent "wrong party")
- the representatives know what group of citizens they represent
- the representatives probably supports many local activities (this
may be good and bad)
2) Multi-member district related (open list assumed in the
explanation below)
- the basic philosophy/benefit is to keep all parts of the country
represented (also single-member districts do this)
- the voters know whom they voted (if that candidate was not elected
then their vote was counted for the other representatives of the
"right party") (only a mental link without any more concrete impact)
- the representatives do not know which voters voted for them or
their party (the voters can thus contact any representative of this
district as "their own")
- the representatives probably support many regional activities (this
may be good and bad)
3) Personal vote related (as in STV-PR without districts)
- the basic philosophy/benefit is to allow voters to pinpoint their
representatives without being bound to a party structure and districts
- the voters know which for which representative their vote was
counted for (only a mental link without any more concrete impact)
- the representatives do not know which voters voted for them (the
voters can thus contact any representative (that is the name as
none:-)) as "their own")
- the representatives do not represent any regions (regional
proportionality not supported)
Note that the needs of close relationship between a representative
and his/her voters, regional proportionality and ability to vote
without being bound by the party structure or by the districts are
all quite separate targets.
Note that there is a strong correlation between candidates that some
voter votes and the ideology that they represent. It is thus typical
that a vote to a candidate is also a vote to the ideology/party that
he/she represents. The total freedom of STV-PR to vote any candidates
of any party may thus quite often be just a nice option but not very
needed (typically just low level of noise in the results). Elected
candidates probably work for their party anyway, so one can not get
rid of these bindings even if the voters would be allowed to bypass
them in their ballot.
The freedom of voters to determine the order of inheritance of their
vote in STV-PR is maybe more meaningful within the party that the
candidate belongs to (vs. between parties). The voter can make sure
that his/her vote will go e.g. to the right wing of his/her party by
listing only candidates that he/she assumes to belong to that wing.
Normal party lists do not support this. But it is possible to extend
them so that they do (e.g. by using subgroups or a tree like
structure within the party). The difference between these two
approaches is again "more freedom" vs. "clear groupings" (that keep
the ballots simple, help the voters to understand the bindings of all
of the candidates, and after the election bind the representatives to
their announced policy).
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
