A view from Finland
It's nice that the Citizen's Assembly has decided to endorse
proportional elections, but I'm a bit worried. How can they make a
sensible decision when there's so little written in English on
party list systems?
If British Columbia decides to use MMP, I hope you choose the
Bavarian system which has open party lists. I noticed
that MMP was criticized for combining first-past-the-post with
closed lists. That is not at all necessary, because it's quite
possible to have the open lists. Besides, the first-past-the-post
element in MMP is cosmetic, because constituency candidates usually
stand on the party lists.
Here's a very short description of the
Bavarian system
, just in case.
The link in the posting to the electoral law and regulations is now
dead and the site has moved to this URL:
They are in German, of course.
Open lists are most easily handled with the Finnish system of
blank ballots and single non-transferable vote within lists. Each
candidate is numbered and you write the number of your candidate on
the ballot. The system can handle municipal elections in Helsinki
with 80 seats and nearly 1000 candidates. Here's an image of our
ballot paper, which can be counted simply by piling up. The ballots
are folded, stamped and inserted into the ballot box. Envelopes are
not used.
We have information on the candidates in a reference sheet or
schedule posted on the wall of the polling booth. Here's part of
one:
It looks like a German ballot paper.
I don't know what happens in Bavaria, but with the Finnish
system of open lists, the parties try to find well-known
independents to stand on their lists. Many of these candidates are
good, because they have experience outside the party machine,
although there are people who don't agree with me.
To ensure local representation I would personally prefer
multimember districts to MMP. If they are of different sizes, a
possibility is Friedrich Pukelsheim's biproportional system
('Double Pukelsheim'), which is being introduced in Zurich. It has
no problems with overhang seats. There's nothing much on it in
English, though, and it's probably not relevant in British
Columbia, but BAZI, the Java election result calculator on Dr.
Pukelsheim's site is useful. It can calculate the results of most
systems.
Proportional representation can easily be watered down by a high
threshold. The high effective threshold of STV unfortunately has
the same result.
I've hesitated to write this because it's none of my business,
but I feel important things may have been overlooked. It's a
momentous decision if you start using PR in North America,
especially if it's a party list system.
levelCMS