Hallo,
I tried to find out which method will be used
to determine the winning option in the upcoming
referendum. However, even the report of the
Special Committee on Democratic Renewal
doesn't specify which method will be used:
http://www.assembly.pe.ca/sittings/2016spring/reports/23_1_2016-15-04-report.pdf
I guess that the IRV supporters will argue that,
when preferential ballots are being used and
no method is specified to determine the winner,
then it is feasible to use IRV.
So let's presume that IRV will be used in the
upcoming referendum. Then I predict the following
outcome:
* "First Past The Post Plus Leaders" will be
eliminated first because the voters will
consider this to be a rigged proposal,
rigged to ensure that party leaders get elected
even when they fail to win a seat in a district.
* "Dual Member Proportional" will be eliminated
second because of two reasons. (1) Biproportional
methods are rather unknown in Canada and the
remaining 4 months will not be sufficient to
explain this method to the average voter.
(2) The concrete DMP proposal prescribes
closed party-lists and only one vote for each
voter, while the concrete MMP proposal prescribes
open party-lists and two votes (a district vote
and a party vote) for each voter.
* "Preferential Voting" will be eliminated third.
* So in the last round, the voters have to choose
between "First Past The Post" and "Mixed Member
Proportional". We already had this situation in
2005 when FPP won against MMP with 64% against 36 %:
https://en.wikipedia.org/wiki/Prince_Edward_Island_electoral_reform_referendum,_2005
Therefore, my criticism is:
1. The proposals are too similar. "First Past The Post
Plus Leaders" is too similar to "First Past The Post".
"Dual Member Proportional" is too similar to
"Mixed Member Proportional".
When the purpose of this non-binding referendum
is to find out what the citizens of Prince Edward
Island really want, then a larger number of options
from a larger range of election methods would be
needed (e.g. Condorcet voting, single transferable
vote).
2. It doesn't make much sense to use IRV to determine
the winning option, because IRV will always lead
to a situation where the voters have to choose
between the two most extreme proposals (FPP and
pure proportional representation) in the final round.
Markus Schulze