[EM] 02/03/02 - STV for Candidate Lists:

[Adam Tarr]

At 06:17 AM 2/3/02 -0500, Donald Davison wrote:

>One: Its high proportionality is only for party.

How else do you define proportionality?  You can't define it on a 
per-candidate basis; candidates are either 100% elected or 100% un-elected..

>Two: The voter is not allowed to cross party lines.

True, this is a weakness, without a doubt.

>Three: The order of the candidate list of each party is suspect.

Only slightly suspect.  Think of Open List this way: once you have 
determined how many seats a party gets, you re-apply those votes in a SNTV 
election within the party to find which candidates take the seat.  While 
SNTV is not the best way to do things (STV is likely better) the distortion 
is not terribly bad.

>The mathematical correct way to determine the order of candidates for a
>party would be to use STV, but then why not use STV for the entire 
>election and drop Party List?

Well, in essence I agree with you.  However, there are practical reasons to 
do otherwise:

- Open party list can be implemented on even the most primitive voting 
equipment.

- Open party list ballots are as easy to tally as plurality, single-winner 
ballots are.  STV Ballots could be a nightmare to tally if used for, say, 
the 52 California representatives to congress.

- Open party list is extremely simple to use, and in fact the voter need 
not distinguish it from SNTV in order to vote in an effective 
fashion.  Moreover, while STV becomes unwieldy when the number of seats 
available becomes very large, Open List stays very easy to use.

- Its simplicity can make it easier to push its adoption.

But are you asking me whether STV is better for multi-winner 
elections?  Yes, sure it is.  Although I think you need to give the voters 
a crutch by allowing the to mix party lists into their vote list, or it 
becomes too much of a pain in a large district.

-Adam