[EM] Another PR method based on ranked ballots

[Steve Eppley]

On 20 Mar 2003 at 10:02, Forest Simmons wrote:
> Here's another idea for PR that makes use of Borda style ballots:
-snip-

I like PR methods in principle, because like the best single-winner 
methods they don't suffer as much from spoiling.  But primitive PR 
methods tend to distort the public's perception of "popularity" by 
equating the number of seats won by a party with its popularity.  
Typically, the party that wins the most seats forms controls the 
agenda and, in parliamentary systems, also controls the executive 
branch.  For example, the Nazi party in 1933 won the most seats and 
went on to be granted a lot of power, but they won a minority of 
seats and were probably the least preferred party of a majority of 
the voters.

A party that wins fewer seats could be preferred by a majority over 
the party that wins the most seats, and in that sense is more 
popular, and should be distinguished as such by the voting system.

So I prefer a PR system such as the following: Each voter ranks the 
parties from most preferred to least preferred, then seats are 
awarded to each party in proportion to the number of voters that 
ranked it topmost, with the following exception:  The voters' 
rankings would also be tallied by a good single-winner method (such 
as MAM), the winning party according to that method would be awarded 
enough extra seats to make it the largest party.  Also, that party 
would be given control over the legislative agenda (assuming the 
legislature operates under typical primitive rules like Robert's 
Rules of Order that require agenda control, instead of a modern 
asynchronous online system that allows each member to propose issues, 
propose alternatives, post arguments, and rank the alternatives).

Another advantage of allowing each voter to rank the parties is that 
the voters' relative rankings of the viable compromises become known, 
and that may be much more important that learning each voter's 
favorite party.

-- Steve Eppley