[EM] Free riding

[Kristofer Munsterhjelm]

Raph Frank wrote:
> On 9/4/08, Kristofer Munsterhjelm <km-elmet at broadpark.no> wrote:
>>  Not necessarily PAV, but a method that's based on Approval and would
>> otherwise be as good as STV, if such a beast exists. What kind of strategy
>> can be used in PAV?
> 
> If a candidate is certain to win, then there is no point in voting for him.
> 
> PAV becomes party list PR if everyone approves the candidates from
> their party and doesn't approve anyone else.
> 
> It also has the same strategic problem that voters who vote for a
> party which doesn't win don't get any representation and can't
> transfer their vote.

Not really. The vote transfers happen indirectly through reduced 
satisfication scores. If a voter doesn't vote for a party and instead 
votes for a group of personal candidates, his satisfication score will 
be lowered for the potential assemblies with only party members in them, 
and increased for the assemblies with independents in them, and so the 
chance of the method picking one of the latter would increase, also.

As for the D'Hondt problem, I suppose it could be generalized to (C * 
v)/(C+s) instead of V/(s+1). I would then pick C = 0.5. However, this 
may still lead to a problem in the left-right-center edge case.

>> If this was a
>> single-winner method, you could have used a cardinal ratings (range)
>> equivalent of approval strategy A, but I'm not sure how you'd make a
>> multiwinner version of that.
> 
> Not sure, for multi-winner, there needs to be some kind of mechanism
> for deciding who has already obtained representation.
> 
> Perhaps, the approval winner is selected, and everyone who voted for
> them has their vote decreased as a proportion of how many votes they
> voted for that candidate.
> 
> This is basically RRV, but when the vote is split over multiple rounds.

One could also have a range version of ordinary (nonsequential) PAV, 
where v is replaced by the sum of the ratings for those that have 
already been elected.

>> Parties that run vote-management can't gain more
>> seats than they have, proportionally, unless they run some candidates as
>> independents. However, to do so would look somewhat dirty and the party
>> might lose genuine personal votes as a consequence.
> 
> This means that parties are considered primary.  Ofc, if you want to
> have a mix of personal and party based votes, vote management might be
> unavoidable.

It would enshrine the existence of such a structure as a political 
party, but I don't think it would be primary about parties. The idea is 
that parties are the ones that have enough organization to run a vote 
management scheme. Therefore, the party list aspect should limit the 
vote management by those from which vote management is most likely. 
Independents, who don't have the clout to perform vote management, would 
all be grouped within the "no official party" pseudoparty and the subset 
elected determined by STV just as it would be for parties (by running 
ballots with nonparty candidates through STV).

The hard part is finding a way of tallying group support. As mentioned, 
first preference wouldn't work, because then you'd get return of Woodall 
free-riding: parties would say "just vote for a friend as a write-in, 
then the order we give you".

> Another option would be Meek's method with max keep values.
> 
> For example (blanks assumed 1):
> 
> A B party supporter might vote
> 
> 1) A (0.75)
> 2) B1
> 3) B2
> 
> This means that they are willing to give 0.7 of their vote at most to
> candidate A.  If party A vote manages, they can't get more than 0.7 of
> that vote.
> 
> In effect, the voter has cast 4 votes
> 
> A>B1>B2
> A>B1>B2
> A>B1>B2
> B1>B2
> 
> This would mean more 'ordinary' methods would still be usable.
> 
> It would allow voters to give a fraction of their vote to a candidate
> who helped them (people would feel guilty not helping the candidate in
> some way).  Also, having the rule would make people aware of the whole
> problem in the first place.

I think that option is, like AERLO, ATLO, and approval cutoffs, too 
complex for the ordinary voter to understand; at least unless it was 
camouflaged as a rated ballot, but that would limit the voter's 
expression since they may want "A > B1 > B2" with B1 and B2 both at max 
transfer value 1, something a rated ballot with max transfer derived 
from the ratings couldn't support.