[Election-Methods] method design challenge + new method AMP

[raphfrk at netscape.net]

> you wrote
> > If a majority has a 100% chance of getting their candidate elected,
> > then there is no incentive for them to trade.? If the voters are 100%
> strategic, they will know this.
> 
> Yes, although some Range Voting supporters try hard to convince us of 
> the opposite, it seems.

Well, your example is under the assumption of perfect information.  

I think in the near perfect balance of your example and without perfect 
information, range or approval would probably elect the compromise as a 
reasonably large number of voters would approve the compromise as well 
as their favourite instead of risking their hated candidate wins.  As 
the majority for A gets larger (say >55%), A's supporters would be advised
to switch to bullet voting just for A.

I think that approval is a good and practical voting method.  

However, the ideal method is one that gives people an incentive for 
honest ratings and elects the candidate that is socially optimal, i.e
a candidate who would be a 'Kaldor-Hicks optimal' candidate.  If there
is more than one optimal, then use random ballot to decide which one,
but a tie isn't that likely.

One issue with random processes is that they don't work well for a 
legislature.  A majority would just keep asking that the vote be repeated
until they win it.

Saying that a re-vote cannot occur unless the situation changes would
require that a definition of a change in the situation be decided.

Also, people have a certain level of distrust for random processes.
I don't think people would accept a President who was elected even
though he only had a 1% chance of winning.  I am not sure what the
threshold is before it would be acceptable (some people would object
to a 49% candidate winning instead of a 51% candidate).

> > Optimal utility via trade requires that voters have something to
> > trade, and fractions of a win probability seems to be quite a
> > reasonable solution.
> 
> I cannot really imagine any other thing unless we consider money 
> transfers...

There is a method called CTT voting where money is used and in theory,
it provides an incentive for everyone to be honest.

However, it requires each voter to act completely independently.  Also,
it weakens the secret ballot.

Also, it only requires payment in a small number of cases, where the
result is close.  In a landslide, nobody has to pay.

http://rangevoting.org/CTT.html

It would probably not work in practice, especially for large
elections.  Most of the issues are listed on the linked page.

For a legislature, you could trade future votes instead of money.  
One issue is that it needs to be a secret ballot for CTT to work,
(otherwise parties can control the votes by disciplining party 
members who vote the 'wrong' way after the vote is completed and
made public).  However, it needs to be a public ballot so that the 
voters know if their legislator is voting in their best interests.

In Germany, motions of no confidence in the Chancellor (head of 
government) are handled by a secret ballot that requires an absolute 
majority of their legislature to pass.  This is used to change the
head of government, so I guess not all legislatures have public 
votes for everything.




Raphfrk
--------------------
Interesting site
"what if anyone could modify the laws"

www.wikocracy.com

 


 


________________________________________________________________________
AOL's new homepage has launched. Take a tour at http://info.aol.co.uk/homepage/ now.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.electorama.com/pipermail/election-methods-electorama.com/attachments/20080508/f505ef84/attachment-0003.htm>