[EM] In defense of the Electoral College (was Re: Making a Bad Thing Worse)

[Aaron Armitage]

--- On Fri, 11/7/08, Jonathan Lundell <jlundell at pobox.com> wrote:

> From: Jonathan Lundell <jlundell at pobox.com>
> Subject: Re: [EM] In defense of the Electoral College (was Re: Making a Bad Thing Worse)
> To: eutychus_slept at yahoo.com
> Cc: "Chris Benham" <cbenhamau at yahoo.com.au>, election-methods at lists.electorama.com
> Date: Friday, November 7, 2008, 10:47 AM
> On Nov 7, 2008, at 8:07 AM, Aaron Armitage wrote:
> 
> > A national plurality
> > election only gains anything in the very rare case
> that the EC result
> > would have been different
> 
> That's true only if one assumes that both candidate and
> voter behavior is the same under NPV and EC, and that's
> really in indefensible assumption in both cases.

I should have specified that I meant gains in terms of better outcomes,
and before 2000 the last election in which a candidate carried the EC
without also carrying the popular vote had been over a hundred years
before. So from that perspective it seems behavior won't change that much.

If you place a value on a voting system because you approve of its
behavioral effects even when those effects have no impact on the outcome,
it's not clear which way that argument cuts. Most of the arguments
defending the electoral college are of the same kind. Maybe what you see
as a closer proportion between population and political importance others
will as ignoring rural voters altogether. You might not like a strong
focus on swing states, but it seems likely that conservative Ohio voters
were less likely than conservative Idaho voters to think Bill Clinton was
sending black helicopters to round them up and send them to secret FEMA
camps, and likewise it seems likely that liberal Ohio voters are less
likely than liberal San Francisco voters to think George Bush should be
executed for war crimes. 

We won't know which case is actually better until we make the experiment,
but there are very good reasons to think that even making the most
optimistic assumptions about what effect a national plurality would have,
the gains will be small compared to a popular vote using other systems,
especially Condorcet. For example, in Condorcet it makes a lot of
strategic sense to appeal to voters who will never rank you first, and
even would never approve of you.