[EM] Minmax under-representaton causes small-bias

Abd ul-Rahman Lomax abd at lomaxdesign.com
Tue Feb 6 08:44:08 PST 2007


At 09:37 AM 2/6/2007, RLSuter at aol.com wrote:
>Mathematical analyses of this issue that are not informed and
>tempered by a good understanding of the issue's history and
>political consequences are close to worthless, if not worse than
>worthless in that they muddy the issue and are used to justify
>one of the most undemocratic aspects of the U.S. political
>system and the least representative legislative body of any
>major nation (and possibly any nation) in the world.

There are two activities of value here: one is theoretical 
mathematical analysis and the other is practical, having to do with 
applications of election methods and, in this case, apportionment 
issues and their effects. The activities are independent, and it is 
not correct to claim that the first activity, in particular, is 
worthless merely because it does not currently impact the second.

This list, in general, focuses on theory. If you want more focus on 
actual political work with Election Methods, the place to look is 
probably the Range Voting list on yahoogroups, which does not confine 
itself to Range Voting but is concerned with election and related 
reform in general.

Now, there is no doubt but that the Senate is in some respects 
"undemocratic." It wasn't intended to be the representative of the 
"people." It was intended to lend balance, "gravitas," to be an elite 
institution. I'd suggest that any replacement for the Senate should 
ideally perform a similar function. As an example of a reform that 
could move the Senate toward democracy without sacrificing its elite 
and state-representational character, the state representation for 
states below a certain population could be reduced to a single Senator.

But, still, the Senate is vulnerable to shifts in the majority, just 
like the House. Some kind of proportional representation could be 
more stable. (When you have representation by district, elected by 
majority vote within each district, a small shift in which party 
holds the majority can create a drastic shift in party 
representation. This would seem to be undesirable if gravitas is what 
we want.) I'd suggest, in fact, Asset Voting with the Senate being 
elected nation-wide. Asset Voting for PR has the benefit that no vote 
need be wasted. You end up with a Senator who either you picked, or 
someone you picked, picked. And you know exactly who your Senator is. 
And quite likely your Senator would be based relatively close to you, 
unless you find yourself affiliating with some group that is thinly 
spread about the nation, in which case all of you could share one 
Senator. Asset, if the candidates use vote transfers by precinct, 
essentially is districting-on-the-fly; that is, a Senator could 
generally represent a set of precincts geographically contiguous, but 
would not be limited to that.

Personally, I don't understand all the fuss about more fixed PR 
methods when Asset is such a brilliant solution, one which Warren 
should more frequently take credit for. (His Asset Voting is more 
complex than what I generally proposed, which is Fractional Approval 
Asset Voting, where the ballot is essentially an Approval ballot, 
but, because no vote is wasted, unlike standard Approval 
single-winner, the votes are divided by the number of approved 
candidates on the ballot. I think most voters would pick just one and 
leave it at that.)




More information about the Election-Methods mailing list