On Sat, Oct 18, 2008 at 8:13 PM, Jonathan Lundell <[EMAIL PROTECTED]> wrote: > On Oct 18, 2008, at 11:26 AM, Raph Frank wrote: > I'm still not getting it. Perhaps I'm not following the mechanism you're > suggesting.
I meant if they actually managed PR, but yeah, it is hard to come up with a specific compact that doesn't have defection incentives. > I do agree that there are cases where a proportional EC with free-agent > electors could have a better (in the sense of more democratic) result than > FPTP--say in 1992, where FPTP elects Clinton, but a PR EC elects Bush1 by > combining Bush and Perot electors, or in 2000 Nader+Gore electors defeat > Bush2 (absent SCOTUS interference, anyway). Yeah, that is what I was thinking of, it allows a majority to be formed if the plurality winner doesn't have a majority. It also allows support to shift over time. > It's hard to imagine the mechanism, though, especially since without > universal (by state) participation, any significant state not playing would > have a strong edge (unless, I suppose, the compact states agreed to > compensate...wheels within wheels). One option would be to assign 80% of the seats in each compact State by PR based on say PR-STV. The remaining seats would be assigned using d'Hondt over the whole compact, by party, but would include seats obtained by the parties in non-compact States. This would mean that if a party loses out due to States existing outside the compact, there is a pool of seats available to rebalance things. Also, it would cancel out the effects of non-compact States, only votes within the compact would actually matter. Ofc, the non-compact States might decide not to say who won until the last minute, so it still has problems. > > The advantage of NPV is that it's simple and doable, even without the > consent of small states currently over-represented in the College. Does that > offset the distinct downside of entrenching FPTP plurality? Maybe so, unless > the alternative is business as usual. Yeah, that is pretty reasonable. If PR isn't possible, my next favourite would be that the NPV used approval voting (or at least allowed states to decide to use approval). ---- Election-Methods mailing list - see http://electorama.com/em for list info