Ohio redistricting proposal poised for failure

[muon2]

You can't avoid considering the political concerns here. There's 4 options:

1. Federally mandate independent redistricting in every state at the same time
2. Implement redistricting commissions done as soon as possible in any state with no order
3. Implement redistricting commissions in a specific order
4. Leave everything as it is

The reality is option 2 is not the most fair (option 1 is) when you consider the status quo and that not every state has the opportunity to set up commissions but because no one is going to do option 1, we're going to have to rely on option 2/3 and the courts until it becomes so lopsided that everyone demands option 1. That's probably the only way for the Electoral College to die as well.

And the parties suck, I'd like to pass proportional representation alongside the federal redistricting law to allow multiple parties but that's even less likely than independent commissions in every state.

This country's electoral system is terrible.

I'm a bit puzzled by the part I've bolded. Most studies I have read would argue that moves to a true proportional system tend to strengthen partisanship over individual representation. Italy is sometimes cited as an example where proportional voting leads to voters ignoring the candidates in favor of the label after the name. Is that the direction you prefer?

On the other hand your goal seems to be multiple parties, which can be accomplished with FPTP systems such as in Canada. The need to form majority coalitions in the legislature tends to drive two of the parties to major positions in any system. The question is how to best provide for alternative views.

[muon2]

As far as I understand it most countries that use proportional representation still have districts via a hybrid system of districts for geographical representation + an at-large distribution of the popular vote for the remaining number of seats so that part of my comment was referring to that lower level of districts and my disdain for the U.S. Democratic/Republican parties.

And whether the system is focused on the individual representative or the party, I don't care too much. I'm more concerned about the issues and FPTP's disproportionate results between the popular vote and the seat distribution gets in the way.

A 10-15% swing (15% more seats than votes) is not uncommon in FPTP, add in some gerrymandering and it can be even worse.

Canada with FPTP:

http://en.wikipedia.org/wiki/Results_of_the_Canadian_federal_election,_2011#Vote_and_seat_summaries

The swing is a predictable result of the statistics making selections based on districts. When a population is divided into districts and one selects one or more representatives from that district it is more probable that a representative from a larger segment of the population will be selected. Voting reflects the majority(s) within a district so those are enhanced.

This impacts the redistricting process as well. You are correct that gerrymandering can force the sing more than would happen naturally, but even without gerrymandering the swing will occur. For example in an evenly divided state like OH it is easy to implement rules for partisan balance to create a set of districts that reflects the political balance of that state as a whole. However in a state like MA it is nearly impossible to implement similar rules to get a split of districts that reflect the overall political leaning of the state simply due to the statistics of the majority.

This is also a problem in many states with minority populations large enough to support the creation of a district under the VRA. In some cases the population is too statistically spread to create a viable district to elect a minority candidate of choice except through extreme gerrymandering. Proportional representation for parties doesn't solve this either, since using a larger area (like the whole state) violates the VRA. The use of large multi-member districts in the south to elect a bunch of white Dems despite a significant black population was a reason the VRA was passed.

[muon2]

Quite frankly I wonder why even have geographic representation at all? Why not just have a single at-large election for every seat? Would legislators really ignore the needs of an area just because they don't individually geographically represent it? I suppose it would be confusing as to which representative you're supposed to write to but is that it?

As a (timely) example, a suburban city here in the area is currently debating a ballot initiative to switch the city council from an at-large model for the councilors to a divided geographic district model and the main complaint from proponents for changing it seems to be that all of the current councilors live in the same area. I could see how that bothers people but honestly if I lived there I would probably vote no on the change.

Your example is exactly the problem with an all at-large system. My experience with local communities is that in at-large systems areas without a representative do get overlooked. They represent a small part of the vote in a large jurisdiction so the representative caters to the base which is often their personal locale.

[muon2]

Well even if it was a big deal on levels above city/county, that's a purely 100% proportional system which as far as I know is not used by any state or country?

The hybrid model most proportional representation (PR) countries use of local districts below an-large distribution to correct for the FPTP swing eliminates the problems of the non-represented minority vote in geographic districts and the overlooked geographic areas in pure PR systems.

Aren't there examples of pure FPTP extremes? Like the Chicago City Council where all 50 members are from the same party?

http://en.wikipedia.org/wiki/Chicago_City_Council

Actually none of the Chicago City Council are elected on a partisan ballot. They are elected as nonpartisan office holders, though all have voted most recently as Dems in partisan primaries which establishes them in a party since there is no party registration in IL. Until 2011 there was an Alderman who voted Pub.

[muon2]

Well there are other examples like the Massachusetts Senate (~90% Democrat) or the Hawaii Senate (95% D with 24D-1R). A 60% popular vote might be realistic in some cases but what good comes from an artificial 90% seat distribution?

A swing from 3/5 or 2/3 to 3/4 or 4/5 won't bother people as much as a swing from 2/5 to 50%+ because the first didn't change the outcome of the election (ignoring supermajority requirements such as amending constitutions) but I don't see why we even need to tolerate this disproportionate swing when there are electoral systems out there that can eliminate it.

FPTP is low on the bar for fair representation. I don't think any country should adopt it for its legislature(s).

Have you thought about cumulative voting as a way to provide some proportionality while maintaining local districts? IL used it quite successfully from 1870 to 1980.

[muon2]

I linked to this proposal earlier in the thread. I'm just curious as to whether or not either of you have read it.

I read it and think it fails on two counts. First is that it is clearly unconstitutional in that the reweighting of votes causes some votes to be worth more than others. This violates OMOV, and strikes me as falling in a category such as counting slaves as 3/5 a person for reapportionment. I get that the author is trying to merge a list system with single districts but in a list system all are elected from the same constituency so there is no reweighting needed - just determine the party proportions then count the votes for candidates within each party.

The second failing is that it purports to maintain a sense of representation within a district, but part of that requires the ability to recognize when a district is more important than a party. The system here puts the party well ahead of the individual in a way that serves partisan interests, but not popular will. I'd rather see a nonpartisan legislature like NE that reduces party roles rather than this plan that accentuates them.

[muon2]

Muon2 and I, I think, have decided that "independent" commissions need a lot of constraints to work well. Otherwise, the commissions tend to be gamed, and those claiming to be "independent" tend to fall considerably short of that. Those without reasonably tight parameters have tended to go off the rails. I don't think either of us have been particularly impressed.

Indeed I would conclude that the commission doesn't have to be particularly "independent" if the constraints are sufficiently tight. In the aforementioned article the author points out that with a single constraint for population equality all sorts of gerrymandering can result.