Proportional Representation- Philosophy

[sa...@equal.vote]

The multi-winner or PR conversation seems to be splitting into two topics. This is a thread to explore the philosophy behind the various PR options. Essentially we're looking bigger picture and talking about goals of PR systems and reforms; Types of representation, electing a representative council, electing a constructive council, and passing representative legislation. Factors that can influence all that include ballot range, max/min numbers of candidates per council, minimum quotas, and more. Also relevant here might be a look at what situations might be better served, if any, by a non PR vote. I'm going to start another thread to talk about algorithms and how best to accomplish various goals. 

As a starter I'd like to introduce the idea that there are three fundamental types of groups which could be proportionally represented: Geographical groups, ideological Groups, and identity based groups. "Prefect" proportionality would guarantee all three, and is probably impossible unless there was a guarantee of having a ton of super high quality candidates with perfect overlap in all three categories. I know that this is broader than the PR definition and conversation that's standard in voting theory, but it comes up a lot and I think it's got value. 

I'd also like to encourage looking through the lens of the 5 pillars of a just PR voting system: Equity, Accuracy, Honesty, Simplicity, Expressiveness. Here's a totally subjective article for single winner and I'd love to see something similar out there for PR options some day: https://www.starvoting.us/report_card

[Toby Pereira]

I wouldn't say you need proportionality based on specifically geographical, ideological and identity-based groups. You need proportionality based on what issues people choose to vote for. They may or may not vote on some or all of those particular things. And a voting system as a "mathematical" thing won't discriminate between these. Obviously if you have geographical regions within a country where there are multiple seats to be won in a proportional election you'll get geographical proportionality better than if you have a whole country as a single region, but in terms of ideology and identity, they're not thing to build into the system as such.

[Jack Santucci]

Dear Sara and others,

It's great to see the interest in PR voting here.

I have spent the past five years studying PR-STV's rise and fall in the United States. I also studied its effect on diversity in candidate recruitment. There is a lot more work to do, and the data are hard to get.

More interesting for the present conversation, perhaps, is why PR-STV became the preferred method in the first place. Short answer: it is very difficult to promote a voting method that acknowledges third-party existence. I go through those issues in this working paper: http://jacksantucci.com/docs/papers/partyattack.pdf.

Thoughts and feedback are welcome.

Jack

--
You received this message because you are subscribed to the Google Groups "The Center for Election Science" group.
To unsubscribe from this group and stop receiving emails from it, send an email to electionscience+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Ciaran Dougherty]

I'm inclined to agree with Toby here.  With a good voting method, you could have candidates running on platforms of ideology, or of identity-based experience, or religion, or even something as silly as favorite color.  It wouldn't matter, because at the end of the day, when the votes were tallied, the candidates whose messages best resonated with the electorate would win, and those whose messages didn't could try again next time.

The only question that we really need to worry about, I believe, is how geographically based the representation should be.   California, for example, has 53 representatives.  That being treated as a single, multi-seat election would lower the degree of misrepresentation from as much as nearly 50% with single-seat districts, to as low as 1.8% representation error.
....but if it were treated as a single, multi-seat, At-Large district, then the candidates that got elected would likely be almost entirely from the top few major metro areas, mostly from the greater SF Bay and Los Angeles areas, with a few from the San Diego and Sacramento areas.  Few to none from "Jefferson," from east of the Sierras, etc, and significant portions of the population living several hours drive from the district office of the nearest congress critter that "represents" them.   On the other side of the coin, if geography is more emphasized, you end up with greater potential for gerrymandering, and increasing non-representation, up to ~50% non-representation, with single member districts.

The balance I lean towards is to try for somewhere between 3 and 7 seats per district, ideally centered around commuting patterns (such as captured by Metro Statistical Areas (as defined by the OMB), so you avoid scenarios such as we have a lot of in the greater Seattle area these days, where significant percentages of the population live in one congressional district, but work in another.

But once you decide the size of your districts (as a function of area and population), the non-geographic proportionality will be decided per election by the electorate, as Toby said.  Just as the electorate currently determines the outcome based on whether they feel Democrats' ideals or Republicans' ideals  (generally speaking) are more important to them in any particular election, with a decent mutli-seat system, they will be able to determine what best represents them for that election, based on all of the elements the various candidates have to offer.  

--

[Sara Wolf]

Good to meet you Toby and Jack! Hi Ciaran, 
I'll check out that article when I get a chance and appreciate your perspectives. So, commuting time between voters and their reps is a big deal for actual accountability It's important that people have a voice and can stay in the loop! I'd say an hour 1 way is about as big as you'd want. On the other extreme, if districts are so small that people are living in one and working in another that's not ideal. Balancing these two factors seems good. It also seems good to balance things like geographical and ideological representation as well as size of government. It also seems good to balance size of thresholds. Too little allows anyone to rise to power, while too big leaves people excluded. 

So how about this: You have a region that is about 2 hours from end to end. You could divide that into 4 districts that are about an hour across each. (2x2 grid.) Each one elects a rep by STAR Voting and does business at a central location where people can come in and testify or present/meet as needed. Then you also elect 4 more reps at large using PR and a STAR Ballot. The local reps focus on local issues, the PR reps focus on broader issues. The PR reps rotate their weekly public sessions between the 4 district seats so that once a month each community can come and meet, present or testify as needed without driving too far.

These 8 reps are also delegates to a larger governing bodies and each rep could equal a seat on Parliament or Congress or whatever. 

Remember that with STAR Voting every vote makes a difference so the single-winners represent the center of public opinion, not just the center of the majority group's opinion. 

--

-Sara Wolk

Chief Petitioner for "STAR Voting for Multnomah County"

Portland Equal Vote

971-222-9364

sa...@equal.vote

​www​

.

​starv​

o

​ting​

​

.

​

​us​

www.equal.vote

“The fact is that FPTP, the voting method we use in most of the English-speaking world, is absolutely horrible, and there is reason to believe that reforming it would substantially (though not of course completely) alleviate much political dysfunction and suffering.”

-Jameson Quinn, The Center For Election Science

[Jameson Quinn]

I'd like to remind people that "degree of proportionality" and "magnitude of districts" are not necessarily in a tradeoff as Ciaran implies. There's also biproportional methods (including PLACE, DMP, LPR, FMV, etc.; as used in Swiss municipalities) which can ensure balanced geographic representation for districts as small as 1 seat each while also balancing proportionality across regions as large as all of California. This also allows for thresholds that are relatively low in numerical terms while still being relatively high in proportional terms, which I think is the best of both worlds: a fair shake for independent candidates, but a good incentive against excessive party fragmentation (a la Israel).

[Sara Wolf]

What do you mean by "This also allows for thresholds that are relatively low in numerical terms while still being relatively high in proportional terms." Any advice on where to get some background here? I'm only somewhat familiar with some of the European systems.

[Jameson Quinn]

2018-05-24 19:25 GMT-04:00 Sara Wolf <sa...@equal.vote>:

What do you mean by "This also allows for thresholds that are relatively low in numerical terms while still being relatively high in proportional terms." Any advice on where to get some background here? I'm only somewhat familiar with some of the European systems.

Biproportional methods aren't common in actual use; the only one I know of is the Swiss municipal system, and the biproportionality there is pretty weak. The methods I mentioned were: PLACE, DMP, LPR, FMV.

The distinction between numerical and proportional thresholds is based on the 25% threshold in PLACE. Say there's a city with 1M voters and a 10-member city council. Each district will have 100K people, so the 25% threshold is just 25K people, or 2.5% of the electorate. That's pretty lower, and it would be even lower if there were more seats in play; it's reasonably feasible for an independent candidate running a money-poor grassroots campaign. But the threshold is also 25% in proportional terms. A party that has, say, 15% of the electorate will almost certainly make it over 25% in its strongest districts, so will get some seats; but one with just 8% probably won't, so their votes will be transferred to other larger parties of their choice (not wasted). That means that those voters will still be represented, but party splintering will be discouraged; it would be almost impossible to have the size-weighted number of parties reach as high as 6, as it is in Israel's dysfunctional party-list-proportional system.

[Ciaran Dougherty]

Too little allows anyone to rise to power

Ah, minor correction, there.  Too little allows anyone that the electorate supports to rise to power.  And yes, that's what democracy means.  

On the other side of the coin, if the thresholds for participation are too high, then politics becomes (continues to be) a rich person's game.

excessive party fragmentation

Jameson, what do you mean by "excessive fragmentation"?  Is the degree of fragmentation truly relevant compared to accuracy of representation?  With a decent voting method, wouldn't a fragmented legislature simply be indicative of the electorate having diverse, possibly disparate interests?

While that may inhibit the passing of legislation, do we really want to pass legislation that does not reflect the will of the people?   Wouldn't the effect of selecting for less fragmentation be the creation of false majorities that could then pass legislation that a true, significant majority of the population want to see fail?  Or false majorities that can vote down legislation that a true, significant majority of the electorate want to see passed?