Should we stop encouraging uninformed people to vote?

[metatron]

We should, switch from first-past-the-post plurality voting to Range Voting, where every voter is encouraged to rate every candidate independently rather than merely choose to grant his or her full support to the lesser evil.

The candidates seeking ballot access should be required to summit a summary of their platform and qualifications for the office (along with a sworn affidavit, making any falsehoods criminal perjury) which would be published before the election and made available for review in the ballot booth. Voters could then be required to affirm that they have reviewed this information before registering their vote for any candidate.

What did i write about this the last time i got triggered?
Ah, right...

"1. Alternative Vote / Preferential Voting / Instant Run-Off are just goddamn awful, terrible and suck, suck, suck, suck.
2. As far as CFC-OT is concerned there is a substantial trend that anybody claiming the contrary is a (non-Kiwi) Anglospherian currently or previously exposed to simple FPTP election schemes (i.e. the only thing worse).
That's systematic."
​

Never mind that Range Voting is even worse than STV and IR, arguably worse than fptp even, because it's so absurdly gamey that a plurality if not a majority if not a vast majority of the population does not sufficiently understand the math and the game theory involved in maximising the impact of their vote.
So it's virtually undemocratic and an elitist ploy at that.
It's basically what an undercover cartoon villain would come up with to sabotage an anti-fptp-movement from the inside.


And why do you people come up with this nonsense, again and again, even though it fails to be approved by the public time and time again, while actual proportional systems are constantly expanding accross four continents?
Why do you ignore the 60 odd countries that use either full list PR or MMP (i.e. most of the worlds democracies) and instead stick to the glorious tradition set by... erm... Ireland and erm... Halfstralia?
Why do you refuse to recognise that RV, STV and IR have 98% of The Bad of FPTP and all they do is making the fundamental undemocratishness of FPTP slightly more palatable in a priviledged largely docile three party environment (such as Australia, Canadia or cosmopolitan southern England), while showing their true ugly face in, say, Belfast?

I have increasingly the impression that you people want to get rid of fptp's undemocratishness and keep it at the same time, by installing overengineered addons that don't work.
In that way AV, RV, IR and STV are the VW clean diesel engines among electoral systems.

 

[Arwon]

Wait did you just call single transferable vote, a form of proportional representation, basically the same as first past the post and other single member constituency systems? Cos that's an odd link to make.

Also hang on the NI assembly would have pretty similar composition under a party list form of proportional representation rather than the STV method. And they use the D'Hondt method to allocate ministers in that power sharing arrangement they have, anyway.

 

[metatron]

Also hang on the NI assembly would have pretty similar composition under a party list form of proportional representation rather than the STV method. And they use the D'Hondt method to allocate ministers in that power sharing arrangement they have, anyway.

I was unclear about this bit. Sorry.
I was implying the usage of, say ranged voting (as the most extreme example) in UK national elections and as a result in Northern Ireland in UK national elections.
I'm not familiar with the politics of the NI Assembly.

Wait did you just call single transferable vote, a form of proportional representation, basically the same as first past the post and other single member constituency systems? Cos that's an odd link to make.

Yes. I did.
It's not proportional. You people merely decree that it is, largely based on the untrue assumption that, say, a 3rd preference is equal to a 1st preference.

This is all fine and well and feels "proportional" if you have an Australian situation.
Now imagine the situation of a party very much unlike Australian Greens that is in severe disagreement with both major parties, to the point of mutual contempt. That party would fail to win seats, or if strong enough would have their number of seats routinely rounded down.
What you want in that situation is that the voters of this party fil in 2nd and 3rd choices - at random if need be - so that you can pretend they are represented accurately and the system appears legitimised.
As soon as they stop doing that and only fill in a 1st preference the jig is up:
You are effectively disenfranchising them in a non-proportional system that in lived reality has a lot more in common with fptp than than with actual pure party list PR or MMP.

 

[metalhead]

I'm staggered that you're so okay with this. In other contexts you might decry such decisions as "racist", or "sexist", why is it okay for that to be the primary driving motor of democracy? An element of that will obviously always be there, but saying that that's how democracy must work, seems a bit much to me.

People have a right to be terrible. You can use shame and try to establish social mores and use the tools of government power to try to eliminate the effects of people being terrible, or even convince people not to be terrible, but that doesn't take away their absolute right to be as racist and sexist as they want. They are still human beings, they still have fundamental rights to think and say and do whatever they please. A lot of times terrible people misconstrue this right as meaning they have a right to be terrible without any penalty or repercussion, but that's their problem.

 

[Timsup2nothin]

Uninformed voters are not the problem. The uninformed vote will split fairly evenly. That's in the nature of being uninformed.

The problem is the misinformed voters, and the people who intentionally misinform them.

 

[warpus]

I include the misinformed under the uninformed label.

 

[Timsup2nothin]

I include the misinformed under the uninformed label.

I think that's an inaccuracy, and could be problematic.

As I said, the genuinely uninformed, if they vote at all, will balance themselves out. So no matter their numbers or how motivated they are to turn out they really can't be a problem. But the misinformed present a very serious danger if they exist in significant numbers and/or are strongly motivated to turn out.

 

[Hrothbern]

When I was a kid, the uniformed, misinformed AND informed voted very simple:
Most christians on the strict, or less strict protestant party, or the catholic party (totalling 50% of the votes in the early 60ies)
Some christians and most non-christians that had low income, on the communist or social democratic party.
Non-christians with higher income on the predecessor of the neo-liberal right.
The youth in increasing numbers on progressive, libertarian splinters that kept growing.

And yet things went pretty much allright.

 

[warpus]

I think that's an inaccuracy, and could be problematic.

As I said, the genuinely uninformed, if they vote at all, will balance themselves out. So no matter their numbers or how motivated they are to turn out they really can't be a problem. But the misinformed present a very serious danger if they exist in significant numbers and/or are strongly motivated to turn out.

You're not wrong

 

[Arwon]

I was unclear about this bit. Sorry.
I was implying the usage of, say ranged voting (as the most extreme example) in UK national elections and as a result in Northern Ireland in UK national elections.
I'm not familiar with the politics of the NI Assembly.

Yes. I did.
It's not proportional. You people merely decree that it is, largely based on the untrue assumption that, say, a 3rd preference is equal to a 1st preference.

This is all fine and well and feels "proportional" if you have an Australian situation.
Now imagine the situation of a party very much unlike Australian Greens that is in severe disagreement with both major parties, to the point of mutual contempt. That party would fail to win seats, or if strong enough would have their number of seats routinely rounded down.
What you want in that situation is that the voters of this party fil in 2nd and 3rd choices - at random if need be - so that you can pretend they are represented accurately and the system appears legitimised.
As soon as they stop doing that and only fill in a 1st preference the jig is up:
You are effectively disenfranchising them in a non-proportional system that in lived reality has a lot more in common with fptp than than with actual pure party list PR or MMP.

Thaat situation doesn't really make a difference whether it's STV or a party list. If a party can't get to the threshold to gain a seat then they don't get a seat in any case. And whether there be preferences like in STV or no preferences like in list or MMP systems, if there's no secondary parties palatable to those voters, they're going to be unrepresented, period.

The key determinant of representation threshold in all cases is the number of members to be elected. In Spain they have each province as a party list constituency, which elect as few as 3 members, or as many as like 40. In a 3-member province like Soria or Teruel if a party doesn't get to 25% of the vote they don't get represented in those areas. In Barcelona things become much more granular because the quota drops to like 3%. Likewise the difference in the granularity and approximation of proportionality in the Dutch or Israeli nationwide electorate is quite marked compared to an MMP system with say a 5% threshold for party seats.

Likewise STV looks substantially different in my local Territory senate race where we elect 2 senators (quota 33%) compared to our Legislative Assembly with 5-member districts (quota about 16%), the 6-member ordinary Senate races (quota 14%), a double dissolution 12-senator race (quota 7%) or the NSW Legislative Council which elects 21 (quota theoretically 4.2% but the last seat gets weird).

And in all cases they're quite different to seats with one member and a quota of 50%.

But in all honestly usually even the fash voters have a preference between Labor and Liberal, which is why we find fairly moderate exhaust rates even by One Nation voters. I'm not clear on why removing that extra piece of information would improve representativity within the constraints of having a 5 or 6 or 12 person electorate.

You also get under STV's individual focus the ability to discern within parties. In both the ACT and Tasmania (where candidate order within party lists is randomised on each ballot) you often see an energetic local Labor campaign elect a new Labor person at the expense of a disliked existing member of that same party. Tasmania once saw a deputy premier ousted for a partymate in this way and in Canberra in 2012 the grassroots student left effectively rolled a sclerotic old Labor right faction member at the ballot box, by winning over a lot of personal votes through an effective campaign in the low income half of an electorate.

There's little funnier than watching Labor and the Liberals campaign against themselves and both parties factional power brokers regularly fail to get their people up or keep them in office - I really like that individualised aspect of the STV system over list systems. (Note our Senate doesn't permit this to happen because it's a pseudo list system)

 

[Cutlass]

Should we stop encouraging uninformed people to vote?

That would extinguish conservative politics everywhere.

 

[metatron]

Thaat situation doesn't really make a difference whether it's STV or a party list. If a party can't get to the threshold to gain a seat then they don't get a seat in any case. And whether there be preferences like in STV or no preferences like in list or MMP systems, if there's no parties choose to those voters they're going to be unrepresented, period.

But in all honestly usually even the fash voters have a preference between Labor and Liberal, which is why we find fairly moderate exhaust rates even by One Nation voters.

The key determinant of representation threshold in all cases is the number of members to be elected. In Spain they have each province as a party list constituency, which elect as few as 3 members, or as many as like 40. In a 3-member province like Soria or Teruel if a party doesn't get to 25% of the vote they don't get represented in those areas. In Barcelona things become much more granular because the quota drops to like 3%. Likewise the difference in the granularity And approximation of proportionality in the Dutch nationwide electorate is quite marked compared to an MMP system with say a 5% threshold for party seats.

Likewise STV looks substantially different in my local Senate race where we elect 2 senators (quota 33%) compared to our Legislative Assembly with 5 member districts (quota about 16%), the 6- member ordinary Senate races (quota 14%), a double dissolution senate race (quota 7%) or the NSW Legislative Council which elects 21 (quota theoretically 4.2% but the last seat gets weird).

And in all cases they're quite different to seats with one member and a quota of 50%.

Erm, nope.

1. The spannish list constituencies are too small. That's gross malpractise, not an inevitable defect of the system.
2. Thresholds are besides the point. You're free to not have them.
3. You still don't appreciate the fundamental point: This is not about granularity. In the view from my side of the fence 2nd and 3rd preferences are worthless.
To be appreciated by me as proportional STV would have to somewhat proportionally (no need for perfection) represent 1st preference votes.
It doesn't.
4. Please try to make a coherent case as to what the upside of STV compared PR or MMP is supposed to be. All that i see so far - same as in the countless run-ins we had over this - is a somewhat undemocratic system with complicated ballots, massive moral hazard, an unintuitive outcome and vague-at-best local representation that can be increased or decreased in negative correlation to the faux proportionality or rather what's left of it.
This is all greatly mitigated in the specific context of the Australian Senate or are just plain acceptable in a senate in general in the first place.
In contrast it remains largely unclear what the upsides of this system are actually supposed to be in a national election of a lower house (outside such things as said spannish malpractice).

Oh, right, the upside is: It's not fptp.

5. Regarding your later edit. Yeah, sure that's peachy and all. Yet again i fail to see how that is elementary or exclusive to STV.

 

[Arwon]

the view from my side of the fence 2nd and 3rd preferences are worthless.

As a voter under STV I simply don't feel this way. A lot of people would vote quite differently without the ability to express their genuine preferences free of the tactical voting considerations inherent to only selecting one party.

Just because I vote for the Greens doesn't mean I don't have a view on the relative merits of the Liberals, Labor, the Nationals, One Nation, and the various other minor parties who sometimes get a run. The preferences reveal electorate views on multiple pairings of parties. And the counting for a final seat even under closed list proportional systems can always come down to any pairing of parties.

Many party list systems would make me forfeit that say in those kinds of races, because I've expressed equal views towards every party not voted for. So I need to make a decision about who's gonna win seats and where my vote might be most tactically useful rather than just who I like (will the Greens win a seat or am I just taking a vote away from a party who can actually get seats?). MMP might be better for that, depending on threshold shenanigans, but it might not.

Preferences fill in extra information which is left necessarily unknown if you're only expressing a view on one candidate or party. In France they run a whole extra round of voting to get around that problem. We just do it all at once.

 

[metatron]

Preferences fill in extra information which is left necessarily unknown if you're only expressing a view on one candidate or party. In France they rub a whole extra round of voting to get around that problem. We just do it all at once.

This is a disingenuous comparison. I am sure you can see why.

As a voter under STV I simply don't feel this way. A lot of people would vote quite differently without the ability to express their genuine preferences free of the tactical voting considerations inherent to only selecting one party.

Ah, feeling...
The thing is, in a PR or MMP that is not somehow messed up by poor craftsmanship (and i appreciate your point on thresholds on that note) all of this is entirely superfluous, virtually by definition.
The additional information is not needed because first preferences are already accurately reflected, usually in a fundamentally more proportional fashion than STV can provide.

I remind you again: To think and speak about this with intellectual rigor and honesty, you have to ween yourself off the familiar perspective of Australian Senate elections and superimpose the system on large scale (think 300-900 members) national lower house elections.

Put yourself in my perspective: I vote for a party. I disagree with the other parties. We usually get 9% of the vote. We get roughly 9% of the seats (slightly more because of the threshold and all that). In that i am represented.
That is a fact. Not a feeling.
With this funny conconction of yours we would still get 9% of the vote (at first, less later, as a function of defeatism) and would get maybe 6% of the seats, due to the faulty system.
That's underrepresentation.
Fact. Not feeling.
That i have the option to have subsequent preferences get molded into stuff i actually don't want, is besides the point.

How right wing nuts feel about their shots in an Australian Senate elections is immaterial to the above reasoning, as i am sure you can understand.
Because - i know i am repeating myself - it's Australian. And also and more importantly a bloody Senate election.
We are talking huge constituencies and massive disproportionality (vis a vis population) as a baseline assumption in that case.
Because... wait for it...
...it's a Senate!

 

[Arwon]

I'm not aware of a system aside from a single nationwide electorate a la Israel or the Netherlands, where the percentage of seats gained equates to exactly the percentage of the vote. In any form of federal system, anything with multiple electorates, and any system with any form of threshold more meaningful than "we have 80 seats in this chamber so you need 1/80th to get a seat", that's not going to translate across exactly. Even the MMP systems and additional member systems don't quite achieve exact proportionality.

Note also that STV developed as a method of achieving a good degree of proportional representation while retaining the ability to vote for individuals including across party lines, and also a connection to local representation. It has a dual mandate in that regard and as such isn't really designed to be purely proportional, so of course it doesn't achieve a mathematical test it's not designed to meet. STV pretty inherently assumes smaller districts rather than larger ones, because in large districts like the 21-member at-large NSW Leg Council vote, it effectively turns into a preferenceless party list prop rep system anyway (only seats 20 and 21 ever go to preferences and they usually don't).

STV is a method for being as proportional as possible within systems that require multiple smaller districts, and every member to have a district. STV really needs 5 or members per seat, minimum, to work in a truly proportional manner, but then that's also true of party list systems as well as we observed with Spain. The Irish system throws up some anomalies due to only having 3- and 4-member districts in many cases. My local Territory assembly elections are fine but not optimal at 5 members (and really should be 7 per electorate as initially recommended). The sweet spot where preferences still matter but outcomes are quite proportional is probably about 7-12

Localism and voting for individuals is a strong cultural preference in a lot of places, but is also sometimes a constitutional requirement (as I believe it is in the Australian senate that the votes be for individuals). In large and diverse countries, particularly those with vast rural hinterlands, with significant national minorities, or with a politically contentious federal structure, pure proportionality is not necessarily the preferred outcome of most voters, so the retention of an element of localism is preferred or required. You're not, after all, ever going to see the United States vote as a single electorate for a unicameral body like Israel does.

I would also suggest "I like one party and have exactly equal views about all other parties" isn't likely to be a common position anywhere, you may be unusual in that... and I don't think it's the only or most appropriate test of representation. "Does my vote for a left party undermine the less-left party's chances against a right party" is also a pretty valid test. In any system with thresholds or quotas above a few percent, that gets tricky to answer without the ability to actively discern views on multiple parties.

 

[Zelig]

The American elections are a great example of the free market not working to create a socially useful, but efficient, outcome. A tremendous amount of money is spent on advertising. Huge, golly huge, amounts. And then there's all the 'free' advertising that comes with news reporting and commentary.

And people have no true clue as to what's going on with a candidate's platform. At best, after all that advertising is spent, people think they should vote for their team's candidate. It's got electrolytes.

I think advertising is a more general, and broad failure of the free market.

 

[El_Machinae]

Yeah, you mentioned last time. I'd agreed after I'd mulled it for awhile.

It's one of those 'when did CFC change your mind' things.

 

[metatron]

I'm not aware of a system aside from a single nationwide electorate a la Israel or the Netherlands, where the percentage of seats gained equates to exactly the percentage of the vote. In any form of federal system, anything with multiple electorates, and any system with any form of threshold more meaningful than "we have 80 seats in this chamber so you need 1/80th to get a seat", that's not going to translate across exactly. Even the MMP systems and additional member systems don't quite achieve exact proportionality. STV really needs 5 or members per seat, minimum, to work in a proportional manner, but then that's also true of party list systems as well. The Irish system throws up anomalies due to only having 3- and 4-member districts in many cases, my local Territory assembly elections are borderline at 5 members (and really should be 7 per electorate as initially recommended).

As i said: We're not talking granularity here. The imperfections in the other systems are just that: imperfections.
The disproportionality in STV is fundamental. It's plainly put not a proportional system. It can sort of get away with masquerading as one, but it's not.

Note also that STV developed as a method of achieving a good degree of proportional representation while retaining the ability to vote for individuals including across party lines, and also a connection to local representation. It has a dual mandate in that regard and as such isn't really designed to be purely proportional, so of course it doesn't achieve a mathematical test it's not designed to meet. STV pretty inherently assumes smaller districts rather than larger one, because in large districts like the 21-member NSW Leg Council, it effectively turns into a preference-less prop rep system anyway. It's a method for being as proportional as possible in systems that require multiple smaller districts and every member to have a district.

Localism and voting for individuals is a strong cultural preference in a lot of places, but is also sometimes a constitutional requirement (as I believe it is in the Australian senate). In large and diverse countries, particularly those with vast rural hinterlands, with significant national minorities, or with a politically contentious federal structure, pure proportionality is not necessarily the preferred outcome of most voters, so the retention of an element of localism is preferred or required. You're not, after all, ever going to see the United States vote as a single electorate for a unicameral body like Israel does.

You still have no explanation here as to how most of this is achieved better in STV than say MMP or DMP.
In one case you have demonstrated that: If disproportionality and "unfair" representation is actually desired, say to lend weight to a minority the way you mentioned, then STV is a very valid option.
As i indicated, this is what upper houses are commonly used for.
Lower houses, not so much. Pears. Bananas.

I would also suggest "I like one party and have exactly equal views about all other parties" isn't likely to be a common position anywhere, you may be unusual in that... and I don't think it's the only or most appropriate test of representation. "Does my vote for a left party undermine the less-left party's chances against a right party" is also a pretty valid test. In any system with thresholds or quotas above a few percent, that gets tricky to answer without the ability to actively discern views on multiple parties.

Oh, of course the other parties are not perfectly equal to me. If you shake me real good i can come up with a preference.
That doesn't justify you stealing a third (or whatever ammount) of my vote.

 

[Arwon]

I like MMP and DMP, they achieve similar goals in a different way and are particularly elegant for unicameral systems. I'm also okay with party list setups, but as in the Australian senate, the inability in many of them to punish high-listed individuals from your party of choice would be frustrating, plus a lot of them don't let you split your vote across individual candidates from multiple parties.

I'm just saying STV is also a method of proportional representation, and that it works in the situations (smaller electorate sizes) it's designed for, and that preferences are good for extracting more information from voters in any situation where any significant fraction (ie, more than a few percent) won't be electing a member with their primary vote.

That doesn't justify you stealing a third (or whatever ammount) of my vote.

That's not quite how the counting of preferences works in STV. If your party doesn't get elected at all, the whole vote transfers or exhausts as determined by your preferencing. If your party does get elected, then the surplus transfers and exhausts in proportion to preferences within the whole set of votes. That still works out to, conceptually, your whole vote either electing your own candidate, or whoever your preferences helped get elected, or outright exhausting if that's what you expressed on the ballot.

 

[Timsup2nothin]

Should we stop encouraging uninformed people to vote?

That would extinguish conservative politics everywhere.

Again, this is the common and dangerous error; failing to distinguish between the uninformed and the misinformed. To extinguish conservative politics requires reduction of the misinformed vote, not the uninformed.