Jump to content
Sign in to follow this  
Nacey

The Brexit Countdown Echochamber of Exasperation

Recommended Posts

The election thread (and associated TS session) was fun and contained a lot of good debate! But now that the election is over we all have to go back to our miserable lives bemoaning the state of everything while angrily shaking our fists at things. Such is the way of things.

But wait! We could have a politics thread and never stop arguing about important issues and linking to stories about a hilariously dumb thing that an MP has done. With the various party leadership contests, the first all-Tory government since 1992, the devolution agenda for Scotland (and elsewhere), the Trident renewal, and the not-so-small matter of an EU referendum in a couple of years' time, I think we can probably find some things to discuss!

Politics can be divisive and we have quite the range of opinions around these parts but I think we've demonstrated the ability to disagree without being enormous jerkfaces to each other. Let's continue along the same lines :) .

Share this post


Link to post
Share on other sites
One of the things that we were chatting about over the weekend was the presentation of the SNP down south compared to up north. 

 

Up here independence barely featured in the SNP campaign, while down south it seemed to be the only thing talked about regarding the SNP. Since the election I've seen various people down south talk about the SNP victory up north as "independence right round the corner", while up here that's not the case.

 

It seems like a disconnect over how the party is presented and something that should probably be solved if any party wants to reclaim seats here.

Share this post


Link to post
Share on other sites

Election fallout and political reform seem like good places to start.

In the run-up some of you will remember that I wrote a big post about why FPTP is good and we should keep it. I would like to state now that the post doesn't entirely represent my opinion, though it isn't too far away. FPTP is dumb and doesn't make a lot of logical sense, but it does produce strong governments. I am really undecided on this issue and have a tendency to argue against whoever I happen to be discussing it with because that's far more interesting than everybody agreeing. Hence my big post before.

With the way that the results came out it was inevitable that people would be talking about vote shares and votes-per-seat and how unfair everything is. Again it's important to understand here that PR may produce a perfectly representative parliament but it does not produce a perfectly representative government and doesn't even necessarily produce a more representative government. Sound backwards? We don't even have to invent an example!

Now party politics are complicated but we can make a fairly valid simplification in saying that the parties can be arranged along an axis (which you might label "left-wing" <-> "right-wing" but it doesn't actually matter) and that most people, if given a second, third, etc. choice would vote for a party that was close to the one that they voted for. This axis typically looks like this:

Grn, SNP, Lab, LD, Con, UKIP

(For simplicity I've ignored the Green party and other regionals)

For this election the vote share, seat distribution, and seat distributions under a PR system are as follows:

PARTY  VOTE  SEAT    PR
Grn     3.8     1    24
SNP     4.7    56    31
Lab    30.4   232   199
LD      7.9     8    51
Con    36.9   331   242
UKIP   12.6     1    82
So the Tories have a majority. If we used PR, what government would we have? Almost certainly a Con/UKIP (324) coalition, with either the DUP or the LDs depending on how the NI seats would work out. 324 is technically a majority but is so narrow as to be difficult to operate in practise. What we can say for sure is that we wouldn't have got a Lab/LD, Lab/SNP, Lab/LD/SNP, or Con/LD coalition because those combinations simply have nowhere close to the numbers required to form a majority.

Is this result better? Is it more representative? In some ways, certainly, since it includes MPs from parties that a majority of the populous voted for. But in 2010 we saw lots of LD voters declaring that they "didn't vote for a coalition" (lol). In some ways a coalition represents nobody. And I'd think a lot of Lab/SNP/Green voters would rather see a straight Tory government than a Tory/UKIP one (but tell me if I'm wrong about that, it's an interesting one). So if you started building preference into this, a coalition involving UKIP could be said to be less representative than a straight Tory government.

This is all a bit vague because "representative" is not a simple word to define. And people would undoubtedly vote differently if they knew that the result was based on PR. Typically you'd expect to see more votes for smaller parties that current "can't win" in a given area. Mainly that would mean more votes for UKIP and the Greens and more votes for the Tories in Scotland. Possible a bigger LD vote too, given how few they got this time around. It's really hard to say how this would all interact and work out. But it feels like there is a good chance we'd be operating under a Tory/UKIP coalition right now.

Food for thought.

Share this post


Link to post
Share on other sites

Do you lose local politics in a PR vote? You can no longer vote for your local mp, on local issues as you are only voting for the national parliamentary seats. Could they instead look at the votes per seat and any party with more than one SD from the average gets two votes per seat.

Share this post


Link to post
Share on other sites
I'm not entirely convinced we can call the current government a strong one as yet. Will depend how much Cameron can keep his backbenchers in line.
 
But in 2010 we saw lots of LD voters declaring that they "didn't vote for a coalition"
If you have a system that frequently produces coalition governments though, people will vote with that in mind. In 2010 a coalition wasn't in anyones minds when they voted since its not something the current system generally produces.
 
I personally would not want a Tory/UKIP coalition. However electoral reform shouldn't be argued against on the basis that it might give less savoury parties more of a say. In the end UKIP got a lot of votes (3.8M, 13%?) and ended up with very little for it.
 
 
As an aside, who do people think will lead the "stay in Europe campaign"? I doubt it will be Cameron.

Share this post


Link to post
Share on other sites

Do you lose local politics in a PR vote? You can no longer vote for your local mp, on local issues as you are only voting for the national parliamentary seats. Could they instead look at the votes per seat and any party with more than one SD from the average gets two votes per seat.

So in some PR systems, like the one used in Germany (Mixed-member proportional representation, or MMPR), you cast two votes: One for local constituency candidate (elected for that constituency by FPTP) and one for national party. If a party wins a higher share of the party vote (say 40%) than the total number of constituency seats (say 32%), they have a separate party list of non-local candidate MPs who are drafted to make up that shortfall (8% in our example) in parliamentary seats - which become the "list seats". In some systems, you can be on both a regional candidate and on the party list, but would get crossed off the list if you win your regional seat.

If a party wins more local seats than its share of the overall government should allow, an overhang occurs and generally other parties are allowed extra "list seats" to restore the overall balance. This means that the overall size of government is not fixed.

Other PR systems work in different ways which allow a range of control over a local representative from full to none.

Closed list PR has voters literally just voting for the name of the party and then seats in government are assigned based on the total relative popularity of the parties in order of the list as pre-decided by the party pre-election.

Open/local list PR systems tend to publish lists of candidates for constituencies in which a number of seats exist, usually 3-10. In the UK you might have a single city or less-populated county as a constituency which would be sub-divided into electoral districts the size of the current constituencies. Voters vote for a single party candidate in their district only, MPs are elected to the constituency rather than the district, based on the overall party performance in the country - if we look at Merseyside as a hypothetical 15-district constituency (in isolation), instead of pulling 14 Labour MPs and one Lib Dem in this election, the overall vote share would be most likely have given 10 seats to Labour; 1 Lib Dem and 4 Tory. The 10 highest polling Labour candidates would receive seats, as would the one Lib Dem seat that they actually won and the four highest Tories (who all came second, though in come cases by only a couple of hundred votes).

Edited by Brynneth

Share this post


Link to post
Share on other sites

This came up a couple of times in the previous thread so maybe this would be a good time to explain how PR works in practice.

Firstly, most PR systems in practice are not perfectly representative of the vote. There are a few different systems and it is useful to understand them and the differences between them. Unfortunately this requires spending an afternoon on Wikipedia, but I'll mention a couple of popular examples here.

Some use a simple regional or national system, others are able to combine traditional constituency representatives with a regional party-based top-up list producing a so-called "hybrid" system. Where such a hybrid system is used, generally the PR system itself doesn't handle this, rather what you do is you bolt two different systems together and modify them slightly to maintain the required level of proportionality.

For example, D'Hondt is a system which allows you to use a simple single-choice vote to elect multiple candidates for a region (or whatever). You can read the Wikipedia page for a fully worked example but the idea is that the party with the most votes wins a seat and then you halve that party's vote. You do this repeatedly until all of the seats are allocated. Each time a seat is awarded, that party's vote is reduced to x/(n+1) where x is their original vote count and n is the number of seats they currently have. So if a party wins their third seat their vote will now b a quarter of what it was originally for the purpose of allocating the next seat (and the same for each other party). This can be modified to support constituency-level seats by having a separate vote for these (using FPTP or AV or something else), allocating the constituency seats, and then modifying the n value described above for each party so that the regional voting already accounts for the constituency seats that they have won. This is a slightly simplified version of the Additional Member System which is used for the Scottish Parliament and London Assembly, amongst others.

The most commonly proposed PR system for UK national elections is Single Transferable Vote. Again, this system only provides for a way to elect regional seats. I recommend reading the Wikipedia page if you want to understand the system but the general idea is that it accounts for second, third, etc. choice votes by allowing votes that were wasted either because a candidate cannot reasonably be elected or received more votes than they needed to be elected are redistributed according to second choice votes in a proportional manner. There is some debate about how this reallocation should work and different elections use different systems. In terms of being representative it is very good system but the maths are somewhat complicated for the general populace to understand which typically makes it a difficult sell in a "should we use this system" referendum. Under STV it isn't possible to add constituency-level seats without severely hindering the proportional nature of the vote. You can make it more local by increasing the number of regions but this inherently makes it less perfectly representative.

Share this post


Link to post
Share on other sites

So I'm trying to find a list of the election results per constituency listing parties by votes / %share, so I can play around with a potential PR map where I assign seats to parties based on relative vote share, but I can't find anywhere that has an actual list. The BBC site is the only place which has a full list of the constituencies from which I can find the vote share for each individual one, but there's nowhere I can just access it as one dataset. This displeases me.

Edit: Complain about not being able to find something and ye shall locate it: http://www.electoralcalculus.co.uk/orderedseats.html

Edited by Brynneth

Share this post


Link to post
Share on other sites

The site I edited to link has a comprehensive set of %ge vote share per constituency which is perfect for what I need - I've been able to import into Excel and set up a number of filters to sort the rows by preference for each party.

I'm going to try assigning seats to parties in rounds based on the overall seat share. If we take the table you posted earlier:

PARTY SEAT

Grn 24

SNP 31

Lab 199

LD 51

Con 242

UKIP 82

Then divide this by the lowest value, round to the nearest whole number and sort highest to lowest:

PARTY SEAT

Con 10

Lab 8

UKIP 3

LD 2

SNP 1

Grn 1

This gives me the number of seats and order in which I'm going to allocate to a party at a time, based on their %ge vote share in that constituency. So the Tories get their top 10 seats, then Labour get their top 8, then UKIP get their top 3, etc, etc. If a seat has already been taken then I'll skip to the next on the list; ex. The Green's 4th highest-ranked seat is in the Labour Stronghold of Stoke Newington which Labour'll claim in the 3rd round, pick skips a seat to Norwich South which doesn't have more than a 40% share for any party.

Share this post


Link to post
Share on other sites

Yeah I have no idea what that system is supposed to be. It isn't going to tell you anything interesting because you literally just made it up and it makes no logical sense.

 

However electoral reform shouldn't be argued against on the basis that it might give less savoury parties more of a say.

It's not about "less savoury", whatever that means. The question is whether a system that requires multiple parties to group together to form a coalition necessarily produces a government that will enact policy that is closer to the views of the country in general than under a system where the largest party typically forms a majority. Where you have something like a Con/LD coalition the answer may be yes since the LDs were seen to block the more extreme Tory policies and pushed through a few of their own that were closer to things that Labour voters would have wanted. But where you have something like a Con/UKIP coalition, even if these parties collectively represent a majority of voters, it can produce policies that are very far away from what the remaining 40% would have wanted to see.

The issue is that which one of these you end up with can have very little to do with the vote share and a lot to do with who the biggest party thinks will agree with them on most points.

Share this post


Link to post
Share on other sites

Short answer is yes. It's mostly just an intellectual exercise to draw a map of what the UK might look like with PR, I couldn't think of a fairer way to allocate seats. I'm about 1/3 of the way through and so far it's only the Greens which haven't been taking seats in which they were 2nd place by a few percent in a marginal area, partly because of their unique nature of not being strongly regionally affiliated.

Share this post


Link to post
Share on other sites

The question is whether a system that requires multiple parties to group together to form a coalition necessarily produces a government that will enact policy that is closer to the views of the country in general than under a system where the largest party typically forms a majority. Where you have something like a Con/LD coalition the answer may be yes since the LDs were seen to block the more extreme Tory policies and pushed through a few of their own that were closer to things that Labour voters would have wanted. But where you have something like a Con/UKIP coalition, even if these parties collectively represent a majority of voters, it can produce policies that are very far away from what the remaining 40% would have wanted to see.

The government at the moment is arguably going to produce policies that aren't close to the views of the majority (63% based on votes), I don't see a proportional systems being much worse than that.
 
 

The issue is that which one of these you end up with can have very little to do with the vote share and a lot to do with who the biggest party thinks will agree with them on most points.

The parties that they think will agree with them still have to have a sufficient number of votes though.  It doesn't matter if the tories think Party X will agree with them on most things if party X doesn't have the votes to actually do anything.
 
As I said though, in a system that's likely to produce coalitions, you can vote with that in mind.

Share this post


Link to post
Share on other sites

 

The government at the moment is arguably going to produce policies that aren't close to the views of the majority (63% based on votes), I don't see a proportional systems being much worse than that.

 

 

That isn't entirely true. The more extreme LD voter and less extreme UKIP voter are likely to be reasonably satisfied with a Tory government or rather, more satisfied with it than the other likely alternative (Labour). You can't just look at the Tory vote in isolation to make that statement.

Share this post


Link to post
Share on other sites
That isn't entirely true. The more extreme LD voter and less extreme UKIP voter are likely to be reasonably satisfied with a Tory government or rather, more satisfied with it than the other likely alternative (Labour). You can't just look at the Tory vote in isolation to make that statement
I've seen thinking that UKIP took more votes away from Labour than the conservatives (working man blaming immigration) so that's hard to judge.
 
If you took a simple 50/50 split on the UKIP / Lib Dem vote and assume that they'd be happyish with the result, you'd still get 53% of the country not happy.

Share this post


Link to post
Share on other sites

If you look at the constituencies in which UKIP came second place, they're a pretty even 50/50 split between Labour and Tory. Presumably a split between Sun readers and Mail readers.

Share this post


Link to post
Share on other sites

I don't see a proportional systems being much worse than that.

"Let's change this thing"

"Will that produce a better result?"

"I don't see it being much worse."

This is the exact point I'm making.

Share this post


Link to post
Share on other sites

And yet the only actual data that we have suggests that it wouldn't produce a particularly different outcome in terms of the government formed, except that UKIP would be involved.

Now before you say the same thing again about the dangers of using the election vote share data to produce a meaningful PR outcome, I know, here's the quote from my original post:

This is all a bit vague because [...] people would undoubtedly vote differently if they knew that the result was based on PR. Typically you'd expect to see more votes for smaller parties that currently "can't win" in a given area. Mainly that would mean more votes for UKIP and the Greens and more votes for the Tories in Scotland. Possible a bigger LD vote too, given how few they got this time around. It's really hard to say how this would all interact and work out.

Are there any historic examples of elections that switched from FPTP (or similar) to PR? I can't find any articles about it, which is a shame, as it would be interesting to see how much the vote changed and maybe we could make some speculations about what would happen to our parties.

Still, I can't think of any reasons for the assumptions above to be incorrect (can you?). It seems rather likely we'd be operating under a Con/UKIP coalition right now. Possibly a Con/UKIP/LD. Can you explain in what sense you see it "potentially being much better"? What vote changes are you expecting and in what sense does it produce a better result?

Share this post


Link to post
Share on other sites
It's not just about government formed though, it affects the make up of the opposition as well which is also important.

 

I'm about to leave work but might type something up later.

Share this post


Link to post
Share on other sites

Interesting factoid from The Register, each SNP MP represents 25,972 people whereas the 1 UKIP MP represents 3.8 million.

Share this post


Link to post
Share on other sites

(That's a factlet, not a factoid.)

When we're done with PR we can talk about boundary changes. I assume everyone who wants votes to be equal and fair is in favour of these too?

Edit: In case you're not familiar:

http://www.electoralcalculus.co.uk/boundaries2013.html

Even though the review will be done from scratch, it is likely to look a lot like this.

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
Sign in to follow this  

×