Tory/SNP DevoMax/boundaries deal??

[No Offence Alan]

Jan 15, 2013 22:35:23 GMT East Anglian Lefty said:

Wales may punch above its weight at the moment, ...

How many MPs for Welsh seats are in either the Cabinet or the Shadow Cabinet?

[cibwr]

Good point Alan on cabinet/shadow cabinet membership. The fiscal transfers to Wales are not unusual, you see such mechanisms in other federal/semi federal countries. However it is something of an indictment of the union that the central government has permitted such disparities to arise. If Barnett was replaced with a needs based formula then Wales would actually receive more money. I don't actually think the neglect of the economic needs of Wales are as a result of a vindictive government - just that Wales is too small to matter to central government.

[erlend]

I have seen Labour supporters (here?) arguing for seat size to be based on population rather than electorate. That seems like a major change. If Labour had a majority would Labour supporters expect that to happen without consensus?

[Deleted]

I have said many times that all issues like this, voting systems, HoL etc should go to a fully independent panel to recommend the best law then that gets put to the house to vote on.

This failed law was just nonsense in how it was devised and the aims it had and deserved to die but we should look at more equal sizes only where it is sensible. I see no justification to reduce numbers.

[The Bishop]

The reduction in MPs was originally a bit of red meat thrown by Dave to the baying mob at the height of the expenses furore, let's not forget........

Despite the fact it will save little money nor improve governance in any discernable way - especially with the "payroll vote" remaining at its present bloated size.

[Davıd Boothroyd]

Jan 17, 2013 11:51:40 GMT erlend said:

I have seen Labour supporters (here?) arguing for seat size to be based on population rather than electorate. That seems like a major change.

Merely reverting back to the old way of redistributing used until 1944. Several other countries do it this way.

It's possible there is an argument to be had over whether to use resident population of voting age instead of all population. But it's the obvious solution to the problem of incomplete electoral registers, which is likely to be made significantly worse by individual electoral registration.

[erlend]

Actually you have a point. But the point others were making was whether it is reasonable to make serious changes without consensus. This would likely be without consensus.

[East Anglian Lefty]

Further to my discussion with the_bullies, I've mapped out a 35 seat Wales, all within 5% deviation (though with a few alternatives involving a slightly greater variability). There are a few awkward seats, but that's the case at present too and the only areas that seem to preclude a convenient solution are Pontypridd and Port Talbot.

With 35 seats, the quota would be 65188, with an acceptable range from 61929 to 68448 electors. I briefly considered allowing Anglesey to form a seat on its own, but I decided against it as you can't call the Menai a significant natural boundary when Bangor has suburbs on both sides of the strait.

If we begin with north Wales, the preserved counties of Clywd and Gwynedd have enough population for 8 seats. You can further subdivide that, making 2 seats from the principal authorities of Anglesey and Gwynedd and 6 from the rest of north Wales, but that makes for a lot of seats right by the lower limit in Clwyd and forces you to either draw an A5 seat or split a town, so it's not my preferred option.

Option 1:



Menai (62,312) - Anglesey, Bangor and a few miles down the coast to get Gwynedd within acceptable limits. More or less draws itself.
Gwynedd (66,929) - merges Dwyfor Meirionydd and the Caernarfon parts of Arfon. Given that Dwyfor Meirionydd already gets fairly close to Caernarfon, this doesn't really make the seat more unwieldy than it already is.
Aberconwy (64,051) - keeps what it already has, then adds a couple of wards along the coast plus four more in the interior.
Colwyn & Rhyl (64,973) - The coastal parts of Clwyd West is added to Rhyl and St. Asaph from Vale of Clwyd. I'd argue this is significantly more coherent than the current Clwyd West, where the coastal bits are widely separated from the parts round Ruthin.
Prestatyn & Flint (67,297) - almost all of Delyn, plus Prestatyn from Vale of Clwyd. Might make this seat a little more marginal, but not by much.
Alyn & Deeside (63,075) - just adds Lockwood ward.
Clwyd South (63,425) - gives up Coedpoeth and points north-east to Wrexham, and in return takes the southern half of the Vale of Clwyd.
Wrexham (63,709) - expands a bit, but maintains it basic orientation.

Option 2:



This is the variant for those who'd rather see fewer authorities split and aren't too bothered about a constituency stretching from Snowdonia almost to the English border.

Gwynedd, Flint & Prestatyn and Alyn & Deeside are unchanged from option 1.

Menai (68,098) differs only in that it grabs Bethesda and environs, which makes good sense on community grounds.
Conwy Coast (61,976) drops most of its rural hinterland in exchange for Colwyn Bay and Old Colwyn.
Vale of Clwyd (62,605) is most of the current seat plus Abergele and scattered rural areas. I don't understand the political dynamics of north Wales very well, but it looks to me like it'd be a pretty safe Tory seat.
Llanwrst and Glyndwr (63308) is my suggested name for a seat that's frankly an abomination. Starting on the A5 outside Wrexham, it draws in Ruthin, but otherwise keeps heading west until Betws-y-Coed, from where it gathers the watershed of the upper Conwy.
Wrexham (62,483) is not much changed from option 1. It's altered only in that it heads south-east, because otherwise the A5 seat would be even stupider.

Next up, Mid and West Wales minus Dwyfor Meirionydd is just the right size for six seats. You could technically get two whole seats out of Carmathenshire and four from Pembrokeshire, Ceredigion and Powys, but that would cause no end of problems, split Ceredigion unnecessarily and make some pretty stupid looking-seats. Instead I've elected to treat them all as a unit.

Option 1:



Montgomery & Radnor (65,025) - there isn't the population in Powys for two whole seats, but there's nowhere outside the county Montgomery can sensibly be connected to. So this is the seat that stays in Powys, heading south as far as Llandrinod Wells to pick up population.
Brecknock and Dinefwr (66,786) - vary spelling according to taste. Crossing over the mountains in south Wales is obviously out, so the Amman Valley is just about the only direction this seat could head in. It just about managed to pick up Ammanford and connecting areas along the A40 without becoming too large. (Note: since drawing these, I've noticed I could put more of historic Radnorshire into Montgomery and Radnor, whilst still leaving this within quota.)
Ceredigion & Teifi Valley (65,716) - Yes, north Pembrokeshire is the traditional option, but Newcastle Emlyn does extent both sides of the county boundary. That and a few villages gets you to the quota with room to spare.
Llanelli (64,008) - largely unchanged, just grabbing two wards which appear to look towards Llanelli anyway.
Carmarthen and South Pembrokeshire (62,724) - not that dissimilar from the existing constituency of almost that name. Grabs the remainder of Carmarthen East, but sheds Narberth and environs.
Preseli Pembrokeshire (63,145) - just adds Narberth.

Option 2:



This is a variant for those who'd rather see Ceredigion go into Pembrokeshire. Montgomery & Radnor, Brecknock & Dinefwr and Llanelli are all unchanged from option 1.

Ceredigion & North Pembrokeshire (64,708) - more or less what it says on the tin. Gets up to the outskirts of FIshguard, but doesn't take the town itself.
Carmarthen (64,912) - I've gone for a short name, as the Pembrokeshire element is only around a third of the seat and doesn't contain any town large enough to be worth mentioning.
West Pembrokeshire (61,965) - notably coastal constituency. Pembroke is only connected to the rest of the seat by the A48 bridge.

That'll do for now. I'll continue with the rest once I've had time to type it up. Criticise away.

[Pete Whitehead]

I like Montgomery & Radnor. Much better than the other options we've seen for splitting Powys

[Sibboleth]

I suppose you could call Llanwrst and Glyndwr 'The A5'...

[the_bullies]

Up until the 1980's there were between 36 & 38 seats (a couple more than necessary as we have seats that are geographically isolated like the isle of wight & the highlands & islands) so I think it's possible to decrease to the original 1970 to 1980 boundaries in some places. Especially in North East Wales (Clwyd) & South West Wales Cardigan & Pembrokeshire. If you publish your South Wales maps I'll let you know what I think.

[East Anglian Lefty]

Moving on to South Wales...

The three authorities of Swansea, Neath Port Talbot and Bridgend would have a theoretical entitlement to 6 quotas, but currently contains most of 7 seats (with Ogmore also taking part of its electorate from Rhondda Cynon Taff). So one seat has to disappear, and given that the Swansea seats have electorates over 60,000 whilst Aberavon barely clears 50,000 and only about 45,000 of Ogmore's electorate is in Bridgend authority, it's clear that the seat that disappears is going to be in the east of this grouping.

The problem is that the population of Bridgend authority, the Afan Valley and Port Talbot is just over 143,000, which is too much for two seats. That's not a problem if you're willing to allow a greater variance of electorate, but if not then realistically you have to split Port Talbot.*

Once you accept this, the rest is fairly simple:



Bridgend (66,579) - not much changed from its present orientation. Gains a trio of wards in the east of the seat on either side of the M4.
Aberavon and Ogmore (66,731) - does not include Aberavon ward. I've checked, and the Sandfields wards are contiguous with the rest of the seat, but I will accept that this is hardly a nice-looking seat.
Neath (65,599) - drops the Tawe Valley, replaces it with Coedffranc, Briton Ferry and the north end of Port Talbot.
Swansea East and Tawe Valley (65,037) - loses Cwmbwrla and Penderry, gains a couple of rural north Swansea wards and the valley of the Tawe.
Swansea West (65,521) - gains Cwmbwrla and Penderry, loses some outer suburbs to Gower. The partisan effect is probably around zero, but it might have tipped this into the LD column in 2010.
Gower (67384) - swaps northern rural wards for Swansea suburbs.

Next we get into the South Wales Central region for Assembly purposes. Here you can get eight seats out of RCT, Cardiff and Vale of Glamorgan. There are also eight here at present, so this is the only region that doesn't need to lose a seat. On the other hand, the electorate is far from evenly spread amongst these seats. Three of the Cardiff seats don't need any change, which Cardiff South and Vale of Glamorgan both need to lose wards and Rhondda and Cynon Valley are each undersized to the tune of about 10,000 electors.

No one authority can stand alone, so there's a need for two cross-border seats. The principle of least-change and the fact that the Cardiff seats need little change would both suggest that the current Cardiff-Vale seat (Cardiff S) should be preserved and a Vale-RCT seat created. I did play around with making the links Cardiff-RCT and Vale-RCT, but that's more change and the results are no better. Ideally, you'd want Cardiff-RCT and Cardiff-Vale, but I'm yet to come up with a version of that I like, because the ward shapes in the west of Cardiff are not helpful.

I therefore propose only one map:



Rhondda (64,334) - this was surprisingly simple. All you need to do is head further south, adding Tonyrefail and Giffard Goch. Insofar as I can see, the transport links look reasonable, though I'm willing to be corrected on this point.
Pontypridd and Aberdare (62,665) - there isn't room for the entirety of Pontypridd community, but there's enough of the town here to justify the name. This is Cynon Valley plus Pontypridd itself.
Llantrisant (62,727) - I've got no idea what to call this, so I've settled for just naming it after the largest element. The south end of RCT, plus enough of Vale of Glamorgan to get up to population. Internal communications aren't great, but neither do they seem appalling.
Vale of Glamorgan (62,777) - loses some rural elements, adds a couple of wards from Cardiff South & Penarth.
Cardiff South & Penarth (68,558) - loses the non-Penarth bits of the Vale it had. I've just noticed this would be outside 5%, but it can always lose a Penarth ward.
Cardiff Central (62,218) - no change.
Cardiff North (66,377) - no change.
Cardiff West (63,360) - no change.

Finally, we come to Gwent plus Merthyr Tydfil, which has an electorate that entitles it to 7 rather large seats. Torfaen and Monmouthshire are both slightly too large to have seats co-extensive with their authorities, although if there was a bit more leeway those would be obvious solutions.



Ebbw Vale (67,601) - Blaenau Gwent gains Blaenavon to get Torfaen within the quota, but proceeding further south would take it straight back out again. I therefore headed west, adding Rhymney. Last time that was included in this seat it was named Ebbw Vale, so I've changed the name back.
Merthyr and Bargoed (67,300) - fairly self-explanatory. If Torfaen were allowed to remain whole, Ebbw Vale could add Darren Valley, which would allow this seat to add Aberbargoed.
Caerphilly and Blackwood (67,309) - Made up Caerphilly and most of Islwyn. Internal communications aren't great, but if they'll do for a local authority they'll do for a parliamentary constituency.
Newport West & Risca (67,103) - and some other bits and pieces to get the Caerphilly seat within the quota.
Torfaen (65,271) - other than one orphan ward, co-extensive with Torfaen.
Newport East (65,730) - picks up four wards from west, drops all but one ward from Monmouthshire.
Monmouth (68,240) - other than one orphan ward, co-extensive with Monmouthshire.

*Theoretically you could instead put the Afan Valley with Neath, but a. that's less than neat and b. doing that would force you to split Bridgend, which is no better than splitting Port Talbot and is also a greater change.

I'm pretty sure I've made some hideous errors here, because I don't know South Wales at all, but I'm also relatively sure that if they're pointed out I can suggest alternatives. Wales with 30 seats might be pushing it too far, but 35 is not unfeasible in terms of respecting community identities.

[the_bullies]

You've fallen into the trap in the valleys which the electoral reform society fell into with its proposals which were just not coherent. In Rhondda Cynon Taff the two Northern constituencies Rhondda & Cynon Valley have rapidly shrinking electorates so really have to be merged (even though its not popular) & there are distinct communities which do not blend easily. You've made the mistake of splitting the 'community of Pontypridd' which the boundary commission itself rejected as unsatisfactory. Cynon Valley & Rhondda do not have communities so there's not a problem there. Taff Ely as the South is called & where I live on the other hand is a commuter area for Cardiff & is a rapidly expanding area around Church Village, Llantrisant & Llanharan. This is where the majority of house building has taken place. And in the local development plan is to take place. It is a coherent area & a distinct community & the seat in some form has been together for nearly 100 years. The problem has been that other undersized seats have pinched wards from around the fringes to alleviate their population decline. Although in 2010 the figure for Pontypridd was 58219 it is already over 60000 in the latest figures. It is likely here that to return of wards previously removed, would present a viable seat that has historical links. This is what the Commission proposed in their final publication. I presented in writing Labour's proposals for our seat to the commission & attended the public enquiry & gave our submission. My proposals & subsequent analysis for our MP were 90% correct in South Wales & most of the ideas presented by me in advance of the initial proposals were accepted by Welsh Labour. I correctly forecast that our biggest problem would be in Gower, Swansea where a Labour seat turned into a LIbDem-Tory marginal. However this changed in the 2012 local elections & in the 2011 Assembly elections. South Wales is a nightmare for boundary changes in the main. But your ideas for Ogmore are good as the original Aberavon seat did cross the boundary at Porthcawl. And Ogmore was the only Bridgend seat. I proposed a coastal Porthcawl-Vale of Glamorgan seat minus Barry called Glamorgan Coastal which did have similarities & communities. But was the 10% where I failed, but I still think I was right.

[East Anglian Lefty]

OK, that's helpful. Do you have a list of what wards would have been in previous iterations of the seats? Assuming, of course, that there haven't been major changes in ward boundaries that make it impossible to get a clear picture.

Obviously if you merge Rhondda and Cynon Valley then the south of those two seats has to go somewhere. I take it that you think the best option is with Pontypridd?

And presumably as Taff Ely is a commuter area for Cardiff it'd fit better with Cardiff than the Vale? I ask because I think I may have found a way of sorting that configuration within 5% deviation, though I have to grab a ward from Bridgend to make the numbers work. It does mean that none of the Cardiff seats can remain unchanged, but none of them is entirely reshaped.

[cibwr]

Jan 17, 2013 11:51:40 GMT erlend said:

I have seen Labour supporters (here?) arguing for seat size to be based on population rather than electorate. That seems like a major change. If Labour had a majority would Labour supporters expect that to happen without consensus?

I would support seats based on population too rather than electorate. After all you are elected to represent the people of the area rather than those entitled to vote. Why should areas with say, higher than average number of retired people, as opposed to those with a younger population have greater representation. And of course if the disadvantaged tend not to register to vote that shouldn't mean that they lose out on representation. So population rather than electorate should be the criteria for deciding the quota.

[cibwr]

Jan 17, 2013 14:03:00 GMT East Anglian Lefty said:

Further to my discussion with the_bullies, I've mapped out a 35 seat Wales, all within 5% deviation (though with a few alternatives involving a slightly greater variability).
Carmarthen (64,912) - I've gone for a short name, as the Pembrokeshire element is only around a third of the seat and doesn't contain any town large enough to be worth mentioning.

The good burgers of Tenby would strenuously disagree... as would the inhabitants of Pembroke :-)

But on the whole a very workable division. The North Powys option is very much what the boundary commission floated at one time in the 1990s in one draft (they combined Brecon with Abergavenny).

[cibwr]

South Wales is very problematic because of the poor east west communication in the Valleys. Merthyr and Aberdare have greater affinity than does Aberdare and Rhondda.... Southern Taff Ely with Northern Cardiff, Pontypridd with Rhondda, Western Taff Ely with the Vale, Southern Taff Ely with Northern Cardiff. Penarth with South Cardiff would work still. Newport with Cwmbran. Then it gets very messy.

[the_bullies]

When doing the Boundaries for South Wales I like to work East to West starting at Monmouth. You can make a constituency from the wards which make up Monmouthshire CBC adding either one or two of the Eastern most Newport City wards, off the top of my head one of which is Langstone. Next we can make another constituency from the County Borough of Torfaen adding Croesyceiliog from Monmouth (East & West). But we need to add another Newport city ward either Caerleon or Malpas, I think Caerleon is the best fit. We can form a coherent constituency out of the wards that make up the centre of the city of Newport, we will take out some of the Western most wards to make a Cardiff East seat later. If we add some of Islwyn's northern most wards around Blackwood, Newbridge & Bargoed to Blaenau Gwent we get a Blaenau Gwent & North Islwyn seat. The remainder of Caerphilly added to South Islwyn gives us a coherent Caerphilly seat. I've added Aberdare & Aberaman to Merthyr Tydfil & Rhymney to give us a seat based around much of Keir Hardie's old seat. Merging Rhondda with Mid Cynon gives us a Rhondda & Mid Cynon seat (that includes the Rhigos Mountain Ward/Road). Adding Ynysybwl, Glyncoch & Cilfynydd in the North & Brynna, Llanharan, Llanharry & Gilfach Goch in the South gives us a Pontypridd seat. The Cardiff seats are almost large enough especially with population expansion so do not need major changes. You can create a viable Cardiff East seat by adding 4 of the Western most Newport wards including Tredegar Park. You can also create viable West & South seats by adding a ward each from Cardiff Central. Cardiff North does need wards around the Southern most part of Caerphilly (Rudri) & this is contentious but is arguable. The Vale is by itself large enough without Penarth. You can create a large enough Bridgend seat by dividing Ogmore between Aberavon & Bridgend.

[East Anglian Lefty]

Jan 17, 2013 21:13:20 GMT cibwr said:

Jan 17, 2013 14:03:00 GMT East Anglian Lefty said:

Further to my discussion with the_bullies, I've mapped out a 35 seat Wales, all within 5% deviation (though with a few alternatives involving a slightly greater variability).
Carmarthen (64,912) - I've gone for a short name, as the Pembrokeshire element is only around a third of the seat and doesn't contain any town large enough to be worth mentioning.

The good burgers of Tenby would strenuously disagree... as would the inhabitants of Pembroke :-)

The inhabitants of Tenby might disagree, but places with populations below 5,000 shouldn't appear in constituency names. And Pembroke isn't in that version.

[cibwr]

Ah remember Tenby is the home of Press Freedom (TM) and the equals sign so it thinks its self very important.