iain
Lib Dem
Posts: 11,453
|
Post by iain on Mar 27, 2016 22:11:53 GMT
Just had a go at a 7 seat Gloucestershire. It would be, quite simply, horrific. It would help the Forest of Dean enormously, but screw both Cheltenham and Gloucester completely. Having had a play around I think the best solution would probably be some kind of west Cheltenham-east Gloucester monstrosity ('Hatherley'). Stroud would then lose little bits to both Gloucester and Cotswold, with the nothern half of Cotswold, essentially Bourton-on-the-Water upwards tacked onto Tewkesbury district north of Cheltenham and Gloucester.
|
|
|
Post by timrollpickering on Mar 27, 2016 22:27:04 GMT
For the North East: Northumberland 3.31 quotas rounds to 3 seats (down 1). If we're using the old rules then the geographic consideration came into play with Northumberland (and also Cumbria with its three dimensional geography - ISTR someone here came up with a monster seat that both took as long to drive end to end as London to Manchester and also risked energy security). It didn't get so much attention, perhaps because it didn't get noticed or fit the general narrative of an anti-Conservative bias. More generally the old rules didn't have a strict allocation of seat numbers but allowed a divergence if this would better meet the other criteria * - it was this more than anything else that pushed up the number of seats Wales had. (* Northern Ireland was constrained to a 16 to 18 seat band.)
|
|
|
Post by East Anglian Lefty on Mar 27, 2016 22:32:19 GMT
There's a simpler option that that, if not a terribly lovely one. Forest of Dean and Cotswold are both co-extensive with their districts; Cheltenham hives a couple of wards off to Tewkesbury (probably on the west side of town? Either way a name change may in order); Gloucester loses some wards bordering the M5 (Quedgley, Grange and Tuffley looks like the best option to me) and the new seat is comprised of northern rural bits of Stroud, Quedgeley and eastern suburbs of Gloucester outside the city itself. It's not pretty, though.
|
|
|
Post by andrewteale on Mar 27, 2016 22:44:17 GMT
London has 72.95 quotas which should produce 73 seats (no change), however the fifth review didn't proceed like that, instead pairing boroughs to reduce the disparity between constituencies, or to reduce the number of seats allocated to a combined area, or both. The fifth review's pairing rules don't work on the current electorates: Bexley (2.35) needs pairing and has an entitlement of two seats; Greenwich (2.29) needs pairing and has an entitlement of two seats; Bromley (3.22) doesn't need pairing and has an entitlement of two seats; but Bexley+Greenwich is five seats and Bexley+Bromley is six.
Taking that problem into account, the following pairings seem to work. Retained from the fifth review: four seats for Barking and Dagenham (1.59) and Havering (2.53); five seats for Bexley (2.35) and Greenwich (2.29); three seats for Kingston (1.49) and Richmond (1.80). New pairings: Brent (2.60) and Harrow (2.30), five seats; Hammersmith and Fulham (1.50) and Hounslow (2.27), four seats; Lewisham (2.38) and Southwark (2.66), five seats; Newham (2.42) and Redbridge (2.54), five seats; Westminster with the City (1.59+0.09) and Camden (1.85), four seats. Bromley (previously paired with Lewisham) reviewed separately with three seats. Hillingdon (previously paired with Harrow) reviewed separately with three seats. Kensington and Chelsea (previously paired with Hammersmith and Fulham) reviewed separately with one seat. Lambeth (previously paired with Southwark) reviewed separately with three seats. Waltham Forest (previously paired with Redbridge) reviewed separately with two seats. Other boroughs with unchanged seat allocations.
These pairings give a total of 74 seats for London (up 1), with the new seat arising from the division of Hampstead and Kilburn.
|
|
|
Post by andrewteale on Mar 27, 2016 22:50:28 GMT
For the North East: Northumberland 3.31 quotas rounds to 3 seats (down 1). If we're using the old rules then the geographic consideration came into play with Northumberland (and also Cumbria with its three dimensional geography - ISTR someone here came up with a monster seat that both took as long to drive end to end as London to Manchester and also risked energy security). It didn't get so much attention, perhaps because it didn't get noticed or fit the general narrative of an anti-Conservative bias. Special geographical considerations were a fairly easy decision for the fifth review. The 2000 entitlements were 3.48 for Northumberland and 5.496 for Cumbria, both of which are above the harmonic mean anyway. Allocating four seats for 3.31 quotas is much harder to justify.
|
|
|
Post by andrewteale on Mar 27, 2016 22:58:53 GMT
Summing up, regional totals for England: East Midlands unchanged Eastern up 2 North East down 3 North West down 2 South East up 6 South West up 2 West Midlands down 1 Yorkshire down 1 London up 1 Total 537 (up 4)
|
|
iain
Lib Dem
Posts: 11,453
|
Post by iain on Mar 27, 2016 23:21:16 GMT
There's a simpler option that that, if not a terribly lovely one. Forest of Dean and Cotswold are both co-extensive with their districts; Cheltenham hives a couple of wards off to Tewkesbury (probably on the west side of town? Either way a name change may in order); Gloucester loses some wards bordering the M5 (Quedgley, Grange and Tuffley looks like the best option to me) and the new seat is comprised of northern rural bits of Stroud, Quedgeley and eastern suburbs of Gloucester outside the city itself. It's not pretty, though. That would probably end with having to split Stroud town.
|
|
|
Post by East Anglian Lefty on Mar 27, 2016 23:38:17 GMT
Nope - I'm not sure of the normally agreed boundaries of Stroud proper, as the area of continuous settlement appears to go beyond the civil parish, but this has got fairly clear separation: Electorate range is 3322 and under the old rules you've got way more leeway than that. If Stonehouse is a problem, you could swap it for Bisley and Chalford or at a pinch just take Berkeley and allow the Stroud seat to be a bit larger.
|
|
|
Post by East Anglian Lefty on Mar 27, 2016 23:57:02 GMT
West Midlands 26.55 quotas rounds to 27 seats (down 1). Coventry (2.99) keeps its 3 seats; Solihull (2.19) keeps 2; Birmingham (9.79) keeps 10. The seat which disappears is in the Black Country, with the Dudley/Sandwell/Wolverhampton group falling from ten seats to nine. Actually, I think you might see the Dudley/Sandwell/Wolverhampton group broken up, with three seats assigned to Sandwell and six to the other two. Which is liable to cause problems round Stourbridge, because some areas have to be tricky whatever the rules are... Although given that Walsall would only have 2.66 quotas, you could combine it with Wolverhampton for 5, then Sandwell and Dudley would sneak just over the harmonic mean for 7 constituencies.
|
|
|
Post by East Anglian Lefty on Mar 28, 2016 13:10:20 GMT
If you review Sandwell separately, you can create 3 seats easily enough but it's not going to be terribly pretty, because you have to divide the Rowley Regis area (or Smethwick, but that would be even worse and wouldn't satisfy rule 5 anyway.) WBE gains Wednesfield and loses Greets Green & Lyng, Rowley and Cradley Heath go to WBW, Warley gets Blackheath. You then have to form a Halesowen and Stourbridge East seat that is exactly as ugly as the name implies and merge the rest of Stourbridge with the western half of Dudley South. Dudley North picks up the remainder of Dudley South and loses a couple of wards in its north to compensate. The Wolverhampton seats then shuffle round. That's pretty hideous, so let's assume you keep Sandwell in the group. In this version, Sandwell is the same except Cradley Heath stays with Halesowen (worsening electoral equality, but on the old rules that ought to be OK.) Halesowen can then add Cradley and Quarry Bank, Stourbridge grabs Brierley Hill and Wordsley and the rest of Dudley South goes with Dudley town centre. Which leaves Gornal, Upper Gornal and Sedgeley to go with the the eastern half of Wolverhampton SE - the best balance of electorate I can get is if it keeps the two Bilston wards, Ettingshall and Spring Vale. Wolverhampton NE then gains East Park and Oxley and Wolverhampton SW takes the balance. The range in electorate is from 62757 to 75604, which you can probably get away with under the old rules. Picture below: Alternatively, we can group Wolverhampton and Walsall on the one hand, and Dudley and Sandwell on the other. This would probably be a good idea, because otherwise Aldridge-Brownhills would have the smallest electorate in the West Midlands. So that adds Pheasey and Walsall South takes Birchills Leamore and Blakenhall. (In an earlier version, it swapped Darlaston for Bloxwich, creating a rather nice Walsall Central, but unfortunately I think that would fail on rule 5 grounds.) The remainder of Walsall North then spills over the Wolverhampton border, adding Wednesbury and East Park (Willenhall and Wednesbury?) Wolverhampton NE then adds St. Peter's, Park and Tettenhall (Wolverhampton North), and the rest of Wolverhampton becomes Wolverhampton South. Dudley and Sandwell is then simplicity itself. Dudley North has to add Coseley East. The only other possible change is to move St. Thomas's from North to South, which reduces the electorate range. Everything else can stay the same. The range in electorate is from 60880 to 73172. Picture below:
|
|
|
Post by johnloony on Mar 28, 2016 13:44:52 GMT
I would actually prefer the boundary review to be done according to the old old rules, not just the old rules. In other words, no pairing of London Boroughs.
|
|
|
Post by East Anglian Lefty on Mar 28, 2016 15:36:03 GMT
If you give Wales the English quota, it's entitlement is 31.09.
Gwent is entitled to 5.85 seats and it no longer makes sense to combine it with Mid Glamorgan, so you can assign it six seats. Torfaen and Monmouthshire get one each and the rest gets four. My best guess is that works out as Blaenau Gwent gains the Rhymney Valley and Aberbargoed, Newport East comprising most of the town, a new seat covering western bits of Newport and Caerphilly town, and the remainder of Caerphilly going in with Islwyn.
South Glamorgan is entitled to 4.55 seats and Mid Glamorgan to 4.43, so combining them for 9 is sensible. Merthyr and RCT works well for 3 seats and you can then put Bridgend in with Cardiff and the Vale. Pontypridd can gains the RCT bits of Ogmore, Rhondda can take Hirwaun and Aberdare, Merthyr gets the rest of the Cynon Valley.
There isn't room for Bridgend town in Ogmore, so you have to add Portcawl instead. Bridgend can then go with western rural bits of the Vale, Barry and Penarth only need one Cardiff ward (almost certainly Grangetown) and then there are three seats for the rest of Cardiff, with Central probably the abolished seat.
West Glamorgan is entitled to 3.85 seats, so that's a loss of one. More than one way to go about that.
Dyfed is entitled to 3.87 seats. Two for Carmarthenshire one for Llanelli and the Amman Valley, one for the rest of the county. Two for Ceredigion and Pembrokeshire - with the cross-county seat reaching just beyond Fishguard.
Powys is entitled to 1.41 seats. You could pair it with Gwynedd, but as it's over the harmonic mean it's more likely they'd invoke geographical considerations to keep two seats.
Gwynedd is entitled to 1.84 seats. Assuming you don't link it with Powys, you can either consider it separately for two, or group it with Clwyd (5.30) for seven. Personally, I'd consider it separately, because there's a very simple split - Anglesey, Bangor and Bethesda in one, everything else in the other.
So that's a loss of two seats of Clwyd. Essentially the choice boils down to whether you have a single Conwy coast seat or not.
|
|
|
Post by Pete Whitehead on Mar 28, 2016 18:34:22 GMT
London has 72.95 quotas which should produce 73 seats (no change), however the fifth review didn't proceed like that, instead pairing boroughs to reduce the disparity between constituencies, or to reduce the number of seats allocated to a combined area, or both. The fifth review's pairing rules don't work on the current electorates: Bexley (2.35) needs pairing and has an entitlement of two seats; Greenwich (2.29) needs pairing and has an entitlement of two seats; Bromley (3.22) doesn't need pairing and has an entitlement of two seats; but Bexley+Greenwich is five seats and Bexley+Bromley is six. Taking that problem into account, the following pairings seem to work. Retained from the fifth review: four seats for Barking and Dagenham (1.59) and Havering (2.53); five seats for Bexley (2.35) and Greenwich (2.29); three seats for Kingston (1.49) and Richmond (1.80). New pairings: Brent (2.60) and Harrow (2.30), five seats; Hammersmith and Fulham (1.50) and Hounslow (2.27), four seats; Lewisham (2.38) and Southwark (2.66), five seats; Newham (2.42) and Redbridge (2.54), five seats; Westminster with the City (1.59+0.09) and Camden (1.85), four seats. Bromley (previously paired with Lewisham) reviewed separately with three seats. Hillingdon (previously paired with Harrow) reviewed separately with three seats. Kensington and Chelsea (previously paired with Hammersmith and Fulham) reviewed separately with one seat. Lambeth (previously paired with Southwark) reviewed separately with three seats. Waltham Forest (previously paired with Redbridge) reviewed separately with two seats. Other boroughs with unchanged seat allocations. These pairings give a total of 74 seats for London (up 1), with the new seat arising from the division of Hampstead and Kilburn. I don't think you've thought that one through. A Hammersmith & Fulham/Hounslow link up is going to create a hideous and incoherent seat of basically Chiswick & Shepherds Bush. Furthermore Hillingdon is pretty undersized for 3 whole seats so there is a far more natural pairing to be done there. It would make more sense to treat the 'West Central' boroughs all together and essentially merge Kensington and Westminster North, hiving off wards in the South and East to augment the other seats. Camden can stand alone for two seats
|
|
|
Post by Davıd Boothroyd on Mar 28, 2016 18:44:31 GMT
The problem with pairing Hounslow with H+F is that there's only one ward on each side which forms the border so the options are exceptionally limited.
|
|
|
Post by Pete Whitehead on Mar 29, 2016 9:41:55 GMT
Another area where a grouping of three boroughs might work better is in Redbridge/Newham/Waltham Forest. I don't know if the old rules allowed for this kind of thing in London though there are plenty of examples of it in the Met boroughs. In this case you could argue that minimal change, as the total number of seats remains unchanged at 7 - whereas pairing Newham and Redbridge necessitates a complete redrawing of all the seats in the are and none of the possible cross border seats look pretty. I wuld envisage Ilford South losing Cranbrook to Ilford North which also gains Wanstead while losing the two Woodford wards to Chingford & Woodford (which would also gain Snaresbrook). Walthamstow would then take Hale End and Valley wards from Chingford and lose Leabridge to Leyton. Leyton (less Wanstead and Snaresbrook) then crosses into Newham to take Stratford and Forest Gate north from West Ham which in turn takes Royal Docks (alternatively Green Street East) from East Ham which is otherwise unchanged
|
|
|
Post by timrollpickering on Mar 29, 2016 10:05:43 GMT
I don't think the rules themselves barred threesomes but the review commissioners were heavily opposed to it for no logical reason at all. But if individual seats are only in two boroughs then it's just BC laziness that prevents multiple combinations.
Newham is always going to be tricky to combine because of the three rivers and heathland surrounding it which leave few natural communities across borders - we may have an area called Little Ilford but just walk from it to Ilford under the North Circular and you'll realise the difference, whilst any seat that tried to use Wanstead Flats as its link would be torn to shreds at a review meeting. The main exception is in Forest Gate and Maryland where development (and the postcodes) flows into Leyton with only the different bins to tell you you've crossed a borough boundary (I used to live on a road with one end in Waltham Forest and the rest in Newham).
The boundary between the two Forest Gate wards is the Great Eastern Mainline. With four tracks in a cutting it's a pretty wide boundary and the limited crossings leave distinct areas either side. The Stratford & New Torn ward boundaries mainly follow railway lines and major roads but there is a somewhat awkward meander through streets in the south east - but it's hard to find border wards that can be easily hived off.
In terms of what to move from East to West Ham, the Royal Docks ward contains areas historically not part of the old East Ham borough (Silvertown and North Woolwich) and is highly detachable due to the docks themselves forming most of the boundary. Green Street East, by contrast, would take a very arbitrary chunk out of the middle of East Ham.
|
|
|
Post by Pete Whitehead on Mar 29, 2016 10:22:01 GMT
Yes Royal Docks was my preference for all those reasons. The alternative I suggested only because it achieves ratehr better equality of electorates, but i'd consider that a lesser priority
|
|
|
Post by therealriga on May 18, 2016 19:30:14 GMT
Let's look into this for England. The quota under the 2000 review rules is 70169. The East Midlands looks like only minimal change: Leicestershire (including Leicester and Rutland) 10.47 quotas rounds to 10 seats (no change). Leicester is 3.00 quotas so would likely continue to be considered separately with 7 seats for the rest of Leicestershire and Rutland. Lincolnshire 7.43 quotas rounds to 7 seats (no change). On the 2000 review rules they usually did try and get close to the quota and Rutland is now a separate county with an entitlement of 0.40. So, if overall number of seats is not an issue, there is an alternative which gets closer to the quota. You have Lincolnshire and Rutland with 7.83 getting 8 seats. Leicester city is reviewed separately with 3 seats and Leicestershire with 7.07 gets 7 seats. Welcome back to Rutland & Stamford!
|
|
|
Post by therealriga on May 18, 2016 19:34:32 GMT
I would actually prefer the boundary review to be done according to the old old rules, not just the old rules. In other words, no pairing of London Boroughs. But London boroughs were paired pre-1974, as the pre-1966 boroughs were smaller. It was only the second (1974-1983 constituencies) and third (1983-1997) reviews which avoided crossing London borough boundaries. So what you want is the new old old review.
|
|
|
Post by therealriga on May 25, 2016 17:41:35 GMT
North West: Lancashire (including Blackburn and Blackpool) 14.98 quotas rounds to 15 seats (no change). Overall the North West loses 2 seats. Lancashire currently has 16 seats, so that's a loss of one for the county and 3 overall in the North West.
|
|