johng
Labour
Posts: 4,850 
|
Post by johng on Nov 25, 2021 17:19:14 GMT
So therefore, is it safe to assess that: - The New Parliament will have a number of members divisible by 3
- The New Parliament will be elected solely by the Single Transferable Vote
- The next elections will be for the New Parliament
In which case, should we open a new thread with the title "Design the new electoral map for Wales for 2026"?
1- I don't see why that would be the case at all. 2- Probably. 3- Almost certainly. This will be put through the Senedd during Drakeford's term as FM.
TBH, I am not sure cibwr is right in what he says. That panel reported in 2017 and the goalposts have moved since then. For one, it reported over four years ago and there is already one committee report, led by Dawn Bowden, that is influenced by it. If you look at the Bowden led report from 2020, you will see clear differences in approach. There is clear evidence of realpolitik insertion over idealism. When that report was written, the 40 Assembly constituencies looked like they could possibly be staying. Now we know Wales is almost certainly going down to 32 Those 40 constituencies paired up would have significant population variations. Ynys Mon and Arfon (most likely pairing) would have 94k residents whilst Cardiff South and Penarth and its smallest neighbour Cardiff Central would have 143k residents. The boundaries are so out of date, it would be foolish, indeed stupid IMO, to use them with any new system.
Another core difference is size. The 2017 report envisaged a smaller chamber that what we will now likely get. The goalposts have moved and are now 80 to 100 with something in the 90s being far more likely IMO.
|
|
|
|
Post by Penddu on Nov 26, 2021 6:24:50 GMT
So our starting point is:
80-100 seats
Based on new Westminster constituencies (32 seats)
Probably STV
So if based on pairs of seats, then 16 x 5 = 80 or 16 x6 = 96. 80 seats more likely...
|
|
|
|
Post by Pete Whitehead on Nov 26, 2021 10:06:49 GMT
How about the 32 new Westminster constituencies each electing two MPs - one male one female. I don't suppose it wouldd be practical to divide the electorate on that basis and have women only vote for the female MP and men for the male MP. This is similar to a (frankly rather silly) proposal by Teresa Gorman many years ago - ticks the gender quota box
Then four regions electing 9 top up MPs. I'd favour using the votes from the combined constituency vote to allocate here so there isn't the gaming that happens with additional votes.
North Wales - Montgomeryshire and the seats covering the former counties of Clwyd and Gwynedd
Mid and West Wales - Brecon & Radnor, the four Dyfed seats and the three Swansea seats
Souuth Wales - the four Cardiff seats, Vale of Glamorgan, Bridgend, Aberafan Porthcawl and Rhondda
South East Wales - the remainder
I'm not especially concerned about the over-representation of Yns Mon but you could give 8 top up seats to North Wales which would give a total of 99
|
|
|
|
Post by East Anglian Lefty on Nov 26, 2021 12:19:53 GMT
I would suspect that STV with 3 member seats linked to the Westminster ones is probably the most likely outcome. Alternatively, given the wide variation in current constituency sizes, you could keep 40 seats, each electing 3 members, but rejig the constituency boundaries.
Rough maths on this, assigning one seat to Ynys Mon and 39 to the mainland:
Clwyd: 6.53 quotas (breaking down neatly into 2 for Flintshire, 3 for Wrexham and Denbighshire and 1.5 for Conwy)
Gwynedd: 1.50 (one seat shared with Conwy)
Powys: 1.79 (you can either plead special geographic considerations and have two small seats, or you can pair them, probably with Monmouthshire)
Dyfed: 5.04 (one for Ceredigion, 4 for Pembrokeshire and Carmarthenshire)
West Glamorgan: 4.93 (Swansea could stand alone for 3, but NPT needs a 10% variance to stand alone)
Mid Glamorgan: 5.63 (RCT can stand alone for 3 and I guess Merthyr could pair with Caerphilly for 3, but Bridgend would need either a 10% variance or a partner)
South Glamorgan: 6.04 (though with a 5% variance the Vale would be ugly, so sticking the west of the Vale in a Bridgend seat might help
Gwent: 7.54 (Newport works for 2, there are various pairings that work for other authorities but it might be awkward in places.)
That all seems reasonably plausible, so I might have a go at mapping that out over the weekend as a thought exercise.
|
|
|
|
Post by therealriga on Nov 27, 2021 20:15:49 GMT
Again read the expert report and their suggestions based on pairing of current westminster boundaries and the varying number of AS that each would have depending on the size of the Senedd and the alternative method using local government units. They also propose a boundary commission and suggest that STV is the most favoured method of election - lists as a poor second using multi member constituencies as above. So therefore, is it safe to assess that: - The New Parliament will have a number of members divisible by 3
- The New Parliament will be elected solely by the Single Transferable Vote
- The next elections will be for the New Parliament
In which case, should we open a new thread with the title "Design the new electoral map for Wales for 2026"?
Strictly speaking, there's no reason why all constituencies should elect an identical number of members under STV or any other PR system. In fact you'd get much more flexibility and better boundaries if you didn't. I did a map a while ago based on a 90-member Welsh Assembly with 4-6 member seats and Ynys Mon as a 3-seater. The downside I can see is that some of the seats are large so I suspect that if they do settle for that system it's going to be 3 or 4-member seats. /photo/1 |
|
cibwr
Plaid Cymru
Posts: 3,599
|
Post by cibwr on Dec 1, 2021 9:41:30 GMT
The best system would be 13 STV constituencies based on the historic counties. At a push, Glamorgan could be two, Cardiff and Not Cardiff That would be a rather odd situation. Radnorshire with its population of 25k and Glamorgan with its 1.3 million (or 950k and 350k if split like that). cibwr - Where is the political appetite for retaining the current boundaries? My feeling is that there is a strong appetite for the opposite and maintaining the link between Westminster and Senedd. Local authority STV seats has its difficulties too. The main one, for me, is that we have too many small authorities which exist for no genuine reason and should be merged into more viable ones.
The expert review suggested the current westminster constituencies or local government boundaries as templates, but said we need a boundary commission to draw up new boundaries. The issue with using Westminster constituencies is that by the very nature of the system those would change at every election - which would be better avoided. Based on the current westminster or local government boundaries with mulit members you can vary the members (matching population changes) rather than altering the boundaries. |
|
cibwr
Plaid Cymru
Posts: 3,599
|
Post by cibwr on Dec 1, 2021 9:49:42 GMT
Also the expert panel said that 3 is too small for a proportional system, ideally they want 4 - 6 members with most around 5. Pairing Westminster constituencies I think will be a stop gap for one election only. If the Richards report had been followed I suspect we would have had STV and new boundaries in time for the third Assembly elections.
|
|
johng
Labour
Posts: 4,850
|
Post by johng on Dec 1, 2021 13:09:42 GMT
That would be a rather odd situation. Radnorshire with its population of 25k and Glamorgan with its 1.3 million (or 950k and 350k if split like that). cibwr - Where is the political appetite for retaining the current boundaries? My feeling is that there is a strong appetite for the opposite and maintaining the link between Westminster and Senedd. Local authority STV seats has its difficulties too. The main one, for me, is that we have too many small authorities which exist for no genuine reason and should be merged into more viable ones.
The expert review suggested the current westminster constituencies or local government boundaries as templates, but said we need a boundary commission to draw up new boundaries. The issue with using Westminster constituencies is that by the very nature of the system those would change at every election - which would be better avoided. Based on the current westminster or local government boundaries with mulit members you can vary the members (matching population changes) rather than altering the boundaries. I don't follow your logic here. Why would Westminster boundaries change 'every election'? Historically, they haven't really changed that much. The upcoming change will be the biggest in quite some time and will sort out some massive anomalies in seat size. We already have a boundary commission drawing up boundaries. You can't vary numbers much on the current boundaries unless you have a huge Senedd with several hundred members. Paired makes it a bit easier, but I think it would be very odd to based Senedd seats on paired old Westminster seats. It's not exactly great with local authorities boundaries either.
You talk about the expert panel, but you forget that there has been a committee led by Dawn Bowden report since then that was influenced by, but certainly didn't parrot that report, and that politicians will have to agree and vote on a system.
|
|
|
|
Post by 🏴☠️ Neath West 🏴☠️ on Dec 1, 2021 13:26:25 GMT
Also the expert panel said that 3 is too small for a proportional system, ideally they want 4 - 6 members with most around 5. Pairing Westminster constituencies I think will be a stop gap for one election only. If the Richards report had been followed I suspect we would have had STV and new boundaries in time for the third Assembly elections. These "experts" seem to be excessively keen on parties winning seats by coming fourth. Looks blatantly partisan to me. |
|
|
|
Post by 🏴☠️ Neath West 🏴☠️ on Dec 1, 2021 14:07:43 GMT
The best system would be 13 STV constituencies based on the historic counties. At a push, Glamorgan could be two, Cardiff and Not Cardiff If we're doing pre-1974 counties, then Cardiff, Swansea, Newport, and Merthyr Tydfil were county boroughs and were treated as counties for the important psephological purposes of voting to go wet on Sundays. Of course, back then three of the four county boroughs were much smaller than their current boundaries. But that would still leave Glamorgan far too large. Perhaps there could also be a Glamorgan Boroughs constituency to cover the middle of Neath, Bridgend, Cowbridge, and Barry. |
|
|
|
Post by 🏴☠️ Neath West 🏴☠️ on Dec 1, 2021 14:30:44 GMT
Some one-member constituencies, some 8 or 9 member ones. The advantage of housing outdated boundaries is that they are not gerrymandered. Maybe we could set a floor at 3 members. Give each county 2 members just for existing, like the U.S. Electoral College, then do the rest by population according to the Huntington-Hill method. After all, over-representing Meirionnydd would be right up the "experts'" farm track. |
|
|
|
Post by minionofmidas on Dec 1, 2021 15:22:02 GMT
The expert review suggested the current westminster constituencies or local government boundaries as templates, but said we need a boundary commission to draw up new boundaries. The issue with using Westminster constituencies is that by the very nature of the system those would change at every election - which would be better avoided. Based on the current westminster or local government boundaries with mulit members you can vary the members (matching population changes) rather than altering the boundaries. I don't follow your logic here. Why would Westminster boundaries change 'every election'? Historically, they haven't really changed that much. The upcoming change will be the biggest in quite some time and will sort out some massive anomalies in seat size. the current legal framework is calling for constant reviews. Obviously these future changes won't compare in size to the ones now in the pipe, but it's certainly a problem in any system where Westminster constituencies are to be paired or otherwise grouped (presumably by a separate, if quicker, review if I know my Brits...) Not an insoluble problem, but nevertheless something that needs to be borne in mind. You don't want a Senedd election on outofdate Westminster elections that mostly closely resemble the new ones just because the regions review hasn't been completed yet - no voter is going to understand that. From that pov it's best to decouple as in Scotland. |
|
nyx
Non-Aligned
Posts: 1,065
|
Post by nyx on Dec 1, 2021 16:13:22 GMT
The best system would be 13 STV constituencies based on the historic counties. At a push, Glamorgan could be two, Cardiff and Not Cardiff That's similar to my proposal for STV constituencies based on current local authorities (I suggested this in another thread a few months ago). It'd be a 98-seat Assembly, which isn't too bad. Only downside is that there'd be some population deviation, but nothing too major.  |
|
Chris from Brum
Lib Dem
What I need is a strong drink and a peer group.
Posts: 9,771
Member is Online
|
Post by Chris from Brum on Dec 1, 2021 16:30:16 GMT
Cardiff and Swansea have to be split, surely?
|
|
|
|
Post by Davıd Boothroyd on Dec 1, 2021 16:40:12 GMT
If you were doing it that way I would intervene in several areas to even up the seat numbers. Anything over five is probably too many, and two is too few.
|
|
myth11
Non-Aligned
too busy at work!
Posts: 2,843
|
Post by myth11 on Dec 1, 2021 20:41:25 GMT
1) I misunderstood what you were saying here. Personally though I think a party should have to win a majority to win a majority government. What could be fairer? 2) I said that voters don't have to play guessing games. Which they don't. If a party only gets one candidate elected when they would be 'entitled' to two, that is because their second candidate is not popular enough. I don't have a problem with that. The issue I accepted was that parties 'have' to play guessing games. Apart from they don't 'have' to, they choose to. 3) This is the same under every voting system? Definitely FPTP. At least this way you get to vote for someone you want to first, rather than just voting for someone you dislike to stop the candidate you really dislike as under our current system. 4) Well it depends how rough roughly is and how much 'a bit' over is. I'd rather it was proportional, but I think top-up lists should be regional to allow for regional parties. That shouldn't affect many seats, and is a payoff I'd be willing to make. There are issues with forcing parties to put up full lists when it comes to STV. 1.The parties will still try to game it with "strong" and "weak candidates". 2. Its increases the chances of "most votes no seats" due to transfer leakage and other parties using point one which will reduce the proportional value of the system. |
|
cibwr
Plaid Cymru
Posts: 3,599
|
Post by cibwr on Dec 2, 2021 9:10:26 GMT
The best system would be 13 STV constituencies based on the historic counties. At a push, Glamorgan could be two, Cardiff and Not Cardiff If we're doing pre-1974 counties, then Cardiff, Swansea, Newport, and Merthyr Tydfil were county boroughs and were treated as counties for the important psephological purposes of voting to go wet on Sundays. Of course, back then three of the four county boroughs were much smaller than their current boundaries. But that would still leave Glamorgan far too large. Perhaps there could also be a Glamorgan Boroughs constituency to cover the middle of Neath, Bridgend, Cowbridge, and Barry. one of the plans in the late 19th/early 20th century had two elected from each westminster constituency for a Welsh house of commons and two elected from each county and county borough to make up a senate..... |
|
johng
Labour
Posts: 4,850
|
Post by johng on Dec 2, 2021 11:48:07 GMT
If you were doing it that way I would intervene in several areas to even up the seat numbers. Anything over five is probably too many, and two is too few. Then you're just creating new constituencies so there'd be no point looking at local authority boundaries. Wales' local authorities often make little sense so I am not sure how anyone can think it's a reasonable proposition to use them as constituencies. We may as well use Boogie's idea of historic counties. There is no perfect system because both rival parties and voters seek to game every system and because the needs of society vary over time. The arguments for change are invariably made in bad faith by parties seeking advantage. The arguments for no change are similarly so. That's often true, but I don't think it's true in this case from Labour's point of view. Moving to a Senedd wholly elected by FPTP would benefit Labour in Wales, but that's not on the agenda. I think that's a shame as the 32 Westminster seats split into three (2 for Ynys Mon) FPTP seats would allow truly local representation.
I think one issue with Cibwr's view is that many Plaid members (the expert panel he refers to was headed by former Plaid candidate Laura McAllister) would actually like an eternal Labour-Plaid coalition. They misuse words like 'fair' to put their point forward when what is actually fair is people getting the government they voted for rather than something cobbled together by politicians after the election. In our system, that's the way it works and is the way it should work.
I also think a few too many of their politicians like the lack of accountability and scrutiny that comes for being elected on obscure regional lists. The idea of putting their record forward to local constituents makes them feel uneasy (quite rightly for a number of them as Leanne Wood recently found out).
|
|
|
|
Post by therealriga on Dec 2, 2021 12:50:44 GMT
If you were doing it that way I would intervene in several areas to even up the seat numbers. Anything over five is probably too many, and two is too few. Then you're just creating new constituencies so there'd be no point looking at local authority boundaries. Wales' local authorities often make little sense so I am not sure how anyone can think it's a reasonable proposition to use them as constituencies. We may as well use Boogie's idea of historic counties. There is no perfect system because both rival parties and voters seek to game every system and because the needs of society vary over time. The arguments for change are invariably made in bad faith by parties seeking advantage. The arguments for no change are similarly so. That's often true, but I don't think it's true in this case from Labour's point of view. Moving to a Senedd wholly elected by FPTP would benefit Labour in Wales, but that's not on the agenda. I think that's a shame as the 32 Westminster seats split into three (2 for Ynys Mon) FPTP seats would allow truly local representation.
The Westminster constituencies often split communities for electoral equality reasons. That will be even more the case with the new 5% maximum deviation. If you were going for a 98-member FPTP assembly it would be far better to draw them from scratch. I can't see any value obtained from them being subdivisions of Westminister seats. |
|
|
|
Post by 🏴☠️ Neath West 🏴☠️ on Dec 2, 2021 12:58:43 GMT
The best system would be 13 STV constituencies based on the historic counties. At a push, Glamorgan could be two, Cardiff and Not Cardiff That's similar to my proposal for STV constituencies based on current local authorities (I suggested this in another thread a few months ago). It'd be a 98-seat Assembly, which isn't too bad. Only downside is that there'd be some population deviation, but nothing too major. I'd suggest setting a floor of 3 and then apportioning the rest of the constituencies by the Huntington-Hill method. For 100 seats that would change the following counties from your map: - Blaenau Gwent 2→3
- Bridgend 5→4
- Ceredigion 2→3
- Isle of Anglesey 2→3
- Merthyr Tydfil 2→3
- Swansea 8→7
The only wrong-way roundings in that distribution are: - Bridgend 4 (4.66)
- Cardiff 11 (11.64)
- Neath Port Talbot 4 (4.55)
- Rhondda Cynon Taf 7 (7.65)
- Swansea 7 (7.83)
Besides those caused by the floor; - Blaenau Gwent 3 (2.22)
- Ceredigion 3 (2.31)
- Isle of Anglesey 3 (2.22)
- Merthyr Tydfil 3 (1.91)
|
|
