New Welsh constituencies: Quick reaction

The embargo has been lifted on the proposals for the new Welsh constituencies, which reduce the number from 40 to 29.

This was always a tough task and under the circumstances I think they’ve done about as good a job as possible. I have mixed feelings about Arfon and Anglesey and Ceredigion. Cutting Caernarfon’s surrounding villages out of the former, and including Llanidloes in the latter feels unnatural to me, when other options were available. But I’m sure others have similar feelings about areas where they’ve grown up.

I do wish they would have pegged the number of MPs at 30 rather than 29, in order to make it easier for the Assembly constituencies to align with those in Westminster (with 30 AMs elected to constituencies and 30 on the list system, giving us a 50/50 split that would be more in line with Scotland).

Twenty-nine is an awkward number and pretty much ensures that Assembly and Westminster constituencies will diverge at some point, confusing the hell out of everyone, not least the political parties themselves.

Who will be happiest with the new constituencies? Leaving aside their obvious resistance to the reduction in the number of urban constituencies overall, probably Labour. The map looks about as good as it could have for them.

Who is punished the most? Again ironically, it’s the Conservatives, I think.

Plaid Cymru will have mixed feelings, but the map isn’t as bad as it could have been for them. The Lib Dems could well lose their only seat, but there's hope for them too.

The Conservatives

Gower will be a tough ask for the Conservatives now that it includes a large portion of Swansea West. I would expect this seat to swing back to Labour.

Their two seats in Pembrokeshire will be reduced to one – they have little chance of taking Ceredigion and North Pembrokeshire.

The Vale of Glamorgan will be split to include parts of Bridgend and Cardiff South. Again, Labour could probably snaffle both those seats.

Vale of Clwyd, Clwyd West and Aberconwy are currently in Tory hands. They will no doubt keep the Colwyn and Conwy seat created along the ‘Costa Geriatrica’ of the north Wales coastline. North Clwyd will probably lose the battle against Gwynedd.

Potentially one of the most interesting new seats is the Frankenstein's monster know as South Clwyd and North Montgomeryshire. While I'd fancy the Conservatives to keep it, it will include prominent Welsh-speaking areas such as Machynlleth and Y Bala, as well as some built-up areas near Wrexham.

The Liberal Democrats were also competitive in this area not so long ago and will be looking to bounce back after the disaster of 2015. 

This seat could become a four-way fight between the Tories, Labour, Plaid Cymru and the Liberal Democrats. I wouldn't like to put money of the outcome.

All in all, the Conservatives could lose half their seats. But given the Conservatives’ tendency to focus on the UK-wide picture rather than Wales in isolation, I’m sure it will be a sacrifice they’ll be willing to make.

Plaid Cymru

For Plaid Cymru the picture is mixed. It means that all of their seats are now swing seats, whereas before they had three pretty rock-solid strongholds. However, they could potentially up their tally from three to four seats.

Carmarthenshire should be safe, although Labour would run them close here.

Things are less clear in Afon and Anglesey. It includes more of the Labour-leaning Bangor than the Plaid-leaning Caernarfon and surrounding districts, which they lose to North Clwyd and Gwynedd.

This will probably be a very close contest between them and Labour – it could edge Labour.

North Clwyd on Gwynedd includes a number of tory-voting rural areas to the north-east. Plaid Cymru should win (knock Wood) but it won't be a safe seat with a massive majority in the same way as Dwyfor Meirionnydd.

Ceredigion could go one of two ways. This is now a very curious seat, with North Pembrokeshire included, which makes perfect sense, but also Llanidloes, which doesn’t.

Llanidloes is cut off from Ceredigion by very large and desolate mountains made famous by the series y Gwyll/Hinterland, and will feel relatively isolated from the rest of the constituency. It would have made much more sense in my opinion to include part of Machynlleth, which has better transport links with Aberystwyth.

How Plaid Cymru and the Lib Dems handle these changes will all be down to tactics. The Welsh speaking parts of North Pembrokeshire will no doubt help Plaid Cymru.

The Lib Dems will need to convince the people of Llanidoes and Fishguard who don’t vote Plaid Cymru to stop voting Tory/Labour and embrace them as the ‘anyone-but-the-nationalists’ party in these areas.

If everyone votes as they do at the moment, the seat will probably go Plaid Cymru. But if they vote tactically, as people tend to do in FPTP elections, the Lib Dems could be in with a shot.

The Result

What do I think the final tally would be?

Plaid Cymru – 3 seats (taking either Ceredigion or Anglesey and Arfon, but perhaps not both)

Conservatives – 6 seats. (Down from 11)

Labour – 19-20 seats. (down from 25)

Lib Dems – 1 or 0, depending on how things go in the new Ceredigion.

This is how I think it would play out, with swing seats in two colours:

The plans will now go out for a 12 week consultation. I will certainly be making a few suggestions!


  1. I think you're wrong about Gower. Now shorn of the old industrial communities linking Swansea with Llanelli (which join Llanelli) the peninsula itself is much more Tory in character, and the old Swansea West has been Conservative in my lifetime. So the new seat will indeed switch allegiances but it's more winnable now for the Conservatives. Though seeing as the bulk of the electorate will be on the west side of the city the constituency will have to be named Swansea West and Gower.

  2. Interesting point Jac. Labour's majority in the current Swansea West looks very healthy to me, while the Conservatives only scraped in by 50 votes in Gower, which made me think it could be a tough ask for the Tories. But you're much more knowledgeable about the area than I am!

    1. Byron Davies narrowly won the existing Gower seat last year, but the part of the constituency that it's planned to link with Swansea West is the rural area, and this is more fertile ground for the Tories, making the new constituency less safe for Labour than the existing Swansea West.

      Any measure to make students vote at their 'home' address (at all elections) would be another blow for Labour in this planned new constituency. Yet such legislation would be welcomed by communities where the indigenous population feels disenfranchised by a permanent but transient element that has no real commitment to or understanding of the area.

    2. Looking at the detailed map again I understand what you mean now. Good point. I think it's still fair to call it a swing seat but perhaps not as favorable to Labour as I had suggested. It's worth noting however that UK polling report's analysis has Labour winning the seat by a thin margin, based on 2015 figures. If Labour's fortunes continues to plummet it could be a v. different story of course...

  3. You're commenting on things as they stand without factoring in the rise of UKIP and the decline of Labour. But overall, quite accurate.

    1. That's all one can do really. I'm just looking at how things stand now - doing that and predicting future political trends a step too far. UKIP's vote may well crater depending on how Brexit goes.

  4. Diolch am y dadansoddiad hwn.

    Rwy'n cytuno gyda Royston ynglŷn â Gŵyr. Mae Gorseinon, Llangyfelach, Penllergaer a Phontarddulais wedi eu tynnu allan o'r sedd, tra bod y Sgeti, Dynfant, Uplands a Chilai wedi cael eu rhoi iddi. Mwyafrif o fil neu ddwy i'r Toriaid.

    Nawr, mae Ceredigion/Gogledd Penfro yn erthyl o sedd. Mae'n amlwg taw hon oedd y diwethaf i'w chreu. Ych: o Dwfftwn i Drefeglwys, gyda darnau o bedair sir a chroesi Pen Pumlumon Fawr ar y ffordd. Gobeithio na fydd yr AS newydd yn tarfu ar Glyndŵr yn gorffwys dan y mynydd!

    Y trafferth ydy, er bod lle yn etholaeth De Powys i dderbyn Blaenhafren a Llanidloes (3,852 pleidlais), fedr De Clwyd/Gogledd Maldwyn ddim fforddio colli ardal Machynlleth a Llanbrynmair i Ceredigion (3,927) heb ddwgyd Caersws a Rhiwcynon ar gyrrion y Drenewydd a Ffordun ar bwys y Trallwng (4,469).

    Bydd bobl Sir Drefaldwyn yn gweiddi allan pan ffordd bynnag aeth hi!

    1. Duw a Gŵyr, ynde! :P

      Credu bod Ceredigion yn gwneud rywfaint o synnwyr - wedi'r cwbwl, roedd y sedd yn cynnwys gogledd Penfro yn y gorffennol. De Clwyd a Gogledd Sir Drefaldwyn yw'r erthyl yfmi - popeth oedd ar ol wedi creu'r seddi eraill!

    2. Newydd edrych ar De Clwyd/Gog Maldwyn eto. Ydy, mae hi yn edrych yn hynod o flêr.

      Darnau o bum sir (Conwy, Dinbych, Gwynedd, Powys a Wrecsam) a phum etholaeth (De, Gorllewin a Dyffryn Clwyd, Dwyfor Meirionnydd a Maldwyn).

      Oes ateb arall, tybed? Mae'r etholaeth newydd yn agos iawn at yr isafswm etholwyr nawr, gyda dim ond 71,097 o etholwyr. Fedran nhw ddim tynnu neb allan, dim ond ychwanegu ardaloedd eraill i greu rhagor o lanast.

    3. Dwi'n credu bod yna ateb arall Emlyn, a rydych chi'n ddod o hyd i'r ateb yn ne Powys. Mae'n posib symud wardiau Cwm Tawe fel Ystradgynlais i mewn i Castell-nedd, sy'n rhyddhau lle i dros 7,000 o etholwyr yn Aberhonddu, Maesyfed a Trefaldwyn. Mae hyn yn golygu allech chi cadw'r Trallwng a pentrefi megis Cegidfa, Trewern a Llandysilio yn Mhowys. Bydd rhaid i chi newid y ffiniau o amgylch llefydd fel Dinbych, Caernarfon a Machynlleth hefyd i helpu De Clwyd a Gogledd Sir Drefaldwyn cyrraedd y cwota ond mae'n posib. Dwi wedi amlinellu'r newidiadau ar fy mlog(shameless plug!).

  5. According to estimates of notional votes from 2015 (, Plaid seem to have done well in terms of an efficient spread of votes to seats. Only Ceredigion a Gogledd Sir Benfro seems like an obvious target now though, unless Labour really collapse in South Wales (not impossible I suppose)

    1. Very interesting, and broadly in line with what I would have expected. Surprised PC aren't more competitive in South Clwyd and North Montgomeryshire however.

      It's difficult to tell how voters will react to the new seats though, as people often vote tactically under FPTP. I don't think you can simply map out the result in old seats on to new ones. I suspect that the months and years up to the election will be spent trying to convince the electorate that it's a two horse race between such and such a party, even if the previous figures don't suggest that.

    2. Sure, tactical voting is the big unknown here. My hunch is that this increased uncertainty will be bad for Plaid though. Plaid's targeting of seats was a bit ropey in the Assembly elections this year - didn't seem to realise how close they were in Blaenau Gwent while pouring resources into Llanelli etc. Mike Parker's book The Greasy Poll also suggested issues with canvass returns. I can imagine the Lib Dems pulling it out of the bag again in Ceredigion...

    3. I think Plaid made the right decision pouring resources into Llanelli, as it was a seat all the experts were saying they were on course to win. The problem was that Labour also thew the kitchen sink at it with the same aim. One problem in Wales is a lack of polling data, and even the one national poll we do get doesn't tell us what's going on in individual seats. The problem is Ceredigion imo was the election in 2010 rather than 2015 - the Lib Dems' majority did come down significantly, but they simply left themselves with too high a mountain to climb the last time. Plaid do have less resources than other parties, but their campaign in Ceredigion does seem to be very effective. As an example, I almost never plug in my landline phone at home. I did so on the eve of the Assembly elections this year as I was due to go on the radio to discuss the next day's events. Within 30 seconds of plugging it in, the phone rang, with a PC canvasser on the other end asking me whether I was going to vote for Elin Jones!

    4. I think in retrospect I wouldn't single Llanelli out as a mistake. Hindsight is 20/20 and I don't know the constituency at all. That said, I feel like Plaid don't tend to see an 'efficient' distribution of votes - they tend to pile up votes in safe seats and fall just short in target seats. In 2016 they had five results where the majority was under 2000 and Plaid were in contention (vote share of >25%): Aberconwy, Blaenau Gwent, Caerphilly, Cardiff West, and Llanelli. They lost all five. It's obviously a small sample and Plaid can't be blamed for having geographically concentrated support - that's just the human geography of the situation. It does make me think that there's a lot of room for improvement in their ground game.

      Maybe the proposed boundary changes will help them. They do look potentially more efficient for Plaid. I would worry that they'll need to improve their ground game to take advantage of that though.

  6. Just a small point - Ceredigion and North Pembs also includes some of Carmarthenshire (Llangeler and Cenarth Wards). Pont-Tyweli ar bwys Llandysul would be represented by two MPs as well as two community councils.

  7. Pont-Tyweli mor bwysig mae angen dau AS yno! Btw, Lesley, shouldn't that be Llandysul ar bwys Pont-Tyweli? After all, Llandysul only has one community council.


Post a Comment