? ?

Previous Entry | Next Entry

Northern Ireland's ghost constituencies

Despite the fact that the Liberal Democrats and Labour are likely to vote them down, the revised proposals of the Boundary Commission have now been published. The major differences with the previous proposals are that
  • Ballymena gets joined to Antrim town in South Antrim rather than to Larne and Carrickfergus, which stay with Newtownabbey in an enlarged East Antrim;
  • Carnlough stays in East Antrim rather than moving to North Antrim;
  • North Antrim gets renamed "Coleraine and North Antrim";
  • the six Omagh wards which were to be added to Fermanagh and South Tyrone stay with Omagh Town in the new Mid Tyrone;
  • the six Dungannon wards of the Torrent DEA get put back into Fermanagh and South Tyrone as they were until 1996;
  • Upper Braniel goes into South-East Belfast rather than Strangford.
There's a gratifying amount of agreement with my own submission, in that they adopted the first four of my six initial proposals (and two others from my subsequent list), even giving me named credit at 6.19 (page 22) for their proposed reconfiguration of the Antrim seats. I think the Commission is completely wrong on its restrictive reading of its own room for manœuvre (pages 5-7) but I can understand why it is convenient to take the approach that they have done.

I have crunched the numbers on the new proposals, give that they remain in play and there is therefore a vanishingly small chance that they will make it to the statute book. The figures below are based on the latest revised proposals; they also reflect a new methodology that I have developed which links the parties' performance at local elections more strongly to the distribution of their votes at Assembly and Westminster elections. This has given me what look like more credible figures. Of course, these numbers reflect only the result of past elections if the same votes had been cast on the new boundaries, and are not in any way a prediction of future voting behaviour, especially since these constituencies are not likely to be used.

North Belfast
w 201042.2%8.5%0.0%0.0%4.7%0.0%1.0%11.6%31.9%
a 201139.5%9.4%0.0%0.0%5.8%0.0%4.9%11.1%29.3%
lg 201138.2%8.4%0.0%1.7%6.9%0.0%5.5%11.7%27.6%

Same boundaries as in original proposal, but my new methodology concentrates support for DUP and SF, Alliance being the big loser. (My original figures looked a little off anyway.) DUP hold at Westminster, DUP hold three and SF two for the Assembly, last seat probably but not certainly SDLP.

South-East Belfast
w 201027.8%20.5%3.8%0.0%30.9%1.2%0.0%13.9%1.9%
a 201135.2%11.2%1.6%7.9%25.3%2.1%1.8%7.8%7.0%
lg 201132.4%12.9%1.5%6.1%27.5%2.7%2.1%8.6%6.1%

Westminster figure has Alliance still ahead; Assembly and local govt put DUP in front but give Alliance squeeze potential. For the Assembly, clearly two DUP and two Alliance, and one UUP; last seat should be SDLP if they get decent SF transfers. (NB Westminster figure skewed by SF withdrawal in South Belfast.)

South-West Belfast
w 20106.4%6.9%0.0%0.0%5.0%1.0%0.0%24.2%56.6%
a 20116.1%6.0%0.0%0.2%5.3%0.7%6.5%16.9%58.4%
lg 20116.2%4.2%0.0%0.0%4.8%0.9%7.4%18.1%58.5%

Unchanged from previous recommendations. Very safe SF seat at Westminster. For the Assembly I see four SF, one SDLP and one Unionist (probably DUP) with SDLP second candidate as runner-up. Various people tried to persuade me that the fourth SF seat is not definite, or that Alliance has a better chance than the Unionists, but I am not convinced.

Coleraine and North Antrim
w 201042.8%15.5%12.3%0.0%4.9%0.0%0.9%9.9%13.8%
a 201144.1%10.9%7.9%5.5%5.7%0.0%0.0%9.4%16.4%
lg 201138.1%16.3%5.8%0.2%3.7%0.0%13.1%9.5%13.3%

I can claim some credit for the name change here, and for the removal of Carnlough from the proposed boundaries. The latter of course makes it more solidly Unionist, safe DUP for Westminster and three DUP, one UUP and one SF for the Assembly. Last seat probably still goes to TUV, though a strong SDLP or (less likely) Alliance candidate in a good year might have a chance.

East Antrim
w 201046.6%27.8%6.1%0.0%11.0%0.0%0.0%4.2%4.4%
a 201148.3%18.7%4.4%1.8%16.5%1.8%0.0%3.1%5.3%
lg 201139.7%20.2%2.2%2.7%19.4%1.1%8.9%2.8%3.1%

Losing the Glens and extending further into the heart of County Antrim makes this a more firmly Unionist seat, safe for the DUP at Westminster. Three safe DUP for the Assembly and one each for UUP and Alliance, the last seat probably being between DUP and UUP (with UUP possibly having the edge).

Fermanagh and South Tyrone
w 20100.6%1.3%0.1%40.3%1.0%0.0%0.4%8.5%47.9%
a 201122.3%18.3%2.4%0.0%1.6%0.0%2.3%10.3%42.9%
lg 201119.3%21.7%1.3%0.0%0.6%0.1%8.2%13.2%35.6%

Bringing in the Dungannon wards rather than the originally proposed Omagh wards further consolidates SF's hold on the Westminster seat, and also opens the possibility that the SDLP might regain their seat not from the Shinners but from the DUP - the Unionist parties combined are a hair below, rather than above, three quotas and while the DUP are slightly ahead from the Assembly election, accidents can happen. However, from the Assembly election, DUP would keep two and UUP one. (NB joint Unionist candidate at last Westminster election.)

w 201014.9%3.9%0.0%0.0%0.6%0.0%7.2%41.8%31.5%
a 201121.6%0.7%0.0%0.0%0.9%0.0%10.8%32.7%33.4%
lg 201117.4%4.4%0.0%0.5%0.9%0.0%7.7%35.6%33.5%

The SDLP remain secure here for Westminster and the Assembly balance looks undisturbed at SDLP three, SF two and DUP one. But SF are not all that far behind, and the SDLP edge depends on Unionist transfers and tactical votes.

w 201023.6%11.5%7.4%0.0%1.7%0.0%0.0%16.2%39.5%
a 201125.8%8.3%4.9%1.8%1.6%0.0%1.7%16.1%39.7%
lg 201122.6%12.5%5.2%0.0%1.4%0.0%3.1%16.8%38.4%

Again my new methodology concentrates support for the DUP and SF, and the latter should easily win the Westminster seat. For the Assembly it looks like three SF, one SDLP, and two Unionists with the DUP having a fair chance of winning both seats. Possible third Unionist instead of third Shinner.

Lagan Valley
w 201048.8%20.9%8.4%0.0%11.2%0.0%0.0%5.7%5.0%
a 201152.0%19.8%2.8%0.0%12.4%1.6%0.0%6.9%4.4%
lg 201150.2%21.6%1.4%0.6%12.5%1.0%0.8%6.1%5.9%

Remains safe DUP for Westminster. For the Assembly, my new methodology pulls more Nationalist votes out of here and into the Poleglass/Twinbrook area which goes into South-West Belfast. But I still think the SDLP should scrape a seat here, along with Alliance, the UUP and three DUP. If not, it will be a fifth Unionist, probably from the DUP.

Mid Tyrone
w 201016.4%14.3%2.3%0.0%2.0%0.0%1.0%14.7%49.2%
a 201119.2%11.2%1.5%0.0%2.1%0.0%5.2%10.7%50.0%
lg 201116.2%14.8%1.7%0.0%0.9%0.0%8.7%13.6%44.2%

Remains safe for SF at Westminsster. The inclusion of Omagh rather than Dungannon wards puts the Unionists collectively safely over two quotas and the UUP safely within reach of the second of those seats. SF should win three and SDLP one, though SDLP look a bit shaky.

Newry and Armagh
w 201012.4%18.3%0.0%0.0%1.2%0.0%1.4%23.8%42.9%
a 201112.7%18.0%1.7%0.2%1.6%0.0%0.2%23.9%41.7%
lg 201111.5%18.1%0.6%0.0%0.0%0.0%5.9%23.2%40.6%

Almost no change from current arrangements and no change from first proposal. SF safe at Westminster, Assembly remains three SF, one each of SDLP, DUP and UUP.

North Down
w 20106.4%20.1%4.6%53.7%6.6%2.8%0.0%4.4%1.3%
a 201143.8%10.8%0.1%2.1%18.7%6.5%10.7%5.9%1.3%
lg 201138.3%13.9%0.4%3.4%18.3%5.6%14.7%5.3%0.0%

No change from provisional recommendations. Lady Hermon remains way ahead at Westminster; without her, DUP are much the biggest party. No change at Assembly level - 3 DUP, 1 UUP, 1 Alliance, 1 Green, with Greens the most vulnerable to a good independent or a well-balanced Alliance ticket.

South Antrim
w 201037.8%21.1%11.4%0.0%6.2%0.0%0.7%9.9%12.9%
a 201140.9%14.6%8.1%0.7%10.3%0.0%0.0%11.4%14.0%
lg 201138.2%18.4%6.4%0.3%8.4%0.0%3.9%12.3%12.1%

Major rearrangements here but the DUP remain dominant at Westminster, and should win three Assembly seats with the UUP and SF winning one each. The last is tough to call but I think Alliance is likely to overtake SDLP on spare Unionist transfers.

South Down
w 20109.0%8.2%3.4%0.0%1.3%2.1%0.0%47.6%28.4%
a 201112.7%11.4%0.1%5.4%2.1%2.6%0.0%35.2%30.5%
lg 20119.5%14.0%1.1%4.5%2.2%3.2%2.1%35.2%28.1%

No change from the first proposals, and no change from SDLP holding the Westminster seat; my new methodology makes things even tighter for the Unionists to keep two seats but my gut feeling is that they would do so, the SDLP and SF also winning two each.

w 201042.4%24.2%4.4%0.0%13.2%1.5%0.0%12.1%2.1%
a 201146.4%17.2%2.3%2.8%16.0%0.9%0.6%9.5%4.4%
lg 201143.3%16.8%1.8%1.0%17.8%2.0%4.2%9.5%3.6%

Still safe for DUP at Westminster. My new methodology gives more Nationalist votes coming in from Carryduff, so should be enough to secure an SDLP seat (NB Westminster vote skewed by SF not standing). Alliance and UUP win one each, and DUP three.

Upper Bann
w 201033.9%26.0%0.0%0.0%3.0%0.0%0.1%12.5%24.5%
a 201127.4%24.9%2.4%0.6%6.5%0.0%0.0%11.2%26.9%
lg 201129.5%24.0%2.4%0.3%3.8%0.0%3.4%12.4%24.1%

On paper the UUP's best chance at Westminster, though still some way behind DUP. Removing Aghagallon takes away potential for a third Nationalist Assembly seat; a well-balanced SF ticket could defeat the SDLP but their vote management here has not been a strong point. Otherwise two DUP and two UUP for the Assembly.

As before, I give the DUP 7 Westminster seats (down 1), SF 5, the SDLP 2 (down 1), Lady Hermon 1 and Alliance 1 (just).

At Assembly level, I put the DUP on about 34 seats (down 4, but up one from my previous calculations), SF on 25 (down 4, and two less than my previous calculations), the UUP on 14 (down 2), SDLP 14 (no change, rather than down one as in my previous calculations), Alliance 7 (down 1) and Greens and TUV keeping their single seats. The reasons for the changes are that my new methodology gives the SDLP rather than SF the Nationalist seat in Lagan Valley, and the rearrangement of the East Antrim/South Antrim borders gives only one nationalist seat between the two constituencies, the DUP effectively gaining what is now SF's seat in East Antrim.

This is all pretty theoretical of course, but today is a public holiday here!


( 2 comments — Leave a comment )
Nov. 1st, 2012 05:26 pm (UTC)
I love the way your analysis so influenced the final decision - and was acknowledged. Next time they should pay you to sit on the commission!
Nov. 2nd, 2012 05:41 am (UTC)
Good to see your name in the report and great analysis. I feel I let the side down by not crunching any numbers but the uncertainty turned me off. The Antrim reboot makes much more sense. Shame they didn't split Loughries though!

( 2 comments — Leave a comment )

Latest Month

April 2023


Page Summary


  • nwhyte
    25 Jan 2023, 13:24
    O tempora! O mores!
  • nwhyte
    24 Jan 2023, 10:34
    Hello! Your entry got to top-25 of the most popular entries in LiveJournal!
    Learn more about LiveJournal Ratings in FAQ.
  • nwhyte
    8 Dec 2022, 12:44
    UK mailboxes aren't waterproof?! That seems like an odd design.
  • nwhyte
    29 Oct 2022, 16:28
    Now I know that "psephologist" is a word.
  • nwhyte
    9 Sep 2022, 11:19
    That would make things less awkward.
Powered by LiveJournal.com
Designed by yoksel