The current 18 constituencies, ranked in order of electorate (using the figures for 1 July 2010) are:
59,612 East BelfastThat's a total of 1,185,426; divided by 15 that comes out close enough to 79,000 voters, give or take, in each new seat. The margin of variation is supposed to be 5% either way; if that's taken as a strict guideline for the whole UK, it's going to be a very tight constraint on drawing the Northern Irish seats, as the upper bound for any UK seat will thus be around 79,500. More likely, I'd have thought, that they will allow the pragmatic option of 5% variation for NI seats around the local NI average, which therefore means seat sizes between 75,000 and 83,000 will be acceptable, so I shall take that as a working assumption below.
60,256 South Belfast
60,287 West Belfast
61,063 East Antrim
61,181 Strangford
61,230 North Down
62,045 West Tyrone
63,972 South Antrim
64,139 East Londonderry
65,453 Mid Ulster
66,132 Lagan Valley
66,597 North Belfast
67,161 Foyle
68,926 Fermanagh and South Tyrone
71,937 South Down
74,181 North Antrim
75,481 Newry and Armagh
75,773 Upper Bann
The first strategic question for the boundary revision will be how many seats are needed to cover Belfast. Adding the four current seats together gives enough voters for 3.12 seats in the new dispensation, so one option will be to simply use the current boundaries, maybe trimmed a little, and cut the city to three seats rather than its historic four. When a previous boundary commission in the mid-1990s tried this, it was met with howls of anguish, and instead the four Belfast seats were padded with a few more voters to bring them up to the bare minimum, an approach also taken by the most recent boundary commission.
But to retain four seats this time, about 12,500 voters would need to be added to North Belfast and about 15,000 to the other three seats. The Belfast seats have already extended to include pretty much anything that can reasonably be defined as Greater Belfast, and a bit beyond. To push the boundaries out further could mean East Belfast absorbing Newtownards, thus gaining a shoreline on Strangford Lough; while West Belfast might similarly have to extend as far as Lough Neagh. I must say it's quite difficult to see this.
So let's consider the three seat option. This would leave the external parliamentary boundaries of Belfast largely unchanged, allowing more freedom for tinkering elsewhere. Given that Belfast Lough is a fairly unavoidable geographical feature, East Belfast would necessarily annex about a third of South Belfast, North Belfast taking about a fifth of West Belfast, and the remnant forming a new South-West Belfast constituency, which was what was proposed in the mid-1990s. I can't see this being too popular, as the burghers of the Malone Road wake up to the probability of Gerry Adams being their new MP.
There is another option. The whole of Belfast City Council - now preserved by the procrastination over local government reform - has only enough voters for 2.15 seats under the new dispensation. What price "North and West Belfast" and "East and South Belfast" as the two new constituencies, the hinterlands hived off to the suburbs? The former would certainly be a Sinn Fein seat, the DUP losing Nigel Dodds; the latter could be a close call between the DUP and Alliance, with the SDLP playing an interesting role.
Elsewhere it's easier to see how the cat might be skinned. Newry & Armagh and Upper Bann are both over 95% of the new quota anyway, so I'm inclined to think they won't be changed at all, having survived the last boundary commission unscathed, and if so they make a convenient anchor around which to construct the rest of my assumptions.
In the three western historic counties, there are currently five seats (Foyle, East Londonderry, West Tyrone, Mid Ulster, and Fermanagh-South Tyrone) whose combined electorate gives you 4.15 seats under the new dispensation. The two northern seats will expand southward, FST will annex some more of Tyrone, and the amalgamated remnants of West Tyrone and Mid Ulster will be the final seat in the west.
The three County Antrim constituencies between them have enough voters for 2.52 seats under the proposed new system. Again, a lot will depend on what is done strategically with Belfast, but it seems inevitable that North Antrim will reabsorb the areas it lost to East Antrim in the most recent revision which brings it up to 78,000, close to the new avergae of 79,000. A South Antrim seat including almost all of Antrim, Larne and Carrickfergus would have an electorate of almost 84,000, which is just above the 83,000 upper limit (and one can easily see marginal adjustments which would bring it down). That leaves Newtownabbey, and whatever happens to the Belfast seats. One can imagine (I think that someone actually proposed) a massive shift of North Belfast to take up most if not all of Newtownabbey, shedding the rest to West Belfast.
South Down probably regains some of the territory around Strangford Lough that it lost in the last revision, but the other County Down seats depend on what strategic decision is made in Belfast. Strangford as a seat has been dreadfully messed with at every boundary revision since it was created, and this time it will probably disappear entirely between a much bigger North Down, an expanded Lagan Valley, and whatever is done with the Belfast seats.
It is tricky to call winners and losers, but simply because the DUP and SF have the most seats, they are also most likely to lose overall - the DUP probably will lose two in the east, and SF one in the west. As I said just after the election, I don't think AV would deliver much different results to FPTP, if the DUP and SF continue to maintain their dominance on their respective sides of the community. Of course, it does meant that the tipping points are in different places, and the fact that tactical voting is unnecessary with AV may also change voting patterns.
The proposed new division, if it is as equal as we are promised, will at least take care of one of my persistent criticisms of the current electoral set-up, which is that Assembly seats should be better mapped to electorate - East Belfast, South Belfast and West Belfast do not deserve a sixth Assembly seat on current figures; South Down, North Antrim, Newry & Armagh and Upper Bann all deserve a seventh. But if the variation of 5% is adhered to as a rigid limit, that problem at least will be solved, even if many more are created.
Comments
Why not? The Isle of Wight probably will. And a seat that crossed over the Irish Sea to make up the numbers is no less logical than getting Gibraltarians to vote in England South West
I can't see this being too popular, as the burghers of the Malone Road wake up to the probability of Gerry Adams being their new MP.
They're far from the only ones with whom such a change would be unpopular! I can't imagine too many people in South Belfast, Malone Road moneybags or not, would be happy about that. SF only got 9% of the South Belfast vote in the 2005 election, after all.
I also drew boundaries for 15 seats on the basis of a 2 seat Belfast and the outflux of wards from South and East Belfast in particular had big knock on effects. North Down had to gain the Ards peninsula, Castlereagh and Newtownards, first mooted in 1995, became a reality. Mid Down (Down and Banbridge councils), "Newry, Mourne and South Armagh" (Newry and Mourne, 9 Armagh wards) and "Armagh City and Upper Bann" all came into existence for the first time.
The west wasn't too bad, Fermanagh and South Tyrone returned to its pre-1996 boundaries, North Tyrone (or Mid Ulster) all of Omagh, Strabane and Cookstown. Foyle gained some 6 or 7 rural wards from Limavady with the rest joining Coleraine and Magherafelt in the revised East L'derry seat. The outflux of Newtownabbey wards from North Belfast took care of the Antrim constituencies.
It's impossible to judge, but given Clegg made it clear "equal size will take precedent over all other considerations (apart from his three exceptions)" (that's Wells, not got text of Clegg open), it's possible they might end up doing something daft.
Personally, I'd argue that an area that has internal devolution (NI, Scotland, London, Wales) should be, if there's conflict, under represented if there's a choice, so I'm inclined to guess 15, but it's possible they might not want to annoy the DUP too much.
Possible amendement to give NI as a specific special case STV given 'special & historic circumstances'?
(there are many ways I reckon we could sell some Tories on some sort of implementation of STV/BPS, and any area having it, even as a 'trial' or 'special case' makes it easier in the future)
Take it Assembly seats will remain mapped to Westminster consituency is a given? They no longer do for Scotland.
Doesn't seem right to me. UK electorate as a whole in May was 45.61 million, NI was 1.17 million. For the UK, take off Western Isles (15k) and Orkney+Shetland (19k), gives you 45.57 million, divide by 598 gets you a quota of 76,200-ish.
NI quota for 15 seats is almost bang on 78,000, so 2,800 above UK quota. For 16 seats it is 73,100-ish, so 3,100 below, and already at 95.9% rather than 102.3%.
That's using 7 May figures, of course; the figures in my original post are for 1 July for NI as a whole, suggesting that there's been a 1.3% jump in the NI electorate in the last seven weeks... I don't have 1 July figures for the UK as a whole.
they might not want to annoy the DUP too much
Well, if the DUP want to be treated as part of the UK, they will just have to lump it.
Possible amendement to give NI as a specific special case STV given 'special & historic circumstances'?
I'd love to see the case for STV as a special case but difficult to see why today's circs are any more specific or historical than those of the last four boundary reviews.
Take it Assembly seats will remain mapped to Westminster consituency is a given?
The NI Boundary Commission used to have the duty of recommending the number of Assembly seats per constituency, even when there was no Assembly (the recommendation from 1983 boundaries was that there should be five each, never of course implemented). Since 1998 the doctrine has been that six seats per constituency is hardwired into the Agreement, even though this was a last-minute Good Friday concession to a party which doesn't exist any more. I have problems with this, as outlined above, but the new proposals deal with it.
You said 'One can imagine (I think that someone actually proposed) a massive shift of North Belfast to take up most if not all of Newtownabbey, shedding the rest to West Belfast.' Im wondering,is this scenario you envisage part of a 3 seat belfast model and also who do you think would win such a constituency: the DUP or sinn fein?
Conal.
The 2 seat Belfast option would require unnecessary changes, and is therefore less likely in my opinion. Also, the complaints of Malone Road residents or any other residents will be less relevant from now on because the incoming boundary legislation will do away with local public inquiry meetings. Instead there will only be a 3 month written submission period, so that the BC can swiftly complete the job at hand. Would you say the 3 seat or the 2 seat option is more likely? On most recent voting evidence, and taking the 3 seat Belfast option, we would have an election result of 2 SF seats and 1 Alliance/DUP seat.
I also agree that the 5 current Western seats (counties Tyrone, Derry and Fermanagh) will be reduced to 4 new seats (Fermanagh-South Tyrone, North Tyrone, Foyle, East Londonderry). Again, no large scale changes are required and the external boundary encompassing the 4 new seats can remain the same as the old 5-seat external boundary. The electorate of 328,000, when divided by 4 seats produces 82,000 electors per constituency. This will leave the West with 2 SF seats, 1 SDLP seat, and 1 DUP seat. However, the new DUP seat of East Londonderry would become a marginal seat because 16,000 new electors from the Magherafelt DEAs of Sperrin and Moyola would be added. This would leave East Londonderry with a total unionist electorate of only 52%, and this majority would be eroded over time due to demographic changes.
Newry-Armagh and Upper Bann are within the quota as they stand, so I see little point in changing them. So, they would return 1 SF and 1 DUP seat.
That leaves the remaining parts of counties Antrim and Down which are currently represented by 7 seats. This electorate of 459,000 will be reduced to 6 seats with an average quota of 76,500. I expect there will be an expanded North Antrim and an expanded East Antrim. North Antrim and East Antrim would both take territory from South Antrim. The remainder of South Antrim would be merged with the northern half of the current constituency of Lagan Valley. Lagan Valley would be the constituency to lose out, with the other half of Lagan Valley being added to the three current Down constituencies (North Down, Strangford, South Down) to form 3 new expanded constituencies. North Down will expand to take the Ards peninsula. The remainder of Strangford will be added to the remainder of Lagan Valley to form Mid Down. South Down will also have some minor additions. These boundary changes appear relatively obvious choices to me. We could hypothesise that this would result in 5 DUP seats and 1 SDLP seat (although Sylvia Hermon would reduce the DUP tally by 1 if she runs in North Down again in 2015).
This would leave us with an overall tally of 8 seats for the DUP (or 6 seats if Long and Hermon repeat their 2010 performances), and 7 seats for nationalism (SF with 5, SDLP with 2). Overall, the changes would result in the DUP losing 2 seats (North Belfast and Lagan Valley), and the SDLP losing 1 seat (South Belfast). SF’s loss of Mid Ulster would be offset by their gain of North-West Belfast. East Londonderry would become a major electoral battleground, as it would represent the target seat for nationalism to win in order for it to achieve a majority of Westminster seats.
Because by the next westminster election, Gregory Campbell will have been this constituency's incumbent MP for 15 years and I think you'll find that the nationalist vote in East Londonderry significantly underperforms its potential strength,especially in coleraine and portrush/portstewart.
A few republican rural areas of Magherafelt council bolted on cannot overcome a long incumbent MP in a seat were the differential turnout between protestant and catholic voters is significant.Anyhow,the DUP are strong in this seat,for example,they have 3 MLA's to SF's 1.