What if #AE17b happens ?

It’s been a tumultuous week, to say the least, at the end of what had been a very boring election, with a number of historic and fascinating political developments going on. There’s been plenty of analysis of the wider political configuration on Slugger for the past while, so I thought I’d have a look at what might happen closer to home, specifically if there is another Assembly election.

If the March 2017 Assembly election was about the nationalist surge, the June 2017 general election in Northern Ireland was about the unionist surge in response. It shouldn’t be underestimated how strong the surge was, with Unionist MPs romping home with the kind of results seen no more than once or twice during the past 40 years in places like East Antrim, East Belfast and even the Paisley heartland of North Antrim. It should be lost on nobody that the previous occasions within the past five decades when Unionists won these kinds of results were associated with the UWC Strike and the Anglo Irish Agreement.

Between these two elections, the combined Unionist vote added 4.4 percentage points,  from to 44.8% to 49.2%. In terms of total votes it went from 358,818 to 398,921, an extra 40,103 Unionist voters turning out, every one of them for the DUP. The DUP vote increased beyond even this, a total of 66,903, belying what everyone already knows – suggesting that above the increased turnout, the DUP attracted around 27,000 voters away from other Unionist parties.

Sinn Féin sustained their results and drew out an additional 14,670 voters to bank 238,915 votes, which was more votes than the DUP had in March, and would have made SF the largest party had the DUP not been able to successfully rally Unionists to its cause.

The other side of the coin is a number of serious defeats inflicted upon the UUP and SDLP. The UUP lost just under 20,000 votes, which clearly must put a number of assembly seats in jeopardy. The SDLP lost only 539 votes, its three seats disappearing under the sheer weight of increased Sinn Féin and DUP turnout. The PUP, UKIP and TUV mostly stayed out of this election, substantially to the benefit of the DUP.

So, if there were an assembly election tomorrow, what would happen ?

There’s obviously a huge risk in triggering an election over the issue of voter fatigue.

However, looking solely at numbers, there is a clear opportunity for the DUP to run another campaign urging supporters to complete the circle by reinstating the DUP and Unionism to the dominant position it lost in March, and returning control over the Petition of Concern back to the DUP’s hands.

It’s almost a perfect storm for the DUP. While on one hand a number of Unionist seats were lost in March due to nothing other than poor Unionist transfer discipline, on the other hand there are no obvious opportunities for Sinn Féin to make additional gains. Moreover, SF would have trouble resisting calls for another election, having urged adherence to the law which, strictly speaking, requires a further election following the failure to appoint an Executive back in March.

The UUP and SDLP, reeling from the loss of five Westminster seats (and the attendant financial support) between them would be poorly equipped to mount a fightback at this early stage and are both vulnerable in a number of constituencies.

Ordinarily, a British government might try to account for the possible cost of squashing the smaller, centrist parties; but with the DUP pulling the strings in the Commons all bets are off for this kind of calculation.

Below we’ll take a quick look at the seats, and I’ll make my prediction of the Assembly makeup and the possible d’Hondt Executive configuration that could follow. The usual health warning applies; these are rough estimates based on a back-of-an-envelope attempt to tie together polls conducted under two very different electoral systems – anything could happen between now and the next Assembly poll. I’ve not done a detailed analysis of small changes effecting earlier stages of each count – but there are a number of scenarios where small shifts in opinion could significantly effect the outcome in a constituency.

Belfast North

Both Finucane and Dodds polled remarkable results here. Nearly 4000 extra voters came out over and above the Assembly election, and it appears that almost all of them went for the DUP and SF. For both parties, their two assembly seats here are completely safe, but the SDLP seat now looks wobbly.

That said, traditionally, the SDLP has always polled in North Belfast well behind its Assembly result and Nichola Mallon has successfully inherited Alban Maginness’ personal vote, seeing off a reinvigorated Alliance challenge. In the event of extra DUP and SF candidates, the result would end up being decided by Alliance transfers which should secure the seat for Nichola.

Prediction 2 SF 2 DUP 1 SDLP (no change)

Belfast East

The story of the election here was one of Gavin Robinson consolidating the DUP’s position, adding 4000 votes from his previous total. Alliance held onto enough votes to keep their two seats safe. PUP voters got Andy Allen elected in March, but in the general election they clearly rallied behind Robinson. If they repeat this pattern, the DUP will take the third seat from the UUP.
Prediction : 3 DUP 2 All (DUP +1 UUP -1)

Belfast South

The Sinn Féin vote here appears to have reached peak in the March election, with several hundred voters peeling off this time in an effort to keep Pengelly out by voting for McDonnell. The DUP carefully calibrated their campaign here, keeping it low key, aiming to squeeze between the warring centrist factions.

In March, the bleary-eyed but valiant Greens stayed #AwakeforBailey, with their candidate just about nicking it when UUP voters failed to uniformly transfer to the DUP. However, if the pro-DUP mood among Unionists prevails from the Westminster election, more UUP voters could be persuaded to the DUP, and the likelihood of transfer leakage should be lower.

(this is my least favourite prediction  …. )
Prediction : 2 DUP 1 SF 1 SDLP 1 All (Green -1, DUP +1)

Belfast West

This is an unusually interesting one, for two reasons. PBP clearly lost ground with respect to the Assembly election here. In an ultra-safe seat that Sinn Féin literally cannot lose, this is not good news for Carroll. Nonetheless, those who did not vote for either SF or the DUP would be expected to mostly transfer to Carroll which should see him elected, although possibly under quota.

Meanwhile, it was really interesting to see the DUP’s Frank McCoubrey poll 5455 votes, 500 votes behind the last Sinn Féin seat in the Assembly election, a point which the DUP will take care to press home on the Shankill Road. However, I still don’t think they’ll quite make it, so:
Prediction : 4 SF 1 PBP  (no change)

North Down

In the general election the only change here was a significant swing from Sylvia Hermon to the DUP. This implies that there is still one clear seat for the UUP, which should keep Alan Chambers safe. Overall, it’s “as you were” in North Down.

Prediction : 2 DUP 1 UUP 1 All 1 Green (no change)

South Down

Normally, Margaret Ritchie would poll strongly on tactical votes. She still polled strongly, but SF’s turnout – which leapt ahead even from the tally in March – ensured that this was not enough, with Hazzard claiming nearly 40% of the vote.

Tactical voting for both SF and the SDLP here overestimate their overall levels of support. In the Assembly election, Alliance’s Patrick Brown was 500 votes behind McGrath which endangers this seat, while SF got two seats over quota with a surplus of 2615 votes. However hardly any of these transferred – which means that the SDLP cannot hope to get the trickle-down of any further SF surge.

I’d quite fairly stand accused of bias if I predicted an Alliance victory (although I’ll say that Brown’s chances are probably better than Frank McCoubrey’s chances in West Belfast, with a roughly similar gap to close), but the SDLP will need to be careful not to allow the loss of the Westminster seat to effect their morale if they wish to retain two assembly seats here.

Prediction : 2 SF 2 SDLP 1 DUP (no change)

Lagan Valley

The DUP romped home here, one of their safest Westminster seats, with the equivalent of what would in an assembly election be well over three quotas. In March, this seat caused a shock when Jenny Palmer’s transfers favoured Pat Catney over the third DUP candidate, Brenda Hale. I suspect, though, that UUP voters are in no mood to shore up the SDLP, which will deprive Catney of the seat. Catney’s transfers on exclusion should, in turn, secure the Alliance seat.

Prediction : 3 DUP 1 UUP 1 All (SDLP -1, DUP + 1)

East Antrim

East Antrim was a bit of an anomaly in the Assembly election, with the UUP surprisingly scoring two seats on the back of a collapse in the UKIP vote (in turn, UKIP had narrowly missed out on the last seat in 2016 – once again due to the failure of UUP voters to transfer to UKIP in sufficient numbers, handing the seat to Sinn Féin).

There was no such ambiguity in this election, with Sammy Wilson reaching almost 57%, the best result of any MP in this seat since the Anglo Irish Agreement fallout period in 1987. If this is sustained, as seems likely, the DUP have three easy seats.

Prediction :

3 DUP 1 UUP 1 All (UUP -1, DUP +1)

South Antrim

Back when the election was called, I opined here on Slugger that this seat might be the subject of a DUP-UUP pact, with the UUP agreeing to stand aside in South Belfast in exchange for a free run to hold their seat here. The DUP had other ideas, correctly anticipating that a return in support to their (locally well known and well regarded) candidate would seal the deal without any pact. I doubt we’ll see any change here.

Prediction : 2 DUP 1 UUP 1 All 1 SF

North Antrim

In the Assembly election the UUP pulled in quite convincingly from the start of the count with Swann getting elected ahead of both the DUP and Jim Allister. However, in Westminster, Ian Paisley romped home with nearly 60% of the vote, a result not obtained even by his late father since the Anglo Irish Agreement period in 1987, and before that since 1974. If that vote holds up, the DUP could return with three seats, however Jim Allister’s cachet remains strong and Robin Swann polled convincingly last time.

I’m going to go out on a limb and predict that North Antrim will want the Petition of Concern reinstated so that the DUP can block gay marriage, causing a DUP swing against the UUP here which knocks out Swann.

Prediction : 3 DUP 1 SF 1 TUV (UUP -1, DUP +1)

East Derry

Not so much dramatic change here. The swing away from the UUP will probably not hurt the independent Clare Sugden. Sinn Féin polled very roughly the same number of votes as they did in the assembly, with the SDLP slightly ahead, so I suspect there will be no change.

Prediction : 2 DUP 1 Ind 1 SF 1 SDLP (no change)


A heartland seat for the SDLP, it provided the shocker of the night when it switched from the SDLP to SF, with SF pulling in 6,577 votes over the result obtained in the Assembly election in March. However, the two parties remain evenly balanced in the seat, so it is unlikely any seats will change hands in the Assembly election. Absent a well-known local figure, SF might judge it unwise to attempt to run three candidates here.

Prediction : 2 SDLP 2 SF 1 DUP

West Tyrone

Here’s a seat where Sinn Féin could actually be in a bit of trouble. Barry McElduff attracted 739 new Sinn Féin votes over the assembly election; but Thomas Buchanan added 2654 votes, about 1401 of which came from the UUP’s Alicia Clarke. Numerically, there is scope for a second Unionist candidate to squeeze past the third Sinn Féin incumbent, but this would only happen if there was no transfer leakage from Buchanan to Clarke.

I’m predicting that there will be leakage in the DUP->UUP direction, hence:
Prediction : 3 SF, 1 DUP, 1 SDLP

Mid Ulster

Interestingly, Sinn Féin lost 752 voters here between the Assembly election and the General –  where the DUP gained 2997, in the face of an overall drop in 4159 votes in the turnout. In March, only 1190 votes need to transfer to the UUP candidate to beat the SDLP’s Patsy McGlone. The SDLP are set to lose this seat unless they come out fighting hard.

Prediction : 3 SF 1 DUP 1 UUP (SDLP -1, UUP +1)

Fermanagh and South Tyrone

This seat was surprisingly interesting during the assembly election. At the second count, 67 seats separated the SDLP from the third Sinn Féin candidate. The elimination of the third SF candidate could secure the seat for the SDLP, but only if SF voters transfer. The failure of SF voters to transfer here could ironically return the fifth seat to the DUP.

The gap with the UUP was wider but still only a matter of 587 votes ahead of the SDLP. A reversal here with the elimination of the UUP would have given the DUP a second seat.

In addition, the Unionist side of the house is harder to call due to the Unionist pact. Since the seat is already dominated by the DUP and SF, the safe but unadventurous bet is that it would remain unchanged provided that the UUP vote holds, and that SDLP transfers upon elimination continue to support the UUP. The UUP seat would probably be safest if Elliott stood for the Assembly; otherwise there is an outside chance for a DUP upset here.

Prediction : 3 SF 1 UUP 1 DUP (no change)

Newry and Armagh

No big surprises here in the Westminster election. Without a pact, the Unionist vote share was very slightly higher, but with four nationalists elected with only one of them (slightly) below quota, there’s not enough to offer the UUP a convincing chance of regaining their seat here.
Prediction : 3 SF 1 SDLP 1 DUP (no change)

Upper Bann

The SDLP win here came as a little bit of a surprise to some commentators. In the Westminster election, there was a swing from the UUP to the DUP which would probably mostly swing back to the UUP in an Assembly election, safely returning Doug Beattie.

Sinn Féin’s John O’Dowd banked 14,325 votes here, against 14,328 in the Assembly election, so interestingly SF actually lost three votes here. They’ll need 2712 votes to get ahead of Dolores, so outside of either a further spike in SF’s vote or a spectacular collapse in the SDLP vote, SF are unlikely to regain their second seat.

Prediction : 2 DUP 1 UUP 1 SF 1 SDLP (no change)


Another 30-year high for the DUP, with Jim Shannon landing 60%, the highest result again since 1987. The nationalist vote is basically nowhere in this constituency, unaffected by the nationalist surge, and there aren’t enough Unionist votes for the DUP to to take the risk of running four candidates.

Prediction : 3 DUP 1 All 1 UUP

Overall prediction and Executive makeup

If there is an Assembly election in the short term, I believe we will see

  • DUP 33 (+5)
  • SF 27 (no change)
  • SDLP 10 (-2)
  • Alliance 8 (no change)
  • UUP 8 (-2)
  • PBP 1 (no change)
  • Green 1 (-1)
  • TUV 1 (no change)
  • Independent Clare Sugden 1 (no change)

The clear story is a consolidation of the DUP vote at the expense mainly of the UUP, with the restoration of DUP control over the Petition of Concern and clear spacing of the DUP as the largest party. The Green and SDLP seats lost to Unionism via transfer leakage will almost certainly returned provided the mood of the Unionist electorate carries over into the Assembly election.

This would yield an Executive of 3 DUP, 2SF, 1 SDLP and 1 UUP. The UUP and Alliance are tied in terms of seats; the tie is decided by the number of first preference votes, which would probably give the seat to the UUP. If Alliance take the Justice Ministry, the next seat would go to the UUP.

Correction 15/06/17

In the original version of the article I’d counted an SDLP seat in Lagan Valley instead of a UUP seat. The UUP would be level with Alliance and would be line for the final Executive seat.