Two big Wedges now define Northern Ireland politics

I was fortunate enough to appear on UTV’s The Issue programme last Thursday – initially to talk about elections but then racism got in the way. Here is what so would have said about the former…

As ever, the old media missed it, but the story of the recent Northern Ireland elections was the rise of hard-line Unionism – at local level, TUV/UKIP/PUP polled over 50,000 first preference votes (the same as Alliance and NI21 combined); at European level, they polled nearly 100,000 (nearly double Alliance plus NI21). For all the wild coverage of the new progressive centre ground, what the disenchanted Unionist non-voter was looking for was a harder line, not a softer one.

For all that, this may be a one-off. At a Westminister election where only X-votes count, the temptation will be for those voters to return to the DUP or the joint Unionist candidate; at a Stormont election where the largest party gets the First Minister, the temptation will be to go with the largest Unionist party (the DUP) to make sure it comes out on top.

What will not be a one-off, I suspect, are the two clear “wedges” now identifiable among the Northern Irish electorate.

The first is between Progressives (my term) and Unionists. In the past, the Alliance Party has consistently scored much lower at “higher-level” elections (such as European) than “lower-level” (such as Council), but this year that was turned on its head when Anna Lo’s 7.1% beat the party’s local election score of 6.7%. Alliance and other Progressive voters opted not to lend their vote to the UUP, as some on the more pro-Union end of that spectrum had in the past, thus ensuring by far the worst ever Ulster Unionist European Election result (over three points worse than ever before) and only 20% of their transfers (where in the past it would have been 40-50%). The ire worked both ways, however. In heavily Unionist towns such as Carrickfergus and Ballyclare the Alliance Party came close to wipe-out at local level as Unionists turned out en masse and transferred heavily to other Unionist candidates while refusing resolutely to transfer to Alliance – and at European level, Jim Nicholson’s low transfer from Alliance was compensated for by a high transfer from UKIP and TUV (which was even more useful given the high Unionist turnout).

The second is between the two Nationalist parties. At the European election, more Sinn Fein voters plumped (i.e. voted 1-Anderson only) than gave the SDLP a second preference. Even of those (barely half) who did transfer, a significant minority went elsewhere. At local council level, this was on occasions even more marked – in Balmoral, fully 20% of Lord Mayor Mairtin O Muilleoir’s surplus went to the Alliance Party candidates. It has long been the case that SDLP transfers tend to favour Alliance over Sinn Fein, but if anything this trend also continued.

Who gains from the first wedge is unclear. Unionists will be delighted by their high turnout, and by exceeding half the vote at the European Election (which they didn’t last time). However, demographics mean that relying solely on those who identify as “Unionist” will soon be a minority pursuit. The Alliance Party will be delighted that their “core vote” in fact appears to be 7% (only a decade on from polling half of that), yet will be concerned that their gradual nibbling of the “soft Unionist” vote seems to have stalled absolutely.

The second wedge has an obvious loser – the SDLP. It now cannot rely on transfers from Sinn Fein at all; on the other side, it now faces a serious challenge from an Alliance Party which has been (deliberately or otherwise) more open about what would be perceived by many as a “Nationalist” tinge. This is something of a trap – to rise to the Alliance challenge, the SDLP instinctively will feel it has to be more overtly Nationalist (or to point to Alliance’s apparent “unionism”); yet this actually puts off its own softer voters, particularly when confronted with an obviously likeable Alliance candidate such as Anna Lo.

The addition of the loss of control of Derry’s Guildhall (in fact made inevitable by the new Council boundaries) will be a huge psychological blow to the party of Hume. In the end, it is not at all clear who the winner of the election in Northern Ireland was – but the loser was clearly the SDLP.


4 thoughts on “Two big Wedges now define Northern Ireland politics

  1. Michael Hehir says:

    When there has been more time to properly analyse the results the conclusions may be more nuanced.

    For example you say, “The Alliance Party …… will be concerned that their gradual nibbling of the “soft Unionist” vote seems to have stalled absolutely.”

    The transfer data has now been published for all of the DEA’s where NI21 stood, apart from Mid & East Antrim. It is clear that around half of NI21 voters transferred to Non Designating parties (Alliance and Green). Like Alliance and Green the remainder of those that transfer do so in significant proportions to both unionist and nationalist candidates. Unlike Alliance and Green, which have historically transferred roughly evenly – NI21 voters transfer three or four times more to unionists than to nationalists.

    In other words the NI21 voters look very similar to voters on the ‘soft unionist’ end of the Alliance spectrum. So it may prove that those “soft unionist” voters are still there – but on this occasion NI21 nibbled at them. I don’t believe NI21 can recover – so whether the Alliance nibbling at the soft unionist vote is really “stalled absolutely” or can be resumed is still open to future developments.

    You also say, “In heavily Unionist towns such as Carrickfergus and Ballyclare the Alliance Party came close to wipe-out at local level as Unionists turned out en masse and transferred heavily to other Unionist candidates while refusing resolutely to transfer to Alliance.”

    Alliance does have issues in North Antrim – but this analysis is not the whole story.

    In the Ballyclare DEA the party fell from 15.2% in 2011 to 9.9% in 2014. That’s 5.3 percentage points, of which transfers suggest 2.4 points went to NI21 and 0.5 between the two Independents – all of which were transferred back.

    But the Unionists did not turn out en masse. In fact total turnout fell from 48.7% to 45.8%. There were fewer transfers from unionists – but this was not due to their refusal to transfer to Alliance. In 2011 Alliance received 4.1% of the 2221 transfers made from DUP and UUP candidates. This year it received 4.1% of the 343 transfers from DUP and PUP candidates (there were no UUP votes transferred). Alliance suffered as a result of a fall in unionist turnout – not an increase.

    In fact Alliance appears to have lost 2.4 percentage points due to its own vote not turning out. Alliance needs to ask itself why those 135 people failed to show. But with the fall in unionist turnout that would still have left it 74 votes short of holding the seat.

    I’ll look at the transfer situation in the two Carrick DEA transfers when the data is published. In the meantime some other observations.

    • The total percentage turnout in the Carrick DEA’s was the same as 2011.

    • Both DEA’s had massive boundary changes, with three old DEA’s being mixed together and then divided in two. The UUP ran two candidates – one with 13 years council experience, the second with 3. Both were re-elected and the party increased its share of the vote. Three of the DUP candidates were long-serving councilors who were all re-elected. No other party ran a candidate who was elected to the council in 2011.

    • It would appear that Alliance lost the large personal votes that Sean Neeson and Stewart Dickson had built up since they were first elected in 1977. The loss of these personal votes would have dropped Alliance to a core vote of about 21%. Instead they scored 15.3%.

    • If this is to be explained by differential turnout alone it would suggests that, using 2011 voting as a base, unionist and independents (who in 2011 transferred more as unionists than anything else) increased their turnout by about 4 percentage points, while Alliance dropped by about 13 percentage points. Alternatively significant numbers of core Alliance voters switched to unionist. This seems to be at variance with performance in many other places.

    Preliminary conclusions:
    • Maybe the party under-estimated the time it would take to build name-recognition and will need to plan future co-options giving the new name a much longer period before facing an election?
    • Maybe the party under-estimated the potential loss of personal votes and may have performed better with one relatively unknown candidate in each DEA, rather than two?
    • The problem or problems which have so depressed Alliance turnout and/or caused previous Alliance voters to switch are not attributable to unionist turnout or to unionist transfer behaviour.
    • It/they look likely to be either specific to Carrick, or to have been overcome elsewhere. Either way Alliance needs to identify it/them.
    • It may be premature to draw wider conclusions from Carrick.

    All that said I agree that the most significant outcome for Alliance in this election was its escape from the trap of the “squeezed” Euro vote. It’s still very early days, but the party now has a solid platform to enter the next Euros with a realistic prospect of taking a seat.

    • Michael,

      I’ve only had time to skim your points but look forward to coming back to them.

      I can probably declare that I have made very similar points in my own private analysis sent to one of the party’s associations. For example, I am relatively unconcerned by the relative loss of first-preference Alliance votes in Lisburn (NI21’s backyard) precisely because they came back and, as you say, are likely to do so directly now with NI21 finished.

      And yes, the point I was trying to make was that places like Carrick are different from, say, Lisburn because there are only Unionist transfers in play. That has always been a problem.

      One thing I would say is that turnout in Unionist areas was *comparatively* much higher than in 2011 or 2009 (the two comparators) – you can read that either as Unionists being more determined to vote or Nationalists less so.

  2. Johng976 says:

    Wow, wonderful blog layout! How long have you been blogging for? you made blogging look easy. The overall look of your web site is excellent, as well as the content! aegdkfaaddeb

  3. Michael Hehir says:


    Have a great holiday.

    I have now examined the transfers in the two Carrick DEA’s.

    As you know the overwhelming majority of voters for unionist candidates restrict their transfers to other unionist candidates until there are no more unionist candidates left. (The same is true of voters for nationalist candidates.) This situation did not occur in any Carrick DEA in either 2011 or 2014.

    Only a very small number transfer to Alliance or Green from the DUP if there is another DUP candidate still available. In 2011 on average about 1.6% of 1st preference DUP voters did so in all the NI DEA’s where they still had an Alliance or Green candidate available. For UUP voters it was 3.1%. These probably represent personal votes. In the Carrick DEA’s transfers to Alliance were above the average in 2011, but dropped back in 2014. This was only to be expected. As I mentioned in my previous response, Alliance were at a significant disadvantage this year running new candidates against well-established DUP and UUP councillors in radically altered DEA’s.

    When they ran out of DUP candidates about 4.8% of DUP 1st preference voters transferred to an available Alliance or Green in NI. For the UUP the figure was about 5.3%.
    Again, Carrick was above the NI average in 2011, when an average of about 12.6% of DUP and unionist-leaning Independents’ 1st preference voters transferred to (mainly) Alliance or Green. In 2014 the average for UUP, unionist-leaning Independents, TUV, PUP, UKIP and Conservative was 7.3%. So as a matter of statistical fact the Alliance runner up in Carrick Castle would have taken a seat if Alliance had received the additional 40 or 50 transfers that the 2011 average would have suggested.

    However it would be unreasonable to expect to maintain the 2011 level. The high Carrickfergus 2011 average may have been a statistical freak. Its principal component was the DUP whose voters were faced with the choice of Alliance, the UUP or not transferring. There was probably as much antipathy for their main rival as admiration for Alliance in the choice some of them made for Alliance. Secondly the unionist voter who had run out of candidates of their preferred party had far more choices in 2014. The biggest component in the 2014 average was the UUP. When its Carrick Castle candidate was elected its voters still had the DUP, TUV, UKIP and a unionist-orientated Independent to choose from, as well as Alliance. In these circumstances a reversion to the NI average does not seem unreasonable.

    Does this represent unionists “refusing resolutely to transfer to Alliance”? Certainly not. Or – at least – no more so than usual.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: