Wednesday 4 February 2009

First thoughts on the European Parliament election

On Thursday 4 June Northern Ireland will vote for its three members of the European Parliament. Although the full list of candidates will not be known for certain until 7 May, the principal parties have already indicated who will stand, and for the minor parties it is frankly of little importance. The election is a game of musical chairs – where previously there were four players and three chairs, now there are five players and still only three chairs.

The players:
DUP – Diane Dodds
Sinn Féin – Bairbre de Brún
UUP/Conservative – James Nicholson
SDLP – Alban Maginness
TUV – Jim Allister

The also-rans:
Alliance Party – have not yet confirmed whether they will stand a candidate
Green Party – Steven Agnew

Issues will, of course, play little or no part in this election. Most voters do not know and do not care where the parties stand on 'Europe', or on the multitude of issues that their MEPs deal with. The election will be a simple headcount both within the two blocks, and between the two blocks.

There are four main areas of interest in the election:
1. the overall unionist-nationalist balance,
2. the break-down of the unionist vote,
3. the break-down of the nationalist vote,
4. the vote for 'other' candidates
These are looked at in more detail below, but first it is useful to put this election in its historical context.

The trend for the unionist vote since the first elections to the European Parliament in 1979 has been downward. From around 59% in 1979 the unionist vote has dropped steadily to the 49% they received in 2004. The evolution of the nationalist vote has been less smooth, but in general it has been upwards, from 31% in 1979 to 42% in 2004. It significantly exceeded its trend in 1999 when a large part of the Alliance Party and 'other' vote went to the SDLP's John Hume in an attempt (which failed) to top the poll. In 1999, while the nationalist vote jumped above its trend, the 'Alliance and other' vote tumbled to 2.3%. In 2004, with Hume gone, the 'Alliance and other' vote recovered, and the nationalist vote reverted to its historical trend.

Much used to be made of Ian Paisley's vote-winning ability, but during his time as an MEP the unionist share of the vote declined constantly, as it did in other elections – local, regional, Westminster. Ironically, the DUP's greatest share of the vote since 1984 came in 2004 when Paisley had retired from Europe and his now nemesis, Jim Allister stood for the DUP, winning 32% of the vote.

There are three seats available, and since the combined votes of the 'Alliance and other' candidates has never exceeded 9.5% – well short of a quota (25%+1) – these votes must be considered to be the core 'neither unionist nor nationalist' vote in Northern Ireland, around 8% on average. Three seats divided amongst four big parties (in the past) must leave one party disappointed. Since the unionist block has always outvoted the nationalist block, and transfers usually stay within blocks, two unionists were elected and only one nationalist. The gap between the two blocks is now much closer, and although unionism will still get more votes than nationalism in 2009, the breakdown of those votes, and the level of transfers will be crucial to deciding whether – in a historical moment – nationalism might snatch a second seat.

The unionist-nationalist balance

The breakdown of the vote in June's European Parliament election between the unionist block and the nationalist block will probably not differ much from the breakdown in other elections. The graph above shows how the two blocks have been approaching parity since the first EP election in 1979, but they have not yet got there.

On the basis of the most recent elections – the Westminster and local elections in 2005, and the Assembly election in 2007 – the relative strengths of the two blocks are approximately: Unionist 50%, Nationalist 43%. The unionist percentage has a downward trend, and the nationalist vote an upward trend. However, the period of time between 2007 and 2009 is too short for any significant change to take place, so it seems inevitable that the unionist block will again outpoll the nationalist block. If there are a high proportion of transfers between the unionist candidates (see below), this should translate again into two unionist seats and one nationalist seat.

The break-down of the unionist vote

One of the most interesting aspects of the election will be the impact of the two major events within unionism since 2004: Jim Allister's breakaway from the DUP and subsequent formation of the Traditional Unionist Voice (TUV), and the common-law marriage between the UUP and the English Tory Party.

Initially derided as a mere flash-in-the-pan by the DUP, or as a temporary usurper of the 'rightful' DUP seat, Jim Allister, by sheer persistence and good timing, has managed to remain a serous threat to the DUP. While few expect him to hold his seat, the DUP are concerned that he will steal their more extremist voters and consign them to an ignominious scramble for the third seat. Allister is playing the same role vis-à-vis the DUP that the DUP itself played vis-à-vis the UUP – that of constantly challenging the other party from a more extreme unionist standpoint. Such a role worked well for the DUP so it is justifiably worried by Allister. The DUP has left such a long string of hostages to fortune in its climb to the top that Allister has a rich seam to mine. And he is mining away, eroding the confidence that extreme unionists used to have in the DUP, and providing them with an alternative. The by-election in Dromore in 2008 showed that while it was still relatively small, the TUV could inflict serious damage on the DUP.

The Paisleyite wing of the DUP is reportedly angry about the manner in which Ian Paisley senior was deposed in 2008, and still hopes to get Ian Paisley junior elected as MP for North Antrim as his father's replacement at the next Westminster election. If the DUP candidate for the EP is identified with the 'modernist' (or non-Paisleyite) wing, some of the Paisleyite voters may vote TUV instead of DUP.

If the TUV manages to steal even 20% of the DUP vote (in Dromore it took 40% of the combined TUV/DUP total; but this was the only election so far in which both parties have competed, so it is not a good basis for predicting the outcome of their next competition), then the DUP might fall below a quota (for the first time ever in a European election). If Sinn Féin poll well, they might top the poll. The DUP may then suffer the double indignity of failing to keep Sinn Féin from topping the poll and having to depend on transfers from minor party candidates, which might not come. When the DUP candidate is eventually elected, he or she may have lost considerable face in the process.

The not-quite marriage between the UUP and the Conservative Party is being talked up incessantly by both sides. However, the 'joint' candidate is none other than the same old UUP candidate whose popularity has been constantly waning for years, and was only barely ahead of the SDLP in 2004. The UUP are desperately hoping for a 'bounce' from their not-quite marriage, but they may be severely disappointed. If a lot of TUV votes do not transfer to the UUP – a party most TUV voters heartily despise – the UUP risks losing its seat to the SDLP. Such an outcome would serious damage the credibility of the UUP-Tory link-up.

A second and less often cited ambition for the UUP/Tory link-up is the hope that it will make James Nicholson more attractive to Alliance Party voters. The eventual winner of the third seat is likely to be decided on the transfers from the 'Alliance and other' voters, and so anything that can make Nicholson less repellent to the centre can only help him. Positioning the UUP candidate (for, despite the hype about the nearly-a-marriage, that is what Nicholson is) as a part of a resurgent soft-right movement may be enough to earn him some extra Alliance transfers, and to keep him ahead of the SDLP's Alban Maginness.

The break-down of the nationalist vote

Another party hoping for a bounce is the SDLP, though their chances are worse than the UUP. Their candidate, the unsuccessful Alban Maginness, lacks any particular appeal – even in his own North Belfast constituency he has been losing ground over the past few years. Although Sinn Féin's Bairbre de Brún is hardly a dynamic MEP she is guaranteed a sufficient vote to get elected – and possibly to top the poll, to the DUP's anger. The SDLP vote should hold steady, though, and if other circumstances went their way, they could just sneak past the UUP for the third seat. Again, much depends upon the transfers of the Alliance Party voters.

The vote for 'other' candidates

In 2004 the assorted 'neither unionist nor nationalist' parties and groups coalesced around a 'unity' candidate – John Gilliland. The novelty of such a candidacy ensured publicity, and some of the more gullible 'centrists' started fantasising about winning a seat. Gilliland got 6.6% of the vote and promptly disappeared back into well-deserved obscurity. But it seems to have led to an expectation of a centrist candidate in 2009, combining the Alliance Party and other disparate groups. So far no such candidate has appeared, and the Alliance Party are keeping their intentions a well guarded secret. The only 'unity' name to have been leaked is Eleanor Gill, whose qualifications for the job are as small as Gilliland's were. She seems to be 'acceptable' to the centre groups because she is (politely speaking) a non-nationalist Catholic.

The Green party have already broken with the 'unity' candidate idea by naming their own candidate, the utterly unknown and unelectable Steven Agnew. He can expect to equal the Greens 2004 total of less than 1%. Hopefully they will not waste too much of the earth's resources on his hopeless campaign.

Historically the 'Alliance and other' vote has tended to be around 8-9% in EP elections, and this time should be no different. The real interest is where their votes transfer to. In 2004 it is likely that more of Gilliland's votes transferred to Nicholson (UUP) than to Morgan (SDLP), but a lot did not transfer at all (it is hard to be sure, as Gilliland's votes were combined with the Greens and Eamon McCann's). In 2009, if the 'Alliance and other' votes go slightly more to the SDLP, if the TUV achieve a reasonable score (say 6% of the total vote), and if the TUV votes fail to fully transfer, this could put the SDLP ahead of the UUP for the final seat.

A possible vote and transfer scenario (not a prediction) might be:


Anonymous said...

If SF only got 26% it would be the first election in some time where their share of the vote hadnt actually increased - (in fact it would have gone down from 26.3). This would also be the first election for the DUP where their vote had actually decreased - down 6%. I would guess the results will be SF 27% DUP 30%. I dont pretend to understand the quota system but whoever tops the poll will presumably bring in her fellow community background politician with her - presuming relatively equal performances by the UU and SDLP.

Do the results get split by constituency? I seem to remember these figures not being realeased last time?

It was Sammy Mc Nally what done it

Horseman said...


My figure of 26% for SF is just an approximation. Maybe they'll get more, but I reckon the total nat vote is around 43%, so if SF goes up, the SDP goes down (and that screws my fantasy about them getting the third seat). I don't think all votes transfer - we didn't get to see that in 2004 because all three seats were filled before De Brun's surplus was distributed. But not all Allister's votes went to Nicholson (only 33482 out of his 38441 surplus).

I think the TUV factor means that the DUP vote must go down this time. The first time for a while maybe, but it is inevitable. Their fear is that they'll have to fight for a seat for the first time ever in Europe. That alone is a big loss of face for them. There aren't enough unionist votes for the DUP to get 30%, unless either the TUV or UUP crash. If its the latter, then the SDLP may be the winners.

I'd love it if they gave the results by constituency, but I doubt if they will. They have never done so before, though they released the turnout and vote by constituency. Maybe that came from the polling stations rather thn the counts, though.

Anonymous said...

Sooner or later SF and the DUP have to peak - and I agree that the DUP will probably show a decline this time out as the TUV will have some impact on the hardline vote.

I wonder if it is worth an enquiry to the electoral office to find out if they will publish the figures by constituency and if not the basis for this. Presumably the Assembly might have an opinion and some influence on this decision if enough notice is given?

It was Sammy Mc Nally what done it

Anonymous said...

Another more negative point is that the Nat. vote % at 43% would be unmoved from the 2004 vote - (and down from 45 % in 1999) that would not augur well for the Nat. electorate exceeding the Unionist electorate and does seem to be out of kilter with the demographic profile of noth communities?

Horseman said...

Whether the nat vote is 43%, or 44% or 42% is not very important. At this stage we do not know what (if any) effect there will be from an Alliance Party or 'unity' candidate. The 45% in 1999 was not all a nat vote - it was mostly nat, but also some anti-Paisley, and some personal for Hume. The graph in the blog shows it clearly - the Alliance and other vote dipped to a very low point as these people voted or Hume. But before and after 1999 they voted for the 'Alliance and other' category, so were only 'borrowed' votes. If the Alliance party doesn't stand this time, then the SDLP vote should increase (and th it the nat total).

It could be that the competition between Dodds and Allister stimulates unionists to turn out in larger numbers, giving the impression of a unionist resurgence and a nationalist decline. Who knows. But this is just one election, and while its outcome may be anomalous, it should be seen in the longer-term context.

The turn-out may also be quite low, in which case the result is less reliable as a gauge of opinion. We'll see how much interest it stimulates as the campaign cranks up.

Anonymous said...

Surely the simplest thing is to take the Nat% and half the Alliance if they stand. If that is not above 45% (although allowance will have to be made for varying turnouts by nats and Unionists)then surely a lot of yours (and my) hopes for project Ulster will be looking a bit forlorn. There really should be some continuing translation of demographics over a 5 year period or the numbers game may be drying up as a source of optimism. I think the % of persons with no religion ( the majority of whom are probably of Unionist background based on differences between for example East and West Belfast) may be distorting the picture favourably towards the Nats.

It was Sammy Mc Nally what done it

Horseman said...

Hi again IWSMWDI,

One reason I am not too bothered about election results, at the margins at least, is that they are very dependent on turn-outs. For different reasons the two tribes may turn out differently on one or other occasion. It doesn't mean that the project has halted or reversed. I prefer to look at the longer term trends which smooth out the peaks and troughs.

The graph above that shows the decline of the unionist vote is one such illustration. Unionism will never again get as much as 59% of the vote in an EP electon. It will have trouble even reaching 50%! The same can be seen at local, Assembly or Westminser elections.

And if you look (as I do) at the demographic picture, especially at the younger ages, it shows that the unionist project is screwed. They still make a lot of noise, but the stats show that unionism is in structural decline, and it is literally just a matter of time until they are outvoted.

As that penny starts to drop they (the unionists) will try to persuade people that voting for 'tribal parties' is so passé, and that we should all vote for 'normal parties' (i.e. British ones). If they can persuade enough Catholics to vote for the British Labour or Tory parties (which are de facto both unionist in principle), then the vote for the capital 'N' nationalist parties may struggle to reach 50%. But personally I think the sectarianism inherent in unionism is sufficienty strong to ensure that most Catholics keep voting nationalist. Only a few unionists seem to get that, and they are constantly undermined by the bigots.

Anonymous said...


I agree with your Nat.Uncle Toms (NUTS) analysis but as I mentioned in my previous post I'm not sure you are taking account of the no religioners in the 2001 census - these appear to be predominantly Unionist in background - if analysed by constiuency.

As we discussed previosly a breakdown by consituency would assist greatly in working out the % turnout by community background. Do know the rationale for not releasing this information is it decided at UK level? If you cant throw any light on this I will email the Electoral Office of Norn Iron.

It was Sammy Mc Nally what done it

Horseman said...


It may be true that the 'no religion' group are mostly unionist in background, but they may not be unionist voters. They may in fact make up a large part of the 'Alliance Party and others' voters, who are often most numerous in the constituencies highest in the 'no religion' group. I don't know if anyone has done any research on this - certainly I don't know of any.

On the constituency-level breakdown of the EP vote I can only guess that it is because for EP purposes there is only one constituency. But it would do no harm to contact the electoral ofice to ask them. Maybe they do have the figures and would release them on request.

Anonymous said...


I will email the Electoral Office and ask them.

re. Non religous - it must be the case that the majority are Unionist background - because if you comapre east and west belfast with similar socio-economoic factors east belfast has a far higher % of non religous. Dont really think that needs any further 'research' - as it can be seen in the other Unionist constituencies.

South belfast which has a higher socio-economic group and is nearly 50% Nationalist has a very high % but I suspect it is a case apart.

In South Belfast's case the Nat% are quite spectacular over the last 2 assembly elections even allowing for an increased Alliance vote - not sure if any boundary changes are partly responsible?

Horseman said...


Good luck with the electoral office!

While you could well be right on the non-religious, I really do find it unlikely that young Catholics are so much less likely to call themselves non-religious than young Protestants. Could part of the issue be that in recent censuses (censi?) a lot of Catholics got into the habit of not declaring their religion, for reasons of safety - and this would be even more important in majority unionist areas where you really didn't want the enumerator (who just might have shared the info with someone) to know that you were the street's sole Catholic!

South Belfast is a really interesting place - the rate at which it has 'greened' is amazing. It isn't just boundary changes either - a LOT of middle class Catholics will only live there - the west is too tribal, and the east too unionist. It is the place of choice for mixed marriage couples too. I guess that the majority of incomers to the constituency have been Catholic, and the entirety of the outgoers were Protestant. Have a look at the blog on Belfast City Council ( - it hows how fast the demography of Belfast is changing, and I think SB is at the forefront of that.

Anonymous said...


", I really do find it unlikely that young Catholics are so much less likely to call themselves non-religious than young Protestants"

Yes, that seems a reasonable point BUT - it could be explained by the tendency of the British-Irish the Unionists to behave more like the British mainlanders and possibly "Protestantism" having generally less stickability - "Catholicism" being more difficult to shake off.

In relation to the Belfast City thread it was I "It was Sammy Mc Nally what done it" what posted the final question on that thread -which is a question which still puzzles me - though I'm probably misreading what you are saying.

Anonymous said...


in relation to my last remark - I can now see what you are saying

"However, in 2001 the electorate of Belfast City Council area was 51.6% Protestant, and only 45% Catholic"

I didnt appreciate that quote above related to the 2001 census.

The 2005 local elctions actually saw Nat% at 48.4 up from 46.5%(2001) - that includeds the WP with the Nats. These are diffcult to understand, even allowing as you do for Unionists voting for 'others'. If you assume only a quarter (2.25% )of the others (9.1) are Nats background - that pushes the Nat background over 50%.

It is the reverse of the Unionist-Catholic arguement - and does seem to happen in other constituencies?

Horseman said...


It is strange that the apparent nat percentage is higher than their percentage of the electorate - I can only assume that they had a higher turn-out rate. Or that the census miscalculated the true community of many of the 'not stated'. Or finally, that between 2001 and 2005 a lot of unionists died or moved out.

Anonymous said...



the electoral office have just replied to my email with the Wesminster Constituency breakdown link for Euro elections. Relatively straigtforward to allocate community breakdown by constituency to calculate a rough Nat/Uni turnout. Will have a go at this.

Seymour Major said...


Firstly, I would like to express my admiration for your work on your Blogsite. Today is my first visit.

"....... But personally I think the sectarianism inherent in unionism is sufficienty strong to ensure that most Catholics keep voting nationalist. Only a few unionists seem to get that, and they are constantly undermined by the bigots"

This is the last para of your comment on 6th Feb 15.30

This opinion could be right or wrong but the right or wrong is the most important question in relation to Ulster's future.

One of the assumptions that you are making is that the tribal vote, which is brought out at elections, would be the same if there was a referendum on Northern Ireland's future. There is some research (which you probably are already aware of) which suggests this might not be the case.

However, if the CU's fail in their quest to attract sufficient numbers of Catholis, then these Catholic "Unionists who never vote unionist" could eventually become permanently alienated.