Alex Kane: East Belfast shows that the era of safe unionist seats is over

“If unionists have trouble holding a constituency like East Belfast, then it won’t be long until they’re having problems in places like North Down and South Belfast.”

That was said to me in the early hours of May 7 2010, by someone who had spent the previous few weeks canvassing for Peter Robinson.

He acknowledged that Robinson had been dogged by personal/political difficulties, while the intervention of a TUV candidate had also hurt his chances. But he made another point – which was of much greater significance than I realised at the time – about what he described as the “small-u unionist” drift from mainstream electoral unionism to a more centrist ground.

In some ways – and I mean no disrespect to her – Naomi Long was helped to victory in 2010 (albeit with a narrow margin of 1,533 votes) by a perfect storm of circumstances. But what mattered most was the sudden realisation that the seat was no longer a bankable unionist one.

Unionists have pulled the Stormont brake, but there’s no need for it to create a crisis - The Irish News view

Toto, I’ve got a feeling we’re not in Tyrone any more

That was a huge fillip for Alliance – and other smaller parties – because if East Belfast was a possibility, then there was space for the significant electoral growth which had eluded the party for years.

I was sitting in a TV studio when the rumours that Robinson was in trouble began to seep through – and I still remember the look on the faces of a couple of DUP backroom staff.

During the campaign there had been concerns that he could take a hit, but I didn’t meet one person who predicted a loss. Particularly a loss to a non-unionist party.

East Belfast had been, since the first post-partition election in 1922, a unionist seat. A safe Ulster Unionist Party seat. A seat where it didn’t matter how many unionist candidates were in the field because there would never be enough non-unionist votes for victory.

And that was the way it remained for decades, until February 1974, when the United Ulster Unionist Council (UUUC) candidate and Vanguard leader William Craig won it on an anti-Sunningdale ticket – beating the pro-Sunningdale and sitting UUP MP, Stanley McMaster.

That was a crucial election for unionism, because in the 12 constituencies the main battle was between unionist candidates who backed Brian Faulkner and the UUUC candidates (DUP, Vanguard and Official Unionists) who opposed the entirety of the Faulkner/SDLP/Alliance power-sharing deal.

In the coming general election we may see another version of the 1974 pro and anti-deal unionist battle: this time with a batch of once-safe unionist seats under threat.

Indeed, in the 1979 general election, Craig – who had softened his stance on power-sharing and rejoined the UUP – was defeated by Peter Robinson. That was a victory of enormous significance, because it gave the DUP a foothold in Belfast and proved that it was capable of damaging the UUP, especially from the right.

It was to be a long journey from then until it became the largest party in 2003, but it was a journey which started with Robinson’s victory, allowing him to build new, professional party structures and organisation from the ground up.

Meanwhile, East Belfast remained a safe unionist seat for another 30 years, with Robinson winning by very comfortable victories. There was a view after the 2010 defeat (not one shared by the person I spoke to a few hours after the result was announced) that the Long win was just a blip and that the seat would return to unionist hands fairly quickly.

And that’s what happened in 2015, when Gavin Robinson regained it with a majority of 2,597 – down over 3,000 votes on Peter Robinson’s 2005 majority. But it’s worth noting that Gavin Robinson was the only unionist candidate (the UUP and TUV stayed out and the local Conservatives fielded a paper candidate).

Robinson boosted his majority to 8,474 votes in 2017, with the UUP fielding another paper candidate, winning just 1,408 votes (3.3%). In 2019 his majority slumped again to 1,819 and again the UUP fielded a paper candidate.

Paper candidates serve a purpose for the UUP: they allow the party to offer choice, but not the sort of choice which risks the seat being won by non-unionists. And it also means that some UUP will drift to the DUP, offsetting those which have drifted to Alliance – particularly after Brexit.

But the fact remains: what was once a jewel in unionism’s electoral crown is now a marginal seat. Yet the threat isn’t coming from nationalism/republicanism; it is coming from Alliance, a party which remains agnostic on the constitutional position.

And that rise in the Alliance vote across Belfast has already cost the UUP and DUP council and assembly seats. It has cost them the North Down seat. It is a vote which is also growing in seats like Strangford and Lagan Valley, as well as East and South Antrim.

It doesn’t follow from any of this that unionism is bound to lose more seats to Alliance, but there is now a mountain of evidence indicating that the era of safe unionist seats is over: and by safe I mean seats that have the sort of built-in unionist majority that allows unionists to compete against each other – which was actually fairly common from the early 1970s to the early noughties.

Interestingly, I think the dangers of unionist seat losses are also likely to increase even under PR if they continue to field three or four parties.

What was once a jewel in unionism’s electoral crown is now a marginal seat

So, what’s the lesson for unionism from all of this? Don’t field much-the-same kind of parties with much-the-same kind of candidates when their only aim seems to be to squabble with each other and then blame each other when – as is increasingly the case – votes and seat numbers are down.

Don’t pretend to be offering ‘choice’ when all you’re actually doing is confusing and antagonising potential voters.

And, most important of all, abandon the nonsense of election pacts for certain seats, which does nothing more that highlight your own electoral/demographic weaknesses.