UK Politics / Paul Krishnamurty / 23 July 2008 / Leave a comment

Politics betting expert Paul Krishnamurty on the Glasgow East by-election - where defeat for Labour could spell the end for Gordon Brown's troubled tenure at No.10

On Thursday, we could well witness a political earthquake, a by-election result that will go down in history as a mark of the end of an era, and possibly the beginning of the end for one of the UK's two main political parties. If Labour fail to hold Glasgow East, their 25th safest seat, then even their staunchest supporters will wonder whether they can win anywhere. Moreover, it could spell a premature end of Gordon Brown's tortured leadership.

The word 'if' is crucial though, as the market trends suggest Labour is poised to hold on. When betting opened a few weeks back, Labour and the SNP were both trading around [2.0]. Now, perhaps influenced by two questionable opinion polls giving Labour a double-digit lead, they've moved into solid favouritism at [1.4].

It's crucial, however, to remember that with political betting, the market is rarely driven by accurate 'information' and odds-on chances get turned over regularly. Famous examples in recent years include Alan Johnson, (sub [1.1] just minutes before it was announced he'd lost the Labour Deputy contest), John Kerry, ([1.25] shortly before it became apparent George W Bush had won a second term), and more relevantly for Thursday's contest, Labour traded at [1.01] before losing the Dunfermline and West Fife by-election in 2006.

In that last case, which came at a time when Labour were in a much stronger national position, they suffered a 16% swing to the Lib Dems who themselves were in turmoil nationally following the resignation of Charles Kennedy. Labour lost because their supporters simply failed to turn out, with their total almost halving. I suspect we will see a similar trend in Glasgow, which suggests betting [1.4] about Labour is a seriously risky business.

Before getting to the individual seat though, it's worth considering the wider picture. Labour are coping badly with opposition, lurching from one crisis to another whilst the SNP have increased their poll lead since taking power in the 2007 Scottish Parliament elections. Alex Salmond is proving a popular First Minister while Labour have just lost yet another leader with Wendy Alexander forced to resign over illegal donations. This election couldn't come at a worse time for Labour.

But then again, this is Glasgow East, a seat they won with a whopping 43% lead in 2005. Those recent polls suggested that, whilst the lead will be cut dramatically, they should hold on comfortably. However, Labour have been consistently over-estimated in opinion polls, both local and national, for several years. As Mike Smithson of politicalbetting.com has argued, their support was similarly over-estimated in polls before the disastrous defeat in Blaneau Gwent, and before a near disaster in Hartlepool four years ago. Those two seats could also be described beforehand as 'Labour heartlands' so perhaps nowhere really is safe anymore.

The common trend among those 'rogue' polls, and the two out this week, is that they fail to ask whether respondents are 'certain to vote'. It's no surprise to hear that respondents in a core Labour seat express a preference for them, but whether they will turn out to vote is another matter altogether. And crucially, by-elections tend to disproportionately encourage those wanting to register a protest vote.

Although I still make Labour favourites, ([1.7] would be my assessment), I think they will struggle to get much more than 10,000 votes, compared to 18,775 in 2005. The key will be how organised, and how tactical, the opposition vote is.

The SNP, who only managed a paltry 5,268 votes themselves, need a substantial proportion of the 7,000 Tory, Lib Dems and SSP supporters from 2005 to put ideological preferences aside in order to vote for the party best placed to beat Labour. Ominously for Labour, this strange tactical coalition was the driving force behind their 2007 Scottish Parliament defeat.

My view then is that Labour, and therefore Brown, should just hang on by their fingertips. But at the prices, the SNP look fair value. My betting strategy will be to have a small lay of Labour and trade out if and when they drift in-running. Its become almost standard on these by-election nights for the media to report that 'it's too close to call', even when that isn't the case.

Two other markets worth a mention are turnout and SNP vote-share. In the former, I think the 30.01 - 40.0% band looks almost nailed on. Above that is highly unlikely and well worth laying at anything under [3.5]. I could imagine it just slipping under 30%, but that must be doubtful given the race is assumed to be fairly close. Marginality always brings people to the polls.

As for the SNP vote share, they registered 17% last time but look certain to top 30% this time. Again, my view is that the 30.01 - 40.0% band looks a very solid favourite, and anyone thinking of backing the overs would probably be better off simply backing the SNP to win the seat.

Tags: 2007 Scottish Parliament elections, Alex Salmond, Charles Kennedy, Conservative Party, Glasgow East By-Election, Gordon Brown, Labour, Lib Dems, SNP, SSP, Wendy Alexander