All eyes on the prize - and on the second prize
Updated on 05 May 2010
The Liberal Democrats have the most to gain from the general election, and Labour has the most to lose. Independent polling analyst Greg Callus looks for Channel 4 News at the battle to be the country's official opposition.
The polls are suggesting a knife-edge finish to the 2010 general election – the current state of the race suggests Cameron's Conservatives will poll in the mid-30s, with Labour and the Lib Dems fighting for second place in the popular vote in the high-20s.
There are outliers, showing a potential Labour collapse below Michael Foot's 23 per cent in 1983, and others suggesting that Brown will not suffer the ignominy of falling to third place at all.
Different polling methodologies are producing broadly the same figures, but with slight variations that could be critical if – as the betting markets suggest – this really is a coin toss between a Conservative overall majority and a hung parliament.
But to what extent should we trust the pollsters? It would be foolish to underestimate the challenge that they face in this election. Unlike in countries like the US, where the major two parties command over 90 per cent of the vote, British pollsters are forced to deal with three major parties, regional four-way split in Scotland and Wales, a separate slate of parties in Northern Ireland, and a collection of minor parties such as UKIP, the Greens, and the BNP.
More analysis from Greg Callus
- Could a minority government deal be done?
- Electoral reform – who would be the winners?
- Voter's guide to engineering a hung parliament
This fragmented picture would be difficult if it were long standing, but the explosion of Liberal Democrat support is but a month old, and the increase in the minor party suggested vote share does not stretch back to the last election.
Quantifying the new shape of the British electorate, even without complications such as the anti-incumbency feeling in the wake of MPs expenses, or the challenges of applying turnout filters, makes this almost incredibly difficult to predict with any degree of accuracy.
This election has a much greater degree of uncertainty than in years past – and years past have not always covered the pollsters in glory either. Most famously, in 1992, the polling industry completely overestimated the Labour share of the vote, sometimes by as much as 8-points. The explanation given was 'Shy Tory Syndrome', whereby supporters of the disliked Conservative government were reluctant to state their support.
Adjustments in response much improved the predictions in 1997, though as recently as last year, pollsters were introducing new ways of overcoming sample bias. Just recently, Ipsos Mori recognised in their polling a disproportionate number of public sector workers, which it believed was inflating the share of the Labour vote, requiring a reweighting on this criterion.
Mike Smithson, editor of PoliticalBetting.com, has long considered "a golden rule" that the most accurate poll will be the one that has the lowest share of the Labour vote. As pollsters in the US Presidential election of 2004 found, supporters of some parties are just more likely to speak to pollsters – a major factor behind the exit polls calling the race for John Kerry.
The stabilising measures should help avoid a 1992-style disaster being repeated, but statistically significant moves recorded by the polls might still not come to pass when polling day comes.
Will the support won by Liberal Democrats from the under-30s (an age group least likely to actually vote) materialise, even if they did register on-time? Those who say they don't know who they will vote for are the most unpredictable – can we have any firm indication of how they will split? And what will turnout be?
The BNP won their second MEP in 2009, not because they increased their vote, but because Labour voters stayed home in unprecedented numbers in Yorks and Humber. The impact of not voting is difficult to measure in an opinion poll, as are the 'wasted votes' for micro-parties.
Even if we could have absolute faith in the polls, the seat calculators used to try and predict how the votes will turn into MPs are under even greater pressure. In previous years, the seat calculators (such as produced by Anthony Wells or Martin Baxter) modelled on modified uniform national swing were accurate enough.
But again, designed to model a two-horse race with a small third party, they are now in uncharted territory. For those who are particularly interested, it would be worth comparing the work done by Robert Ford of Manchester University for PoliticsHome, the proportional swing models of Robert Smithson (VIPA) and Nate Silver (538.com), or the extensive analysis conducted by punters such as Andy Cooke at PoliticalBetting.com.
What happens next?
A couple of questions emerge of the utmost significance. Imagine that the majority of polls are correct, and the Lib Dems outperform Labour to come second in the popular vote.
1) Is there any possibility that the Liberal Democrats could come second in seats, making Labour only the third largest party in the next Parliament?
2) Would the fall of Labour to third place in the popular vote mean further decline and a loss of power for a generation (a two-party system of Conservatives vs. Liberal Democrats)?
On the first question, it is difficult to see how the Lib Dems could overwhelm Labour to become Her Majesty's Loyal Opposition in the next parliament.
Taking figures from one of the scenarios charted by Nate Silver, it would require the Conservatives to win around 366 seats (forming a government with a majority of 62-or-so) off 36.9 per cent of the popular vote, with the Lib Dems winning 130 seats off 30.5 per cent, and Labour shrinking to 126 seats and only 23.1 per cent of the vote. This is by no means impossible, but it would require a stroke of significant good luck (strong Lib Dem performance, Labour collapse, and a Conservative share large enough to cost Labour dearly, but not so great as to take too many seats from the Lib Dems in the South West of England).
The second question is contingent on the first – if the Lib Dems don't manage to come second in seats, in spite of coming second in the popular vote, it is unclear what could happen next. One possibility is that we could see three roughly-equally-sized parties as the standard mode of British politics for the near future. This is eminently plausible, but not as interesting an outcome for those fond of political drama. The other theory suggests that 'first past the post' (FPTP) systems, like that used for Westminster elections, lend themselves to two-party duopolies, and that either Labour will sink into third party status, or the Liberals will resume the position they occupied prior to the TV debates in 2010: the largest of the minor parties.
Without a change in the electoral system (to PR, or STV, or AV+) and the normalisation of coalitions, I have some sympathy for the latter view. The very construction of the chamber at the House of Commons suggests a dualism, of government vs opposition, of two opposing dispatch boxes dominating the political conversation.
The Liberal Democrats would undoubtedly not being doing half as well in the polls without the requirement for equitable coverage during the election campaign, especially their inclusion in the TV debates. In the long-periods between elections, the media will tend to paint quasi-objective narratives, reflecting both (rather than all) sides. The electoral benefit in being the official opposition (which is determined by second largest party in terms of seats, not votes) cannot be underestimated.
For that reason, the Lib Dems and Labour face a critical couple of months.
If the Lib Dems are able to become the official opposition, it could well resign the Labour party to minor third party status for many years to come, as anti-Tory voters and donors rally behind their new best hope. If the Liberal Democrats are the third largest party still, then they need to do whatever is necessary to effect electoral reform, by offering themselves in coalition, otherwise their achievement in coming second in the popular vote will quickly be as easily forgotten by the electorate as the performance of the SDP in 1983 (almost second in the popular vote, but nowhere close in terms of seats won). If they cannot manage this, then they should hope for another election as soon as possible, to benefit from their staging performance in May, and the nascent belief that they can overcome Labour with just one more push.
For Labour the challenge is quite the opposite. If they hold on to second place in terms of seats, in spite of coming third in the popular vote, then they need to resist another general election in 2010.
The longer they occupy the opposition dispatch box, the more faded the memory of the potential of the Lib Dems to be one of the two major parties. They need to solidify their position as the official opposition, and resist being consigned to third party status – that means a long, non-election period in which the Labour leader and the Conservative leader continue to face each other at Prime Minister's Questions, and the media report them second in the news bulletins.
To have to fight a second election in 2010 would not only cost Labour money it doesn't have, but might also allow the Lib Dems to build on the swell of support they are expected to get on 6 May, and to catapult them to becoming the official opposition in a two-part manoeuvre.
The Lib Dems have the most to gain from the May 2010 election, and Labour has the most to lose. Coming second in vote share won't be the best result Nick Clegg could hope for, though it would be a significant achievement.
The perfect storm might mean becoming the second largest party in a continued FPTP system: an outcome that could be better for the Lib Dems than coming third and getting electoral reform as the junior partner in a coalition.
Coming second in the popular vote seems to be the best Labour can now hope for – to fail to be the second largest party would be the most disastrous result for a government since women were first allowed to vote in 1918.
Even if they can hold on as the official opposition, Labour must hope that the new government survives long enough for belief in the Lib Dems to subside – which might prove a predicament if facing a Cameron administration with little or no majority in the House.
The collapse of the Labour party is by no means assured – but the circumstances exist whereby their permanent position at one of the House of Commons' dispatch boxes is at least open to reasonable speculation.
Greg Callus is the deputy editor of politicalbetting.com and co-author of the Guide to the 2010 election.
