The LibDem Collapse, Fact or Fiction

The latest polls indicate that if another general election were to be held tomorrow Labour would sweep back to power with a majority (assuming uniform swing) of 46 over all other parties. The Conservatives would lose 45 seats, despite maintaining their 2010 share of the vote. The Liberal Democrats meanwhile would lose 41 of their 57 MPs, leaving them representing a paltry 16 Constituencies. In 7 short months since the 2010 election, 6 in every 10 Liberal Democrat supporters would have defected elsewhere; they would be back to levels of support last seen in the 1970s under Jeremy Thorpe.

To believe this you have to trust the polls. And why not; since the polling debacle of 1992 pollsters have had a pretty good run, and their record of accuracy in the last two general elections has been among the best on record. It is possible to argue that the polls are right about the collapse in LibDem support, but it is also quite easy to find reasons to worry that the polls are telling porkies.

The polls are right. The argument might go that the LibDems have built their support by being a soft centre party to which disaffected Tory and Labour voters can most easily turn, as well as those who are confused about who to support or don’t know who should get their vote. True Liberal Democrat supporters can be found among those who voted for them up to and including the last general election, but not that many.

Jo Grimond (leader of the Liberal party from 1956-1967) once said he didn’t think it was a good idea for the Liberals to have an election manifesto. That wasn’t such a silly idea. Being the party largely supported by those who, for one reason or another, reject the two major parties can work well for a soft centre opposition, but not one that is in coalition government with the Tories. Presently, the LibDems can no longer present themselves as the reasonable, easy alternative for disaffected Conservative or Labour voters, or for those who cannot make up their minds between the Conservatives and Labour. So a collapse in support for the LibDems is to be expected.

The polls are wrong. Poll respondents are asked to imagine that there will be a new general election tomorrow. Given only a moment or two to think about that unlikely scenario, they are then asked to say how they would vote. In reality, with the next election four years or more away, ordinary people will not have given their current vote intentions much serious thought, if any. Nevertheless, respondents are asked to choose between four discreet alternatives pretty much on the spot (Conservatives, or Labour, or the Liberal Democrats or another party). The question thus assumes that, at a stroke, the coalition has been formally wound up, or breaks down, that all the main parties have already gone their separate ways and have campaigned on different manifestos for votes in an election campaign ending tomorrow.

It can be argued that the old vote intention question has been made redundant (perhaps temporarily) by the establishment of the coalition simply because it asks about four alternatives at a time when respondents can only really choose – or realistically imagine a choice – between three, “this lot” “the other lot” or one of the minor parties.

The result in Oldham and Saddleworth gives further cause for concern. Dangerous though it may be to read too much into by election results, it did not exactly fit in with the poll story of a collapse in LibDem support.

Perhaps, therefore, polls suggesting that the LibDem vote has disintegrated should be taken with a pinch of salt, though some erosion in their support was perhaps inevitable once they decided to go into government with the Conservatives. But in addition, perhaps, pollsters should start to think more about other questions – apart from the standard 4 choice vote intentions – that might for example measure support for the coalition, its policies and performance and attempt to gauge whether voters think the government is moving Britain in the right or wrong direction. Tracking those things over time might be more meaningful (and indeed useful) than slavishly reporting vote intentions (in one case daily!) for a non existent election and a choice of alternatives that does not currently exist.

Voters will only really know about how they will vote at the next election when they also know what the choices will be, and that will only happen when the Conservative and LibDem leadership decide how they wish to present themselves to the electorate next time, jointly, separately or in some loose electoral arrangement, with or without some agreement on key election pledges.

In the meantime those hanging on the next set of vote intentions should, I suggest, also keep the salt near at hand.

Nick Sparrow was formerly head of polling at ICM