When bad polls are good news for governments

By Paul Goodman
^{Follow Paul on Twitter}

Comres snapshot: Labour 47%, Conservatives 30%...

Matthew Barrett reported yesterday evening that the latest Comres poll indeed gives Labour, if the last of two samples in the survey is considered, a lead of no less than 17%: the precise figures are Labour 47%, Conservatives 30%.  I want to explain why bad polls can be good news for governments - and why even this one isn't as bad as it looks.

As Matthew pointed out, this sample was taken in the aftermath of the Sunday Times's Cruddas story.  They are responsible, together with a second sample taken earlier, for the headline finding which he reported: the biggest Labour Comres lead since 2005 - one of 10% (quite big enough in itself).  It comes with other worrying findings.  Over at UK Polling Report, Anthony Wells writes that "in ComRes’s poll there is a hefty Labour lead amongst 55-64 year olds, and a single digit Tory lead amongst over 65s...Keep an eye on it, but it’s starting to look like the Conservatives may have taken a knock amongst those affected by the granny tax."

...But it's not as bad as it looks

But now let's look at the first Comres sample, and see how the poll as a whole compares with two others published this morning - and undertaken by Populus and YouGov respectively.

That sample apparently represents some two-thirds of the whole Comres poll.  Taken before the Cruddas affair, it showed a Labour lead of four points - which is roughly in the range of recent polls.  Populus, like Comres, took about two-thirds of its sample before the Cruddas story and a third afterwards in its survey for the Times (£).  It finds a Labour lead of four points - the same figure, of course, as that ComRes first sample.  The Tories are down three points on 34, and Labour down one on 38: "others" rise by three points to 16%.  Finally, YouGov for the Sun puts the Conservatives at 35% and Labour at 42%.  Wells says that this poll was conducted wholly on Sunday and Monday. Oh, and since you asked: the LibDem totals are 11 per cent, 11 per cent and 9 per cent respectively.

The key is economic trust...

It may be that the second Comres sample is only the first of a similar series of findings which will stretch from now on until general election day.  But I doubt it.  The evidence simply suggests that the party has taken a knock from Cruddas: the effect may linger and then again it may not.  In any event, neither Populus nor YouGov's findings are as bleak for the party as Comres's.  Wells also notes that "the margin of error on a third of a poll of 1000 people is huge, and it may just be that ComRes got more Laboury people on Sunday".  He also cites an ICM poll over the weekend which found "a big Tory lead amongst the elderly, so the traffic isn’t all one way".

But the most suggestive findings are those on the economy.  The Times writes that Populus has found "a six point swing from voters saying they trust David Cameron, Nick Clegg and Mr Osborne more to manage the economy towards Mr Miliband and Mr Balls".  But its report goes on to say that some 60 per cent of those sampled picked the coalition team to “manage the economy in the best interests of Britain” while only 40 per cent plumped for Labour’s team.  Matthew quoted a YouGov poll  showing that 60% of voters want economic strength rather than economic fairness (32%) - and that, on that measure, the Conservatives are ahead by 31% to 24%.  The Times reports no evidence that the 50p rate cut has damaged the party.

...And as long as the Government has more of it, it can take bad polls as a compliment in hard times

The Tories are still bruised by the effect of the Thatcher years, in which they came to be seen as competent but uncaring.  I do not want that effect to be repeated, and honour those strive to ensure that it isn't.  But I believe that shifting public opinon to recognise that caring doesn't equal more government spending - and indeed that such spending can be harmful - is very much a long-term project.  Voters tend to take a chance on caring when times are easy: when these are harder, competence is prized higher.  If the party can maintain a lead on competence and economic management during hard times, it's in a good position to endure bad polls - of which there are two kinds.

The first are those that show the electorate has simply stopped listening to the party in question - such as those that dogged John Major and William Hague.  The second are those that reflect irritation, disappointment, exasperation, even fury - all of which, however discomforting, are forms of engagement.  Margaret Thatcher's governments received a mid-term thumbs-down from polls on a regular basis.  This didn't prevent her from winning three successive elections, two of them with whopping Commons majorities.  She was taking difficult, unpopular decisions.  So did George Osborne in the parts of the budget to which voters seem most to object.  He should take their displeasure as a sign that he's doing at least some things right.