This appears to have been a relatively encouraging week for Remain in the polls. Of six polls published between Monday and Wednesday that had used the same polling method as in a previous survey, five identified at least a small shift in favour of Remain. In the event, the one that did not – by ComRes for The Sun – was over a week old, and a second ComRes poll (for the Daily Mail and ITV News) joined the throng of swings to Remain.
We should though perhaps bear in mind that the date of the relevant previous poll varies considerably. In some cases this was just a week ago, while in others it was as much as a month past. However, even if we ensure the comparison is made systematically by in each case comparing the latest figures with those for the equivalent poll conducted a month ago, as is done in the table below, we still find that there has been a small swing to Remain.
Change in Remain share of Referendum Vote Intentions, March-April 2016
Most of the swings are small, and individually could simply be the result of the chance variation to which all polls are subject. Meanwhile, as the only poll conducted by telephone to put Leave ahead, the ORB poll of mid-March increasingly looks as though it was a bit of a ‘rogue’, and that thus we might want to regard the change being reported by that company as something of an outlier. But when five polls all show the same trend, it becomes much more difficult to reject the claim that there really has been some movement amongst voters as a whole.
That said, Remain should be cautious in taking pleasure in these results. Three of the polls in our table, those conducted by ComRes, Ipsos MORI and ORB, were conducted by phone, as was a second poll from ICM that is not included in the table because it was the first referendum poll that ICM have undertaken by phone, and which reported a 54% Remain vote. While all four of these polls put Remain well ahead, at 54-56% the range of support they report for staying in the EU is still well below the 60% that telephone polls were reporting on average during the winter. That the gap between the figures reported by phone and internet polls is now much narrower than the near ten point gap that was once in evidence is affirmed by the fact that ICM reported just a four point difference between the poll they conducted by phone and their most recent online poll. (Both polls were conducted over exactly the same period.)
At the same time, this week’s polls have also been consistent in reporting that Leave voters are more likely to say they will make it to the polling station. For example, both ComRes and Ipsos MORI report that Leave voters are nine points more likely than Remain supporters to say they are certain to vote in the referendum, while ICM put the gap at six points in their online poll and seven points in their phone survey. ComRes put the two camps five points apart on this criterion. At present, of these companies only ICM are taking voters’ reported likelihood of voting into account in calculating their headline vote intention figures. Doing so makes a small but detectable one point difference to ICM’s estimate of the Remain vote.
But why might the Remain vote have increased at least a little? That is rather more difficult to determine. One central issue in the campaign is, of course, the economic consequences of remaining in or leaving the EU. So perhaps the Remain’s argument that leaving would be economically disadvantageous may be beginning to percolate through to voters (though we should remember that the latest polls predate the publication of the UK government’s claim this week that leaving would cost every household £4,300 a year)? But there is limited evidence that this is the case. True, ComRes report a three point increase in the proportion who think that leaving the EU would be a ‘big risk’ for the economy, but this was counterbalanced by a two point drop in those who said it was a ‘slight risk’. Meanwhile, TNS report that at 36% the proportion who think that being in the EU is beneficial for the UK’s economy is much the same as the 35% who were of that view in February.
There is also some suggestion in some of the polls that Conservative voters in particular may have swung around in favour of Remain. ORB suggest there has been a ten point swing in favour of Remain amongst those who voted Conservative last year, while TNS report a nine point swing. However, there is no such movement in ICM’s or Ipsos MORI’s polls, while the necessary crossbreak is not available in ComRes’ mid-March poll. Meanwhile, we should bear in mind that the sample size of many of these polls means that they contain limited numbers of Conservative voters, and that thus quite substantial swings can occur purely by chance.
What does appear to be the case, if ORB are to be believed, is that the Remain campaign has become more visible (perhaps because the government’s controversial leaflet setting out its reasons for wanting voters to vote to Remain is now being distributed to voters in England). Now 44% say they have heard more from the Remain campaign, while just 25% say they have heard more from Leave. A month ago the figures were very different at 39% and 40% respectively. The last week has certainly shown the richness of the resources available to the Remain side to make an impact, including the civil service and the support of Barak Obama. We will learn from the next clutch of polls whether this advantage means Remain have indeed begun to gain some momentum.