More YouGov, Ashcroft and Tactical Voting

There were a couple more bits of polling for the Westminster election released earlier this week. First of all, The Times published the results of a Scotland-wide YouGov poll. Second, Lord Ashcroft released the results of two further constituency polls, this time in Edinburgh North & Leith and in Edinburgh South. Neither makes very happy reading for the Labour party.

YouGov’s previous poll, conducted ten days earlier. put the SNP on 49%, Labour on just 25%. That represented a record SNP lead so far as YouGov were concerned. The latest poll replicates those figures and thus adds weight to the suggestion that the SNP may be pulling even further ahead. Certainly our poll of polls of Westminster voting intentions now has the party at 49%, higher than it has been at any time since we introduced it after the referendum.

Apart from asking about voting intention the YouGov poll also examined once again whether anti-SNP tactical voting might save many seats from falling into nationalist hands. These data show once again that if they were convinced that the race locally was between the SNP and a party other than their own, Conservative and Liberal Democrat voters would switch to the SNP’s opponents in greater numbers than to the SNP itself. At the same time they also show that a fair proportion of Labour voters would switch to the SNP rather than to the Conservatives or the Liberal Democrats.

Thus, faced with a Labour versus SNP race locally 44% of Conservatives say they would back Labour while just 8% would support the SNP. The equivalent proportions of Conservatives that say they would switch in a Liberal Democrat versus SNP contest are 46% and 4% respectively.

Meanwhile, no less than 55% of Liberal Democrats say they would back Labour against the SNP, with just 9% switching the other way. The figures are very similar, 56% and 11% respectively, for Liberal Democrats faced with a Conservative versus SNP battle.

In contrast while 24% of Labour supporters say they would switch to the Conservatives in order to defeat the SNP, rather more, 31%, say they vote SNP to keep the Conservatives out. Equally while 35% would back the Liberal Democrats against the SNP, 25% would want to support the SNP.

Even if taken at face value, these numbers do not suggest that tactical voting is likely to deny the SNP more than a handful of seats. To establish what difference it might make I have repeated the exercise I conducted using the equivalent data from YouGov’s previous poll. That is, first I have used the poll to estimate what share of the vote would be won by third and fourth placed parties in each seat that the SNP are projected to win (assuming the change in party support since 2010 in every constituency matches that across Scotland as a whole). Then I have assumed that those backing these third and fourth placed parties would switch to the SNP and their principal opponents in the proportions given above, and calculated the estimated level of support for the top two parties following that switching.

In the absence of tactical switching the YouGov poll implies that the SNP might win 54 seats, Labour four and the Liberal Democrats one. Taking into account the possible impact of tactical voting as outlined above reduces the SNP tally by five, with four seats being saved by Labour while the Tories now hang on (just) to Dumfriesshire. This estimate is similar to the one of seven seats that was derived from applying the same procedure to YouGov’s previous poll.

Moreover, as on that occasion, in some seats this estimate implies making some heroic assumptions about the willingness of the supporters of parties that are now reckoned to be in third place locally to accept that their party has no chance of winning. In particular it assumes that Conservatives in East Renfrewshire and Liberal Democrats in Dunfermline accept that their local candidate’s cause is now a forlorn one.

Meanwhile Lord Ashcroft’s additional polling in Edinburgh provides Labour with little comfort either. True, Labour’s vote is estimated to be holding up better in both Edinburgh South (down just one point) and in Edinburgh North (down 9 points) than it is across the country as a whole.  This, however, is not because the SNP advance is any weaker in the capital; the party’s support is estimated to be up by 29 and 33 points respectively, an advance that is still enough to overhaul Labour locally.

The party whose support has collapsed in the two seats is, according to Ashcroft, is the Liberal Democrat party.  Its previously not inconsiderable vote in both seats (the party was a close second in both in 2010) is estimated to have dropped by 26 points in South and 28 in North.  These drops are bigger than both the 14- 15 point drop in the party’s support being registered in Scotland wide polls and the average drop of 15 points registered by Ashcroft in three seats that an incumbent Liberal Democrat is trying to defend. As we noted in our previous blog, it is arithmetically inevitable that the party’s support must be falling by more than 15 points in some seats, and these results suggest that this is most likely to happen where the party was strong but did not actually win in 2010. Such a pattern will limit the party’s prospects when it comes to trying to chart a post-May 7th path to recovery.

Topics: Elections, parties & leaders

About the author

John Curtice is Professor of Politics at Strathclyde University, Senior Research Fellow at ScotCen, and Chief Commentator on the What Scotland Thinks website.