Lord Ashcroft published another set of his now famous constituency polls – in this case of eight seats in Scotland – on Friday evening. Three of them are Labour held seats that Ashcroft previously polled in January or February this year – East Renrewshire, Glasgow South West and Paisley & Renfrew South. Four are Liberal Democrat seats, Berwickshire, East Dunbartonshire, North East Fife and Ross, Skye & Lochaber, in only the last of which has Ashcroft has polled previously (in late February). The final seat is the Tories’ sole seat north of the border, Dumfriesshire, which the former Tory treasurer also previously polled in February.
The results do nothing to undermine the expectation that the SNP will do extraordinarily well on May 7th. The party is placed ahead in seven of the eight seats, including all three of those being defended by Labour – amongst whose number, of course, are the seat of the current Scottish leader, Jim Murphy, and that of the UK party’s election supremo and Shadow Foreign Secretary, Douglas Alexander. The only seat where the SNP are not placed ahead is in Liberal Democrat Michael Moore’s Berwickshire seat, where the Conservatives are placed just one point ahead in what looks (as one would anticipate) like a close three-way fight.
Indeed one of the notable features of these polls is that, much like the recent YouGov and TNS BMRB Scotland-wide polls, they suggest that support for the SNP has increased yet further. The party’s estimated share of the vote is up in every single one of the five seats in which Ashcroft polled earlier this year. On average the increase is as much as five points. The odds on this being a statistical accident occasioned by the random fluctuations to which all polls are subject are very low. Meanwhile at 30 points the estimated average increase in SNP support across all eight polls is commensurate with a SNP Scotland wide vote of 50%, in line with the 49% and 52% recorded by YouGov and TNS BMRB respectively.
The poll uncovers three other findings that should concern the unionist parties. First, even after people are asked how they will vote in their particular constituency, a form of wording that clearly increases the estimated Liberal Democrat vote in the four constituencies they are defending, the Liberal Democrat vote is down on average by 15 points, exactly in line with the 15 point drop in the party’s Scotland-wide share of the vote. There is no sign here of the local popularity of the party’s MPs enabling them to stem the outgoing tide.
We should not be surprised. There are plenty of constituencies in Scotland where the party won less than 15% of the vote in 2010, and where its vote thus cannot possibly decline by as much as 15 points. The three Labour seats in which Ashcroft polled all fall into that category and in these the average decline in the party’s vote according to Ashcroft is just seven points. The Liberal Democrats must be losing votes more heavily in some places where they were previously stronger. All that the local popularity of their incumbent MPs seems to be doing is counterbalancing this inevitable pattern.
A similar pattern is evident in Labour’s case too. In the five seats polled by Ashcroft that the party is not defending, its estimated vote is down on average by just nine points. In the three seats that the party is defending its vote is down by as much as 23 points. The only difference is that there is little sign in Ashcroft’s polling that in the Labour held seats inviting voters to think about how they will vote in their own constituency serves to reduce the estimated SNP lead.
Finally, this latest round of Ashcroft polling serves to underline how difficult it will be to promote effective anti-SNP tactical voting in this election. In three seats the estimated SNP vote is 48% or more, while in two others the party’s percentage lead is in double figures. Under these circumstances it is at least very difficult and in some cases impossible for tactical voting to make a difference. Where it could make a difference is where there is a substantial third party vote for one of the unionist parties, such as the 25% with which the Conservatives are credited in East Renfrewshire, the 28% the Liberal Democrats are estimated to have in Berwickshire, and the 20% Labour tally in Dumfriesshire. But it seems unlikely that Liberal Democrat voters will accept that their man has no chance of winning in Berwickshire (after all Ashcroft shows the local contest to be a very tight three party fight), while Conservatives in East Renfrewshire and Labour supporters in Dumfriesshire may be reluctant to accept that their party has no chance of winning a constituency in which their party was until recently at least quite competitive.
Still, it seems that there is one limitation to the SNP surge. Despite the fabled strength of its party machine and its greatly increased membership numbers, it appears that it is being outnumbered on the ground. In each of the eight seats voters are more likely to say they have received a leaflet or a call from the party defending the seat locally than they are to say they have heard from the SNP. Faced with the challenge, unlike their opponents, of fighting everywhere, the party is, perhaps, having to spread its resources rather thinly. It remains to be seen whether this will make a difference to its seat tally on polling day.