Election watchers and political anoraks have been spending much of the day absorbing the eagerly-awaited Scottish constituency opinion polls from Lord Ashcroft. The SNP have been polling very strongly in national Westminster opinion polls since the failed independence referendum, and have consistently polled leads over Labour of over 20%. The big question for those trying to make sense of the upcoming election has been “Do these strong national polls for the SNP translate to possible gains in a first past the post Westminster election?” The answer has proved to be an emphatic “Yes”.
Sixteen constituencies were polled; fourteen where Labour are defending the seat, and two where the Liberal Democrats are the incumbents. All except one, Glasgow North East, show the SNP first when it comes to voting intention. These are some seriously out-there swings; for instance Dundee West, which Labour won handily in 2010 over the SNP by 48% to 29%, is now showing the SNP leading Labour 59% to 25%, a 27% Labour to SNP swing. That is an astonishing change in sentiment.
The question then becomes to what extent the polls in these particular constituencies can be extrapolated to Scotland as a whole. In a previous UK-wide forecast, I had taken Holyrood results and shifted the results to reflect the changing fortunes of the parties in Scotland since the last Holyrood poll. I was able to compare the figures from the Lord Ashcroft polls to my forecast numbers from adjusting the Holyrood results, to see if the numbers were comparable. They actually turned out to be reasonably close to each other. On average, my forecast was 1.3% lower than the Ashcroft polls for the Conservatives, 0.5% higher than the Ashcroft polls for Labour, 3% higher for the Liberal Democrats and 2% lower for the SNP. I shifted my forecast numbers for all 59 seats up or down for each party by the variance between my forecast numbers and the Ashcroft polls to correct the forecast model to be closer to the polls.
Finally, I decided to give each MP defending his or her seat a 3% bump. MPs defending their seats tend to receive a bump between the polls and their actual election results, and the 3% figure was suggested by FiveThirtyEight for UK elections.
Finally, to the results. In Scotland in 2010, the electoral map looked like this.
Labour were the largest party on 41, followed by the Liberal Democrats on 11, then the SNP on 6, and the Conservatives had one seat. Using the Ashcroft poll adjusted, Holyrood based forecast, then the map for 2015 looks like this.
The SNP would have 50 seats, Labour would be down to 6, the Conservatives would lose one but gain two to leave them with two seats, and the Liberal Democrats would be left with one MP, in Orkney and Shetland.
Were this scenario to happen, it would make it essentially impossible for Labour to have a majority in the House of Commons, unless there is a significant surge in Labour support in England between now and the election. Anything can, of course, happen between now and then. But if Lord Ashcroft’s polls do, in fact, turn out to be close to reality, then it may turn out to be very difficult for Labour to form a majority government ever again. The two party system in the UK would effectively be dead.