General Election 2001 : North Of The Border

With three polls now in from Scotland, we can take a brief look at the electoral scene north of the border. (Not that you would know there have been any polls in Scotland from reading the London-based newspapers.) As the first post-devolution election, it will be fascinating to see how, if at all, the existence of the Scottish Parliament affects voting patterns, turnout, and the concerns which Scottish voters raise with their candidates.

Of the three polls, only ICM's in the Scotsman was definitely conducted wholly after the election was called; the Scottish Opinion survey in the Daily Record straddled the announcement date, and the System Three poll for the Scotland500 was reported in the Sunday Herald on 13 May only as having been conducted "last week".

The first point to note is that, even though the Scottish Opinion survey deviates widely from the other two in Labour's share of the vote, the impact in seats is comparatively small. Scotland has few marginal seats, and very few where Labour is vulnerable.

Seat projections from published poll results (assuming uniform swing across Scotland since the 1997 election)

*Lib Dem total for 1999 includes two seats for Orkney and Shetland, which comprise only a single seat for Westminster elections

Of course, these seat projections should not be taken as gospel even if the final voting turns out to match the first week's polls. Individual constituencies will behave differently, especially those with well-known candidates. It is hard to see Charles Kennedy losing his seat, for example, even if there is a Lib Dem slump in other parts of Scotland, and Malcolm Rifkind's chances in Pentlands are not necessarily entirely bound up with the fate of his party's other 71 candidates.

The System Three and ICM voting figures agree fairly closely, showing Labour's share lower than in 1997 though higher than in the constituency section of the Scottish Parliament election in 1999. The Nationalists, correspondingly, stand higher than at the last Westminster election though lower than their Holyrood performance.

It has been very clear since 1999 that many voters make a clear distinction in their party preferences between the two Parliaments. The monthly System Three polls for the Herald regularly ask the same sample of Scots how they would vote for Westminster and how for Holyrood; invariably, Labour performs better and the SNP worse for Westminster; the difference is usually at least five points, and has touched double figures. Scots are divided on the relative importance of Westminster and Holyrood (32% told ICM that Holyrood is "more relevant in dealing with issues that affect you", 21% that Westminster is more important and 42% that they are of equal importance), but they don't necessarily accept that devolution has removed Westminster's influence over devolved policy areas such as education and the NHS. (38% think Westminster has most influence over the quality of the NHS and schools in Scotland, only 27% Holyrood, again from the ICM /Scotsman poll.)

And who is to say they are wrong? Government spending in Scotland is tied to the Barnett formula, which governs the allocation of central government money round the UK and gives Scotland more than its per capita share. An NOP poll reported yesterday found more than half the British public think Scottish spending should be brought down to English levels. If Scots want to prevent that, should they vote Labour or SNP?

If the Nationalists could poll their Holyrood strength for Westminster, they would be breathing down Labour's neck in the race for most votes in Scotland -- but in seats they would still struggle to reach double figures and Labour could still hold two-thirds of the seats in Scotland even if level on votes with the SNP. Meanwhile, the Lib Dems will be trying to keep their seats in double figures even while only the fourth-placed party in votes; and the Tories, the majority party in Scotland less than half a century ago, will probably be privately satisfied if they can win a single seat this time.