What will happen in the 2007 local elections?

by Stephen Tall on April 3, 2007

About this time last year, I wrote an article for PoliticalBetting.com, attempting to predict the parties’ shares of the votes based on their ICM opinion poll ratings. And, if I say so myself, I didn’t do too badly.

I guesstimated Con 39%, Lab 25%, Lib Dem 29% – but noted in the comments thread that followed: “my instincts are that Lab and Lib Dems will be slightly closer together than my extrapolated figures”. And so it proved.

The BBC’s projected share of the national vote was: Con 40%, Lab 26%, Lib Dem 27%. Respected psephologists Colin Rallings and Michael Thrasher projected: Con 39%, Lab 26%, Lib Dems 25%.

So I’ve updated my workings from last year to see if it points towards this year’s result…

The figures below show the projected national share of the parties’ votes in that year’s local elections compared with the ICM poll rating (in brackets) immediately prior to those elections:

1998: Con 33% (31%), Lab 37% (48%), Lib Dem 25% (16%)
1999: Con 34% (28%), Lab 36% (50%), Lib Dem 25% (17%)
2000: Con 38% (32%), Lab 30% (45%), Lib Dem 26% (15%)
2002: Con 34% (29%), Lab 33% (45%), Lib Dem 25% (18%)
2003: Con 35% (30%), Lab 30% (42%), Lib Dem 27% (21%)
2004: Con 38% (33%), Lab 26% (38%), Lib Dem 29% (22%)
2006: Con 39% (34%), Lab 26% (32%), Lib Dem 25% (24%)

The Tories have added between 2-6% to their ICM national rating in recent local elections; Labour have dropped 6-15%; and the Lib Dems have increased 1-11%.

If this pattern were to repeat itself in 2007 – using the most recent ICM poll (19th March), with the Tories on 41%, Labour 31% and the Lib Dems 18%, as our bench-mark – we might extrapolate the following shares of the vote this May: Con 45%, Lab 21%, Lib Dem 24%.

Hmmm, well I don’t know about you, but I’ve not convinced me. Despite Tony Blair’s best efforts in the last six months to destroy the party he helped re-build by clinging onto power far beyond his political utility, I cannot see Labour’s vote dipping to 21% this year.

Nor can I see the Tories scaling the dizzy heights of 45% – which would exceed what Tony Blair managed in 1996 – though I may, of course, be proved wrong.

My wet-finger-in-the-air projection gives: Con 42%, Lab 25%, Lib Dem 25%

Seems plausible to me. We’ll see.

Incidentally, the final half of my original article stands the test of time (I think) – and even bears repetition:


The stubborn refusal of the third party to lie down and quietly die points to a wider trend: what has been termed the de-alignment of British politics.

If we look at the combined, average general election vote shares of the two big beasts of post-1945 politics, the Tories and Labour, and compare them with the combined, average vote shares of the Lib Dems and other parties for each of the last five decades, the fragmentation of voter loyalties is clear:

1950s: Con/Lab 92%, Lib Dem/Other 8%
1960s: Con/Lab 89%, Lib Dem/Other 11%
1970s: Con/Lab 80%, Lib Dem/Other 20%
1980s: Con/Lab 72%, Lib Dem/Other 28%
1990s: Con/Lab 75%, Lib Dem/Other 25%
2000s: Con/Lab 70%, Lib Dem/Other 30%

Voters no longer identify tribally with one political party based on their self-perception of class or religious interest (or their parents’ views). In 1964, according to the British Election Study, 48% of Tory voters identified strongly with their chosen party, compared with 51% for Labour. By 2001, the figures were 14% and 16% respectively.

As voter turn-out has declined, transforming the electorate into a selectorate, the remorseless march of the de-alignment process has continued apace. The cosy Tory/Labour duopoly is coming to an end.

And, however much Messrs Brown and Cameron might prefer to ignore such a reality, this is the new politics with which all parties are going to have to deal.