What do local by elections tell us?

Thursday’s local by-election results were among the worst that I can remember for the Conservatives. The Party lost five out of seven seats it was defending, and saw its vote share fall sharply almost everywhere. By contrast, Labour and the Liberal Democrats performed extremely well, making a net gain of four and two, respectively. Naturally, both parties’ activists have been heartened by this, and Conservatives, judging by the blogs, have been perplexed.

Currently, opinion polls have been giving the Conservatives leads that range from 8% to 17%. Do these local results, being actual, as opposed to notional, votes, prove that the polls are wrong? Almost certainly not.

To begin with, one has to look at the parties’ starting points. The Conservatives hold as many local council seats as Labour and the Liberal Democrats combined. The local elections of 2006 to 2009 gave the Conservatives huge leads, in terms of vote share, and huge gains in terms of seats. They were not just winning the marginal seats, but chipping away at their opponents’ bedrock. It is hard to see what else was left for the Conservatives to gain, unless they were to stage an unlikely comeback in authorities like Sheffield and Liverpool. In each of those years, the Conservatives had leads, in terms of projected vote share, of 14% to 20%; the Conservatives could still enjoy a lead in double figures, and be losing large numbers of local council seats.

As a result, the Conservatives are defending many more seats than either of their main opponents, in local by-elections. Indeed, since the last round of local elections, on 4th June, the Conservatives have had to defend 42 seats, compared to 11 for Labour, 16 for the Liberal Democrats, and 16 for Independents, and smaller parties. As a result, they have a good deal more that they can potentially lose than either of their principal opponents.

Nevertheless, the Conservatives have lost heavily over this period. They have suffered a net loss of 15 seats, while Labour has made a net gain of seven, and the Liberal Democrats, a net gain of six. And, this is at a time when opinion polls have shown a large Conservative lead. So does this matter?

The answer is no. I’m stating the obvious, but local by-elections, over the course of several months, will give a good indication of the parties’ standing in local elections, but will give no indication of the parties’ standing in the coming general election.

By way of comparison, from May 1996, up till the general election, the Conservatives made a net gain of 24 seats in local by-elections. Between May 2000 and the 2001 general election, they also did well, with a net gain of 26 seats, according to Keith Edkins’ archive (www.gwydir.demon.co.uk) But, in both general elections, they were heavily defeated. And this should cause no surprise.

Local elections (and particularly, local by-elections) are heavily influenced by local issues, as well as the intervention, or withdrawal of minor parties and independents from contests. General elections are decided on national issues, and while support for the minor parties is growing, it remains marginal in most constituencies. When it comes to predicting the winner of a general election, look to the national opinion polls, not local by-elections.

Do these local by-elections tell us anything?

I think they indicate that were there to be a stand-alone round of local elections, next year, the Conservatives would lose a significant number of seats, even if they had a fair-sized lead, in terms of vote share. However, the local elections will probably be held on the same day as the general election, which should enable the Conservatives to do better than one might expect in the local contests. Alternatively, if Gordon Brown goes for a March election, and (as expected) loses, the Conservatives may well make gains in local elections held a few weeks later, as Labour voters are likely to be demoralised.

However, the Conservatives are past their peak, in terms of local council contests. Assuming that they win the next general election, they will almost certainly suffer big seat losses in 2011, 2012, and 2013, even if their vote share is reasonable for an incumbent government.

Sean Fear used to write a regular feature for the site on Friday evenings. Now he makes the occasional guest contribution. Long-time PBers might recall that in the 2004/2005 period Sean was just about the only Tory who regularly posted.