Ukip gained a strong second place after the brief and intensive Eastleigh by-election campaign. The party has since claimed that if the campaign had lasted longer, they would have won the by-election. Does the evidence support this claim?
Ukip had momentum leading up to the ballot. Opinion polls and reports from the ground suggest that Ukip were gaining supporters during the campaign, that the LibDem vote was staying roughly constant and that Labour and the Conservatives were losing support. Ukip had a credible candidate, who was a defector from the Conservative Party and able to empathise with Conservative-inclined voters.
The result was that the LibDems won with a majority of 1,771 over Ukip in second place. To have won the by-election, Ukip would have needed to have picked up the support of at least 1,772 people who did not vote or who voted Conservative, Labour or for one of the minor party candidates, or alternatively to have converted half this number of people (886) who voted for the LibDems. Was either of these, or a combination of the two, likely?
The turnout at 52.8% was high for a by-election, but lower than the 69.3% who voted at the last general election. 41,616 people voted in the by-election, compared to 53,650 who voted at the general election, meaning that 12,034 fewer people voted in the by-election. Potentially there was a pool of non-voters, some of whom might have been persuaded to vote Ukip. If the turnout had increased at all, the momentum of the contest suggests that a disproportionate percentage might have voted Ukip, as a newer, less well-known competitor. However, it would have been almost impossible for even a troglodyte elector in Eastleigh not to have been aware of the by-election and, indeed, many locals were becoming fed up with the invasion of the constituency by outside politicians and activists. A longer campaign may well have resulted in voter fatigue and a lower turnout, rather than an overall increase.
So, probably a minimal number of non-voters could have been tempted by Ukip to the ballot box. But, what of the possibility of converting some of the other parties’ supporters? To erode the LibDems’ majority entirely, Ukip would have needed to convert 886 LibDem supporters. This seems very unlikely as the LibDem vote had already fallen from 24,966 in the general election to 13,342. For LibDem supporters angry at the party’s scandals, their protest in many cases will have been abstention. The remaining LibDem voters had stuck with the party through thick and thin and were probably almost entirely loyal. If they had switched parties, it would have been unlikely that many of the remaining hardcore LibDem supporters would have swapped to Ukip, as a result of a longer campaign.
What about the possibility of finding the 1,772 votes needed among Labour supporters? This also seems very unlikely as the Labour vote was already squeezed down to a loyal core of under 10% of the voters. Labour support showed no flicker of increase, even after Andrew Rawnsley’s rather wild suggestion in the Observer that Labour might win the by-election.
Likewise, the 2,056 voters in total for the minor parties will mainly have had their own reasons to support candidates who almost certainly had no prospect of winning. There was probably little potential for converting many of these to Ukip.
This leaves the potential for Ukip to have picked up the 1,772 voters, or at least most of them, from the Conservatives. This was a big ask, involving enticing 16.8% of the Conservative voters over to Ukip. Although this seems rather unlikely, it is not impossible to imagine. The Conservatives were losing support during the campaign, but the opinion polls still put the party in contention to win. Had the by-election campaign gone on long enough for opinion polls to show that the Conservatives were not going to win, but that Ukip might, then possibly a significant number of supporters may have switched in order to vote tactically to keep the Lib Dems out.
Ukip’s claim, although very unlikely (especially as many people voted early by post), is not totally implausible – and in many ways more credible than predictions that Labour might win, or that the LibDems faced annihilation.