Rochester & Strood – Analysis of Polls and Parties by Survation’s Patrick Brione
The Rochester & Strood by-election is over and UKIP have their second elected MP in Westminster. The winning margin, 7.3% or 2,920 votes, is less than some commentators had been predicting, in part based on polling snapshots during the campaign, but to put it in context this is one of the largest swings in by-election history, with UKIP moving from 0% of the vote in 2010 to over 42% in this by-election. This despite the heaviest campaigning effort the Conservative Party has ever conducted against UKIP with no fewer than five campaigning visits from the Prime Minister in person, over 100 Conservative MPs – many making multiple visits including most cabinet ministers.
Comparing the Polls
Looking at how the polls have fared, there were four published by-election polls during the course of the campaign, two from Survation (one for the Mail on Sunday, one for Unite the Union) and one from ComRes during October, and then one from Lord Ashcroft from 7th – 10th November. The results from each poll are outlined in the table below.
The first thing to be noted is that all four polls showed UKIP ahead, which clearly contributed to the transformation in expectations from a likely Conservative victory at the time of the Conservative Party Conference, to a widely-expected UKIP triumph by the end of the campaign.
The fact that the final UKIP margin of victory was less than that shown in all four polls has led some commentators attempting to portray the result as some sort of success for the Conservative Party, but of course had there been no polling published during the campaign period, today would almost certainly be being viewed as a devastating blow to originally hopeful Conservative plans for defeating Mark Reckless and deterring further defections.
In terms of how accurate the polls were, it should be remembered at all times that, in the words of Lord Ashcroft, polls represent “a snapshot, not a prediction”. No polls were conducted during the final week of the campaign, so none of the polls will have captured any late movement as a result of the very heavy campaigning efforts going on in the seat.
Lord Ashcroft’s own poll, conducted closest to the actual vote, gave the closest result with a 5 point deviation from the final result across the main three parties, compared with 8.4 for the late October Survation poll, 10 points for the ComRes poll and 14 points for the first Survation poll. In other words, the polls all converged towards the final result as the campaign progressed, with the earliest poll being farthest from the result and the later polls being the closest.
This suggests that the polls were accurately capturing real shifts that were going on during the campaign. In particular they show a diminishing Labour vote share, from the mid-20s at the start of the campaign down to the 16-17% the party achieved at the end, as Labour voters were squeezed tactically by the heavy UKIP and Conservative campaigning into voting for whichever of the two main contenders they disliked the least.
Consequences for UKIP?
In terms of the implications for the four major parties (or five – as the Green Party had a sucessful by-election), for UKIP the result is a truly compelling success. Whilst Clacton was easy for some to write off as an outlier – a uniquely popular MP winning in a uniquely elderly, socio-economically deprived constituency – Rochester & Strood represents winning in something approaching an archetype of “middle England”. If an MP defecting to UKIP can win a by-election in Rochester & Strood, then similar defections elsewhere could potentially succeed across a very wide range of English constituencies.
There are 164 constituencies where the Conservatives won in 2010 on a lower share of the vote than they did in Rochester & Strood. MPs in any of these places who are considering defection to UKIP must now reckon they have at least a reasonably good chance of retaining their seats. Meanwhile Conservative incumbents in many seats must worry that they risk losing either to Labour or, increasingly, directly to UKIP as a result of the continued UKIP success. The message “vote UKIP, get Labour” has less traction than ever when election results like this show it is genuinely possible to “vote UKIP, get UKIP”.
UKIP’s Track Record In By-Elections
Looking at the wider picture, UKIP have now come in the top two parties in 9 out of the last 10 by-elections in Great Britain. This is compared to 4 for the Conservative Party and 6 for Labour. In other words, even where they are not winning, UKIP are positioning themselves as the main party of opposition to whoever is the incumbent in a very wide range of places.
Other Winners and Losers
The only other party who could be said to have done at all well out of the by-election would be the Greens, who at 4.2% recorded their best result since the 2010 general election, in a seat where they scored only 1.5% at the last election.
Significance For The Lib Dems?
The Liberal Democrats lost their deposit in fifth place with under 1% of the vote; their worst ever by-election result. Nevertheless, the LibDems will not be overly concerned with this result in a place they had never had any intention of significantly contesting. Liberal Democrat voters are considered to be the most likely of all voters to vote tactically in elections – something that cost the LibDems votes in Rochester but will help them win votes elsewhere. Their attention and resources are entirely focussed on the key seats that they will realistically contest next May and the outcome in Rochester has virtually few to no implications for those places.
Labour too appear to have made a decision at an early stage not to spend central resources campaigning in this by-election, seemingly content to allow UKIP to squeeze their votes in order to deal an embarrassing defeat to the Conservatives. This means that Labour’s low vote share in Rochester can not be extrapolated to those seats that are actually on Labour’s target list. On the other hand, their decision not to contest the by-election in the first place does seem to betray a certain lack of ambition – this was a seat which Labour held throughout the New Labour era and which a Labour Party with hopes to rebuild a majority that reaches beyond its Northern and London heartlands would have wanted to fight for.
A Very Bad Day for David Cameron, Lynton Crosby and Grant Shapps
For the Conservatives, though, this result is a particular tactical embarrassment. From the moment of Mark Reckless’ defection they had announced their intention to throw the proverbial kitchen sink at the constituency, to punish Reckless for his decision to defect and warn off other potential defections.
Amidst a celebratory mood at the party conference, the party confidently predicted success in Rochester. All MPs were instructed to visit the constituency at least three times and over a hundred MPs did so on multiple occasions; David Cameron himself made 5 visits. The party had shipped in activists from across the country to bombard the constituency with canvassing – their resources extended as far as being able to pull off a logistically impressive 6am mail drop to households across the seat on the morning of the vote. They even ran a high-profile “open primary” to select their candidate, hoping to generate extra publicity and rally public support around a democratically chosen, popular local candidate.
Despite all of this, the party failed to retain a seat with a 10,000 vote majority, against an opposition party which had not even fielded a candidate at the previous election and had no previous representation on the local council. The implications for the Conservatives are potentially dire. There now remains the overshadowing possibility of further defections to UKIP. Even the mere potential of such defections is enough to sap party morale and potentially lead to suspicion and paranoia driving a still further wedge between the leadership and Eurosceptic backbenchers.
General Election – There May be Trouble Ahead
In May 2015, Conservative resources that were so focussed on Rochester will be spread across the whole country – they will not be able to mount the kind of defensive rearguard actions against UKIP that they attempted in Rochester, whilst simultaneously contesting battleground seats with Labour across England. UKIP, meanwhile, will be able to focus their increasingly professional resources on a much smaller list of seats, in many cases in places where the incumbents are unused to having to fight hard for re-election. After today, it must be conceded by even their most ardent sceptics that UKIP have genuine prospects of success in 2015.
-by Patrick Briône
Director of Research, Survation