This contest has been billed as UKIP’s chance to have its first MP elected. I think this is unlikely. Here’s the 2010 General Election result for Newark.
But before I explain in more detail why a UKIP victory is unlikely, let us give a bit of background about the constituency.
This is a largely rural constituency. The town of Newark-on-Trent, plus its southern suburb of Balderton, comprises about a third of its population, with the rest scattered along the A1 and River Trent along the eastern side of Nottinghamshire. It includes the small market towns of Southwell and Bingham and villages such as Sutton-on-Trent, Tuxford, Collingham and North Muskham.
The most northern ward, Rampton, is famous for Rampton Secure Hospital, with inmates including child murderer Beverley Allitt.
The majority of the seat is within the Newark and Sherwood district, but its southern five wards (around Bingham) are in the borough of Rushcliffe whilst the northern three wards are in Bassetlaw district.
Newark once a regularly voted Labour: they held the constituency between 1950 and 1979, losing it to Conservative Richard Alexander that year. But in 1983, Labour voting mining areas in the centre of Nottinghamshire were removed through the creation of the Sherwood constituency, making it much easier for Alexander to win in the next three elections, before succumbing to the 1997 Labour landslide in one of the more unexpected results that year.
Labour’s victor, Fiona Jones, was not well-fated. The third place Liberal Democrats complained that Jones’ campaign had broken election spending limits and lied on her expenses submission. Most of the claims were quickly dismissed, but the charge that the rent on the constituency party offices had not been included led to her and her agent to become the first people to be convicted of corrupt practices under the Representation of the People Act 1983.
Her prosecution was a bit bizarre: if she was guilty of this she was certainly not the only candidate to be so: I would estimate that the number of candidates who have over-spent election expenses once you throw everything in including constituency office rent could run into the hundreds since 1983, particularly as the limits established have become seen as ridiculously low (currently around £12k once a candidate is formally announced), but she remains the only person ever convicted. So her prosecution was fairly perverse and was overturned on appeal a month later.
But the damage was done and she lost in 2001 to Patrick Mercer. Jones descended into the alcoholism that killed her six years later.
Labour cannot win back Newark barring a freak landslide. The boundary changes in 2010 were relatively minor over all but harmed Labour here: the Labour inclined town of Retford was transferred northwards to the safe Labour Bassetlaw constituency whilst the Rushcliffe territory around Bingham was added.
Labour took 48% of the vote across Retford this year, to the Conservatives’ 26% and UKIP’s 23%: an edge of 1,300 votes. In contrast, the last elections in Rushcliffe (2011) had a Conservative vote of 47% with Labour on 28%, a Conservative lead of 1,500 votes. That is a net change in the Conservatives’ direction of 2,800 on a lower local election turnout. It is likely that this difference would have been enough to prevent Jones winning in 1997, given her 3,016 majority then.
There is little evidence to show that the ground has shifted enough for UKIP to win through here, although there current media coverage pretty much guarantees them a strong second place.
First, let us have a look at the political landscape in the constituency. This is the current state of play of council seats here: a sea of blue. Labour hold one ward in Newark-upon-Trent itself and are competitive in other wards in the town, the Lib Dems have seats in Southwell and Sutton-on-Trent. But there is a conspicuous absence of UKIP purple anywhere.
Most of those elections (other than in Bassetlaw, which held elections for a third of its seats this year) were back in 2011. But last year’s county council elections would draw you to a similar conclusion: whilst UKIP won 18 seats in neighbouring Lincolnshire, they won none here or anywhere else in Nottinghamshire. And whilst Labour won control of the county, they won nothing within the boundaries of the constituency.
The chart below aggregates the votes per party across the county council seats that are wholly or partially within the constituency.
The polling for the by-election seems to bear this out. Last week, Survation published a poll for The Sun for the by-election (brackets show change since the general election):
Con: 36% (-18%)
UKIP: 28% (+24%)
Lab: 27% (+5%)
LD: 5% (-15%)
So, a massive improvement by UKIP but still 8% behind the Conservatives and pretty much tied by Labour.
A similar position, with a larger Tory lead, emerged from a poll published by Lord Ashcroft this Monday:
Con: 42% (-12%)
UKIP: 27% (+23%)
Lab: 20% (-2%)
LD: 6% (-14%)
I could be wrong (as I was with Bradford West) but it is difficult to see how UKIP can win today. For them to break through, they needed some big campaign incident to shift the ground further since the European elections two weeks ago and that has not happened.
It looks as if Farage was wise not to throw his hat in the ring and save himself for Thanet South.