The Labour party has at least a year to prepare for the next election, and there is no shortage of people offering advice on which direction it should turn. Some have suggested that Labour need to pay more attention to satisfying its core, working-class vote. There may well be merit in that, but any such strategy needs to look at the evidence on where the election will be fought and lost. And on such a reading, one thing is clear: Labour’s majority depends on the south.
At the last election Labour won with a majority of 66, but the boundary changes have reduced it to around 48. That means 24 losses would deprive the party of its majority. It’s a quite different scenario from the 2001 and 2005 elections, when losing David Lock in Wyre Forest or Melanie Johnson in Welwyn Hatfield was sad, but of little consequence to the survival of the government. Now every seat is crucial.
An analysis of where the marginals lie shows that the battleground if Labour is to hold an overall majority is in the south of England – the swathe of seats in north Kent (Gillingham & Rainham and Dartford, plus three seats notionally reassigned to the Tories under boundary changes – Sittingbourne & Sheppey, Rochester & Strood and Thanet South) which the party unexpectedly gained in 1997 and have miraculously held since; the south coast seats (Hastings & Rye, Hove, Portsmouth North, Dorset South); five seats in London (Croydon Central, Battersea, Hampstead & Kilburn, Islington South & Finsbury, Ealing Central & Acton); and a scattering elsewhere in the south (Harlow, Crawley, Basildon South, Stroud, Watford, Milton Keynes North and South, Oxford East).
Altogether 18 of those first 24 seats are in the south. So are several of the seats currently held by Labour which would now be notionally Conservative on the new boundaries (the three aforementioned Kent seats, plus Somerset NE, Finchley & Golders Green, and Enfield North); strictly speaking these don’t count as Labour-held marginals, because they are already assigned to the Conservatives in the calculations of the majority – but try telling that to the incumbents and activists.
What do these seats have in common? Relatively low levels of trade union membership; relatively low council house occupancy; relatively short and shallow traditions of electing Labour politicians. On the old boundaries Battersea is ranked first in the entire country for the proportion of people in managerial level jobs – the highest classification. Crawley was 469th on employment in manufacturing. Gillingham, the number one Tory target, was 402nd – and with 77.9 per cent owner occupation, ranks 92nd in the country on that score. Almost all these marginals are now in Conservative or Lib Dem-controlled councils.
If Labour wants to win these seats, it will need to reach beyond the core vote. Doing more on council housing or trade union rights might or might not add to Jon Cruddas’ six and a half thousand majority in Dagenham, but that won’t be where the election is decided. If we visualise a core Labour voter as being a blue-collar trade unionist, living in a council house, the bad news for Labour is that such people make up considerably less than a tenth of the population, and given their concentration in the Labour heartlands, even less in the marginals. Think again.
There are other battlegrounds, of course. Even if another majority is beyond Labour’s grasp, it would still hope to be the largest party. The seats which take Labour from almost certainly being able to form a minority administration (around 310) to no longer being the largest party (around 290) are disproportionately in the Midlands: Northampton South, High Peak, Edgbaston, Loughborough, Stafford, Broxtowe, Burton, Redditch, Rugby, Tamworth, Derbyshire South, and Worcester. Again, the demographic news is not good for those who want to see a core vote strategy pursued. While some of these seats do have some characteristics which traditionally favour Labour – leafy Edgbaston has well above average council housing – most look solidly average English seats: and these days that means owner occupation up around the 70 per cent range, with small minorities in manual occupations.
It is only if Labour is reduced to hoping to hang on to enough seats to deprive Cameron of a majority that the battleground shifts to the north. But even here, the marginal seats are not necessarily in the most working-class areas. Blackpool North, for example, is 76.1 per cent owner-occupied – ranked 156. Wirral South is 82.6 per cent owner-occupied – 22nd in the entire UK – and is comfortably in the top 200 seats on most measures of social class.
Elections go on every year, of course, in local government. It’s a little dangerous to read across from locals to a general election, given the far lower turnout, but we can learn something from relative regional shifts at the local elections. If we take as a given that the south is particularly important in Labour’s efforts to retain power, the news is again not good: the drop in Labour’s vote at the 2006 local elections was much sharper in the south than the Midlands, and it held up best of all in the north. This was less apparent in 2007, although that may have been because Labour was close to rock-bottom by then.
This effect is further corroborated by regional opinion polling. In August 2007 – when, conveniently, Labour held a 3 per cent lead over the Conservatives, the same as at the 2005 general election – ICM for the Guardian did an analysis of all their polls since Gordon Brown had become leader, as against their 2005 campaign polls (giving a very respectable sample size). This showed that a 0 per cent national swing masked a 4 per cent swing to the Conservatives in the south-east and south-west government regions. Labour’s position had strengthened elsewhere to compensate for this – but that is not where the ultra-marginals are.
None of this is likely to be an adverse reaction to Gordon Brown – he can hardly be held responsible for bad local election results when Tony Blair was leader. It seems much more likely that southerners (and to a lesser extent Midlanders) were positively attracted to the David Cameron approach. If the north continues to resist that message, it may at least keep the Conservatives from winning a majority, although that may appear cold comfort.
Elections in May will tell us more. As annual crops of wards coming up for election go, 2008 is not a particularly big one, but there will be enough in each region of England to get a good idea of further regional variation. London is the big prize up for grabs, but win or lose, Labour should be wary of reading anything into Ken’s performance because people really do vote very differently (as they were always supposed to) in a mayoral election than a traditional one. Indications from the last borough results (2006) suggest that the picture in London is if anything even worse than the rest of the south, however.
The upshot is that all Labour’s destiny as the party of government rests on voters in the south – and that the signs are that since the 2005 general election the party has not done enough to please these people. To say voters in the south are more middle class than voters elsewhere is slightly misleading: they probably work in an office rather than a factory and they probably own their own homes, but by those criteria they are only slightly more affected by the modern world than everybody else. By all means Labour should do right by the most disadvantaged, but Britain has changed in the last 40 years, and the coalition needed to win an election has changed with it.