For all the huffing and puffing in the House of Lords over the Parliamentary Voting System and Constituencies Bill and the possible replacement of first past post with the alternative vote, a switch to the latter would only make a small difference to the outcome of the next general election. So Labour should shun defeatist talk of a new era of hung parliaments and coalition governments and focus instead on an alternative electoral strategy designed to win an overall majority in the House of Commons.
While Labour’s vote has fallen and fragmented, all is far from lost. The party can win again, providing it learns from the past and recognises how and where voting behaviour has changed. In the Smith Institute’s Winning Back the Five Million – Understanding the Fragmentation of Labour’s Vote, our head of research, Paul Hunter, shows that, with the right strategy, the electoral arithmetic can work for Labour.
Electoral tactics alone cannot turn the tide, but a closer look at the polling numbers reveals there is no correlation between the dispersal of Labour’s vote since 1997 and the likelihood of more hung parliaments in the future. A study of the voting patterns indicates there is little evidence to support the “southern discomfort” thesis that Labour lost key seats in the south of England. Further, the election data shows there is no longer a working-class core vote for Labour and that the party must broaden its appeal beyond the inner cities to suburbia.
In 1997, Labour swept to an unprecedented victory, gaining a record 419 seats. Thirteen years later, the party sank to its worst election result since the hammering it got in 1983 – and, barring that, since the advent of universal suffrage. Between 1997 and 2010, Labour lost a staggering 4.9 million votes – one third of its support. The electoral coalition it had built disappeared, with just 29 per cent of those voting giving their support to Labour last year and a mere 19 per cent of the total electorate turning out to put a cross next to the local Labour candidate’s name.
This is a grim picture, but the vote that Labour secured in 1997 did not disappear overnight. It was eroded over time. Rather than a straight swing to the Tories, a rather more complicated fragmentation occurred.
Some of Labour’s loss can be explained by a drop in turnout – 1.6 million people fewer voted in 2010 than 1997. However, other parties picked up more votes. The biggest winners in absolute terms over the three elections since 1997 were the Liberal Democrats. They secured a six-percentage-point increase of the vote share and a 30 per cent increase in their own vote. Their support has since collapsed and may now struggle to reach half what it was in 2010.
For Labour to win again, it will need to repair its dissipated electoral coalition and capture the “progressive majority”. The problem is that any simple targeting of the Lib Dems to gain votes doesn’t add up. While the Lib Dems gained 1.5 million votes since 1997, over the same period, some 2.6 million votes went firmly rightwards – to the Tories, the UK Independence Party and the British National Party. There are simply not enough Lib Dem votes in marginal seats for Labour to win a outright majority.
The Conservatives were disappointed with the 2010 election result. Traditionally, they would have advanced where Labour lost. They actually gained just 1.1 million votes, representing around a five-percentage-point increase. Since 1997, the Conservatives have increased their vote by only 11 per cent.
It is in terms of seats more than votes that Labour was soundly beaten last year. Of the 94 seats Labour lost, 87 were won
by the Conservatives.
Now Labour’s electoral strategy should recognise that the Tories almost certainly gained more from Labour than the figures imply. That’s because some erstwhile Tory supporters switched to UKIP and the BNP. Indeed, both Labour and the Conservatives were hurt by the dramatic rise of the fringe parties. Together, these took over 1.5 million more votes in 2010 than 1997. They may remain minor parties, but their vote share has moved from a paltry 0.65 per cent to a significant 6 per cent.
Much of the present debate about electoral reform centres on the seemingly ineluctable pluralisation of voting patterns, leading to more hung parliaments in future. Supporters of this thesis explain that, over recent years, a smaller and smaller number of seats are now marginal ones. The growth of the Lib Dems and the nationalists in Scotland and Wales means that increasing numbers of constituencies are not in the hands of the two main parties, while the proportion of votes cast for both Labour and the Conservatives is now lower than any time since 1918.
Under first past the post, it is the way seats divide up between the main parties that determines the result of elections. Certainly, votes for Labour and the Tories have fallen from around 97 per cent to 67 per cent. However, Labour and the Tories combined shared 87 per cent of the seats at the 2010 election – roughly the same as 2005, 2001 and 1997, and only 6 per cent below what it has been for any general election since 1974. It would be wrong to assume there is a correlation between the two parties’ share of the vote and the size of majorities. What matters is the Labour-Tory share of the seats.
Current opinion polls put the combined standing of the two main parties back to more than 80 per cent. They also show huge unpopularity with the Lib Dems’ decision to form a coalition, which will probably result in far fewer Lib Dem seats next time. Unless their ratings soar back to 20 per cent-plus, it is difficult to see how Nick Clegg can even contemplate breaking the two-party mould.
The Lib Dems are extremely unlikely to garner as many votes at the next election as they did in 2010. On paper, that should widen the scope for Labour gains. But a Lib Dem collapse does not quite translate into a Labour victory, although it extends the number of seats the latter can contest on any meaningful basis. And that makes another hung parliament less likely. This becomes significant if we look beyond the traditional marginal constituencies to seats where the two-party split is between 40-60 per cent. In these “extended marginals”, the Conservatives hold 90 seats. Labour would need an 8 per cent swing from the Tories to win all of them. A 6.5 per cent swing ought to secure a working majority.
If the Lib Dem vote does collapse, Labour could gain as many as 30 seats from the Conservatives. That would not be enough. However, if Labour gains relatively more from the collapse in the ordinary and extended marginals than the Tories, then a much smaller swing (votes taken straight from the Conservatives) is required – maybe as low as a 4 per cent (rather than 6.5 per cent). While all this is conjecture, it does show that Labour cannot win on the back of a fall in Lib Dem support alone. But it does make the task of beating the Conservatives somewhat easier.
Perhaps Ed Miliband’s biggest challenge if he is to form a government is to win back enough of the five million voters who used to support Labour. Britain is still very much divided by social class. Voting patterns are no exception. However, the country’s demographics are changing fast and this has major implications for Labour’s vote.
There has been much hand-wringing in Labour circles about the collapse of the C2 vote (skilled manual workers). The figures are horrifying. There has been an 11-percentage-point fall (a 7.5 per cent swing towards the Conservatives), with a corollary DE (semi-skilled, unskilled and unemployed) swing of 7 per cent. Clearly, Labour needs to re-engage with those who used to be regarded as its core voters in order to win again.
However, on closer inspection, the electoral map is a lot more complex and a political strategy based solely on the core vote would be an extremely risky one. The way social grades are split suggests that each grouping represents a quarter of the electorate. In fact, this is not the case. Not only do C1s out number other grades (and ABC1s outnumber C2DEs), there is also the impact of turnout. The data is unequivocal in showing that the poorest in society don’t vote. The difference between AB and DE turnout is nearly 20 percentage points. If you weight the social grades, it shows Labour cannot possibly win on the back of working-class support alone.
At the 2010 election, Labour gained more votes from ABs than from C2s and similar numbers from C1s as DEs. ABC1s contributed slightly more to Labour’s tally than C2DEs. That said, while Labour’s support among ABs looks fairly strong, there is a delicate balancing act to be managed between the social grades. Labour needs to win back more DE voters, but not at the cost of its AB vote.
Back in 1992, for example, Labour secured just 19 per cent of the AB vote (compared with 30 per cent in 1997 and 26 per cent in 2010). If Labour’s AB fell back to 19 per cent, it would require a relatively larger (9 per cent) rise among DEs to secure the same national support as it did in 2010 (this is against the backdrop of rapidly falling DE support). So there is no turning back to the idea of a working-class majority for Labour and to pretend otherwise would be counter-productive.
Although Labour’s performance between age groups at the 2010 election is broadly in line with national trends across the board, there are some worrying factors to take into account. While Labour leaked rather a high percentage of 18 to 24-year-olds, it was hurt a lot more by the four-percentage-point drop among voters aged 65 and over, who represent a larger cohort and vote in much bigger numbers.
However, the most costly loss to Labour was among those in the 35-44 age range. This group represents 20 per cent of people who vote and Labour managed to drop from 41 per cent to 31 per cent in five years – representing just under a third of its lost support. Overall, Labour’s fall in support among 25 to 55-year-olds (which represents around 50 per cent of those who voted) was twice what it was for all other age groups combined.
Labour also has to confront issues around gender. Among women, the biggest drop was with C2s (down 15 percentage points), with both the Lib Dems and Conservatives reaping the rewards. For men, the biggest drop was for DE voters. Depressingly, they seem have largely switched to the BNP.
For Labour to succeed, it must have a vision for the country that resonates with all groups in society. It also needs to have better spatial awareness. The Smith Institute’s report focuses the spotlight on this geographical dimension of the missing five million and contests the notion that Labour was most undone in the south of England.
The Policy Network’s recent paper, Southern Discomfort Again, draws attention to the fact that Labour won just eight out of 139 seats in the south-west and south-east regions. This would suggest there is a strong case for concentrating on winning back the south in order to put a Labour Prime Minister in Number 10 Downing Street.
However, the statistics are misleading. Between 1997 and 2010, Labour lost more seats in London, the eastern region and the West Midlands. Regardless of this fact, the main problem with a southern-centric approach to analysing voting trends is that it is not weighted for the size of the region (the south-east has more seats than any other region) or Labour’s performance in these places over the years.
Analysing the percentage point change (looking at Labour’s percentage of the seats in 1997 or 2005 and comparing it with 2010) gives a far more accurate picture. Using this approach, we see that between 1997 and 2010 Labour lost worse in the east of England, the West Midlands, East Midlands, London, and Yorkshire and the Humber.
There is no evidence to support an electoral strategy based on “southern discomfort”. Labour has traditionally fared far worse in southern England than elsewhere and to focus on recapturing an area which has never been won by the party in the first place (Labour secured just 28 per cent of the seats in 1997) would be comparable to a Conservative strategy based on winning Wales and Scotland.
Of course, Labour can and must do better in the south of England. But there are bigger gains that can be made elsewhere. Labour’s policies and voting strategy must go beyond an emphasis on the south and take a much more assertive stance on boosting jobs and growth (and winning back key marginals) in the Midlands and north of England.
While Labour’s “southern discomfort” may be exaggerated, even a cursory examination of the electoral data reveals a serious problem of suburban hostility to Labour. Many of the party’s lost supporters live in suburbia, yet these areas have proved hard for Labour to reconnect with.
Much has been made of Boris Johnson’s suburban-biased electoral strategy for the 2008 London mayoral election. That is hardly surprising, given the electoral map of where he won in the capital. If a picture is worth 1,000 words, then Labour should take a hard look at this one. By speaking to the hopes and fears of the suburbs, Boris Johnson was able to win. Without their extra support, Ken would still be the Mayor of London.
The same could be said for the general election of 2010. Labour failed to connect with voters in suburban towns across the country – many of which are swing seats. These places have a social mix similar to the average in the country, but are neither part of Labour’s urban stronghold nor its heartlands. Labour’s success may in part depend on its ability to relate to the needs and aspirations of suburbia. Contrary to what some may think, these are not at the behest of the Tory Party.
The inquest into Labour’s defeat has not yet reached a conclusion, but one important lesson to be learned is that relying on Tory or Lib Dem unpopularity or the so-called core vote will not be enough. Labour can win back many of the vanished five million to form a majority, while a future of hung parliaments should be regarded as a false prospectus. To win again, Labour needs a different electoral strategy based around policies that are attractive to voters in the places where its support has declined the most.
Paul Hackett is director of the Smith Institute. Winning back the Five Million – Understanding the Fragmentation of Labour’s Vote by Paul Hunter is available from the Smith Institute website www.smith-institute.org.uk
