The bombardment has started. Already the air is thick with election claims, insults, polls and analyses. Within weeks the noise will grow yet more intense, as the campaign proper gets under way. If you wish to protect yourself from the more dubious assertions, look no further. Here are seven pillars of electoral wisdom. Keep these in sight and you will stay safe.
The election will be decided by the whole country, not a few thousand voters in a handful of seats
After each election someone pops up and says something along the lines of: "had 10,000 voters in 20 seats cast their votes differently, the result would have been different". There is a narrow arithmetical truth in this. If a party wins a seat by 1,000 votes, then they would have lost had 501 of their supporters switched to the party that came second. Apply that logic to the 20 most marginal seats, and you end up arguing that just 10,000 votes matter and the other 30 million don’t.
There are two problems with that proposition. The first is philosophical. Imagine a swimmer crossing the channel from Dover to Calais. After hours of effort she arrives at the French coast. Had she stopped ten metres short, she would have failed to complete the crossing. To coin a phrase, the result would have been different. But that is true of every ten metre stretch of her swim. To complete the crossing, she needs to swim all 34 kilometres. By the same token, to win a parliamentary seat, a candidate needs all 20,000 votes (or whatever the local winning-post number happens to be); to single out the tiny number that comprises half-the-majority-plus-one is absurd.
Second, even the narrow calculation can be done only retrospectively. In 2005 Bill Rammell won Harlow by 97 votes. Robert Halfon, the losing Conservative, might well have kicked himself, thinking, "if only we had persuaded 49 more Labour voters to support me instead, I would have won". The trouble with that logic is that he fought the campaign not knowing those numbers. He started the election needing to overcome the 5,228 majority that Rammell achieved in 2001. To do this, he had to keep the Tory base solid, get his supporters out to vote on the day, prize votes away from Labour, and persuade as many Liberal Democrat, UKIP and Veritas supporters as possible that their vote would be wasted unless they backed him. Suppose he said to his campaign team at the outset: "I have had a vision of the result. If we carry on as we plan, we shall fall 97 votes short. Let’s junk our strategy, concentrate all our efforts on identifying 49 Labour supporters whom we can persuade to vote for us". Do you think a wise agent would have (a) hailed him as a tactical genius, or (b) told him to lie down and not get up again until he had come to his senses?
Nationally, the story is a bit more complex. Most seats are safe. We know that Bolsover will stay Labour and Beckenham will stay Tory. The real battleground comprises at most a third of the 650 seats being contested. But this still means that millions of votes "really" matter. And even then, surprises can happen. During the 1997 election, the BBC produced a wonderful "bible" containing detailed information on every constituency. What did Crosby, Hove and Harrow West have in common? The "bible" tagged them all as "very safe" Conservative seats – and all fell to Labour. In more recent years, Labour too has suffered losses in such strongholds as Blaenau Gwent and Bethnal Green & Bow – a further warning that parties must nurture their heartlands as well as protect their marginals.
Forget "Essex man", "Worcester woman", "soccer mums", "Mondeo dads" and other stereotypes. They don’t decide who wins elections.
There is an obvious attraction in relating dry statistics to human beings. To many journalists, a real voter is worth a thousand polling respondents. It is hard, and not necessarily right, to resist the temptation to find a typical voter in a marginal seat and seek her political views. Such exchanges can illuminate how real people view the parties and their leaders.
The problem comes when such encounters are invested with cod sociological authority. This happens when a handful of voters are given a label, and we are told that they hold the key to 10 Downing Street. This is invariably nonsense. Take "Worcester woman". This phenomenon was "discovered" by the Sunday Times shortly before the 1997 election. The paper decided to put an opinion poll story on its front page. Someone noticed that Labour seemed to be doing particularly well among women in the West Midlands, and dramatised this by putting "Worcester woman" in the headline. The label stuck, but two vital truths were ignored: first, the subsample of West Midlands women was far too small to justify the assertion that they were especially pro-Labour; second, none of the poll’s respondents lived in Worcester.
It is rare for particular demographic groups to move very differently from the electorate as a whole. To be sure, there are exceptions: Muslim voters and students deserted Labour in significant numbers in 2005 over Iraq and tuition fees respectively; the party lost a handful of seats it might otherwise have retained. But broadly speaking, movements from one election to another are notable more for their uniformity than their variation. When Tony Blair swept to power in 1997, Labour gained ground among rich and poor, men and women, young and old. The levels of support among these groups varied widely; but the swing from Conservative to Labour was significant in every group. If David Cameron is to win the coming election, he, too, must win converts in every demographic group.
The Tories do NOT need to do anything special in Scotland or the north of England to win the coming election
Two related myths need to be demolished. The first is that Conservative support in Scotland and the north has haemorrhaged more than in the rest of Britain since they were last in power; the second is that they need to do particularly well in the northern half of Britain this time to achieve an overall majority. Neither is true.
Here are the figures. Across Britain as a whole, Tory support in 2005 was 10 percentage points lower in 2005 (33 per cent) than 1992 (43 per cent). Over the same period, Tory support fell by the following amounts in northern Britain: Scotland 10 points, the North-East 14, North-West 9, Yorkshire and Humberside 9. Only the North East stands out, and it is the smallest and least significant region.
There is some truth in the perception that the Tories have ceded ground in the north. But this was mainly a phenomenon of the 1960s and 1970s. Among the factors that contributed to this were: the decline of the Tory protestant working class vote, for example in Glasgow and Liverpool, as class came to matter more than religion; and the growing divergence in economic and employment trends between north and south. The larger point, though, is that the Conservatives won their largest victory (in 1983) and their least expected victory (in 1992) after the north-south divide opened up.
In the coming election, yes, the Tories must make significant gains in the north in order to win power. But there are plenty of seats within reach. Thirty-four Labour seats in England’s three northern regions would fall to the Tories on an eight per cent swing. As long as the national tide flows Cameron’s way, enough northern seats would change hands to give his party victory.
Two qualifications should be entered. First, most of Labour’s super-marginals are concentrated in the south, as the maps show. Adjusting the results of the 2005 election to the new boundaries, Labour is defending an overall majority of 48. If it loses 24 seats to the Tories, it loses that majority. Sixteen of the 24 most marginal Lab-Con seats are in the South. But, on their own, such paltry gains would represent a dismal failure for the Tories. At the very least, they want the extra forty-plus gains they need to become the largest party; ideally, they want the third tranche of seats that would secure Cameron an overall majority. And plenty of seats in the second and third tranches are in the north.
Second, Scotland may well behave very differently from the rest of Britain. But it doesn’t matter (that is, in terms of the arithmetic at Westminster: the impact of the election on the dynamics of Scottish politics could be enormous). The Tories hold just one Scottish seat. If they do incredibly well, they could win another seven, including, as a very long-shot, unseating Jim Murphy, the Scottish Secretary in Renfrewshire East. Gains of one or two are more likely. In the context of the 116 gains the Tories need to make across the UK to win an overall majority, Scotland is a side-show.
There is a significant geographical dimension to the coming election. The seats the Tories need to win are mainly suburban and small town seats. It is no accident that so many of the stations on the Thameslink line from Bedford to Brighton are in Lab-Con marginals – their commuters comprise the broad social mix of voters who decide elections.
The performance of the Liberal Democrats could shape the future of British politics
When the Tories last won a general election, in 1992, only 20 Liberal Democrat MPs were elected to the House of Commons. By 2005, the number had tripled to 62. All by five of the extra 42 Lib Dem seats had been Conservative seats in the 1980s. Plainly the Tories would find it far easier to gain the 116 seats they need to win power if they could win most of these lost seats back.
Will they? It is instructive to look back to what happened in 1979, the last time The Tories won power from opposition. Overall, the Conservatives share of the national vote climbed eight points, while the Liberal vote share fell by four (figures similar to those indicated by today’s polls). Had these changes occurred in each of the 13 seats being defended by the Liberals, seven would have fallen to the Tories. In fact only three changed hands. In the other four, the Liberals gained votes and kept the Tories at bay. The same pattern may well apply this year: national trends might imply, say, 20 Conservative gains from the Lib Dems; in practice the Tories may win far fewer. And the Lib Dems might actually capture one or two Tory seats. For example, boundary changes have made Solihull a Tory marginal; but the actual incumbent, on the old boundaries, is Lib Dem MP Lorely Burt. I would not advise betting too much against a “Lib Dem gain” here on election night. All in all, I would be surprised if the number of net Tory gains from the Lib Dems climbs above single digits.
The Tories face another handicap. There are 17 seats that were Tory in Margaret Thatcher’s time where the Conservatives are now in third place. Nine of them passed through Labour’s hands on their way to the Liberal Democrats. (They include eight of the 12 seats that the Lib Dems captured from Labour in 2005.) With the exception of Watford, which is now a three-way Labour marginal, all are probably beyond reach of the Tories this time round.
In short, there is almost certain to be a much larger block of third party and minor party MPs than there used to be. In 1979 there were just 28 of them; so Margaret Thatcher was able to govern with a comfortable overall majority, with 70 more Tory than Labour MPs. By 2005, the 28 had become 92. This year, if the Tories lead Labour by 70, Cameron will almost certainly NOT secure an overall majority, unless the Lib Dems collapse.
Even if the Tories win outright this time, then as long as at least 70 MPs belong to the Lib Dems and minor parties in total, future general elections will start throwing up hung parliaments. They may even become the norm. Which is why the performance of the Lib Dems matters for the future of British politics and not just any possible manoeuvrings that might take place over the weekend following the coming election.
Issues don’t decide elections; valence does
Cast your mind back to the Romsey by-election a decade ago. The Tories were defending one of their safest seats. Immigration was the dominant issue of the day, especially in the south-east, where asylum seekers from Sangatte, near Calais, were arriving daily. The Tories sought to exploit the popular mood in their by-election campaign. But the Lib Dems, whose immigration policies were backed by few voters, captured the seat with ease.
Their victory demonstrated the power of valence politics. In this context, “valence” has nothing to do with bedding. It is a term political scientists employ to bring together such judgements as competence, character, authenticity and performance. “Valence voters” contrast with “positional voters”. For example, if you feel strongly that greater private involvement in health care is a good thing – or a bad thing – you have a positional view of the NHS. If you don’t feel strongly either way, but simply want a prime minister you can trust and a government that will ensure you are treated promptly and effectively when you fall ill, you are a valence voter.
Now I, and perhaps you, hold strong views about many aspects of public policy. Most voters – and a large majority of the floating voters who decide elections – don’t. They are valence voters, who care mostly whether politicians are decent, honest, capable and generally “on my side”. Thus any benefit the Tories gained in Romsey by emphasising their tough stance on immigration was massively outweighed by the view that they were cynically exploiting the issue for electoral gain. The Lib Dems won the valence battle – and the by-election war.
What was true about immigration in Romsey then is true about virtually every issue nationally today. Just as Margaret Thatcher won the valence war with successive Labour leaders in the Eighties (“she’s strong, knows what she wants, and will sort Britain out”), and John Major lost it in the nineties (“he’s weak and leads a divided, sleazy and incompetent party”), so the coming contest between Brown and Cameron (and Clegg) will be decided by the valence judgements of millions of voters who have yet finally to decide their vote.
Stuff happens
Already in this young century two European elections have been upended by unexpected events. In 2002, Germany’s Chancellor, Gerhard Schroder, was heading for defeat, his SPD administration mired in scandal and economic failure. But less than six weeks before the elections, the River Elbe burst its banks, the Bitterfled dam collapsed and much of eastern Germany was stricken by floods. Shroder’s decisive response and his compassionate manner revived his popularity; his Red-Green coalition won a narrow majority.
Eighteen months later, Spain’s centre-right government looked set to be re-elected when, three days before polling day, a series of bombs exploded in Madrid’s commuter train network, killing 191 people and injuring 1,800. Initially the Government blamed Basque separatists, but within hours it became clear that Al Qaeda supporters were responsible. The opposition socialists benefited from the government’s bungled response and won the election.
So: could some deus ex machina (hopefully less calamitous than a flood or a bomb outrage) alter the course of British politics and produce either a Tory landslide or a fourth Labour victory? By definition, we can’t tell in advance. There is, however, one “known unknown”: the TV debates between the party leaders. If any of the party leaders drops a mighty clanger, he will suffer.
Polls have improved but will never be perfect
Pollsters were traumatised by their failure to predict the clear Conservative victory in 1992. All have since changed their methods, albeit in different ways; and new companies have sprung up, determined not to repeat the mistakes made 18 years ago. So will we be certain on election morning what that night’s results will bring? Here are three reasons for caution.
First, telephone polls in 2005 appear to have retained a slight Labour bias. Looking at polling trends through the campaign, we find that Labour’s support remained stable; but two-thirds of telephone polls overstated Labour’s final vote-share by three points or more; none understated it by this amount. All the final polls came close to the result, though a slight tendency to overstate Labour support still remained. Things may be different this time; but we will not be able to tell until afterwards.
Second, even the best-conducted polls are subject to sampling error. If each figure is within two points of the actual result, statisticians would regard that as perfectly acceptable; figures just one point adrift would be regarded as a good polling outcome. But statistical accuracy is not the same as political accuracy. Even one point errors can change the political significance of the election. Suppose the final polls indicate: Conservative 40 per cent, Labour 30 per cent. If the actual result is 39-31, Britain is likely to have a hung parliament; but if it is 41-29, Cameron is likely to secure a clear majority.
Third, polls measure votes; what matters ultimately is seats. Votes can translate into seats in unpredictable ways. In February 1974, the final polls pointed to a narrow Conservative lead in the popular vote. Headlines predicted victory for Edward Heath. The polls were right about who would win the popular vote – but Labour emerged slightly ahead on seats, and Heath had to resign. The polls were blamed unfairly for getting the election wrong.
The best advice is to follow the dictum of Adlai Stevenson, the Democratic candidate for the US presidency in the 1950s: "polls should be taken but not inhaled".
Peter Kellner is President of YouGov. An edited version of this article appears in the March 2010 issue of Prospect.
