Women are off politics, or at least politicians. No one really knows why, though it's not hard to think of quite a few good reasons. But the fact is that, when it comes to opining on Gordon Brown as the next prime minister, or David Cameron as the one after, a third of women invited to express their views say they don't know.
This might reflect a perfectly rational reluctance to sound certain when they aren't; even a quarter of men polled are prepared to admit that they have not made up their minds. But political strategists, like nature, abhor a vacuum, almost as much as they relish the chance of moulding a statistic to suit a particular purpose. And so, the women's vote is back in the frame.
Because politicians need women's votes: it is one of the triumphs of psephological analysis of the past ten years that it is now understood that women decide elections. The discovery that, without their tendency to back the status quo and vote Tory, there could have been non-stop Labour governments since 1945, and that only Tony Blair's seduction of the female vote in 1997 restored Labour to power, has played a big part in shaping the way politics has been played in the past decade.
Even in these past few months, when the Blairites were still casting about for an anyone-but-Gordon candidate, Brown's alleged lack of appeal to women was a key line of attack. That big clunking fist, which for a brief minute or two looked like a genuine Prime Minister's Questions endorsement - how quickly it became a double-edged sword. Women, gentle souls that we are, do not like big clunking fists.
Yet, in parallel with the slow, quiet transformation to a kind of feminised politics, the influence of women on political outcomes has been treated as a secret weapon. It has been anonymous, almost surreptitious. New Labour always found feminism, like anything else that might be described as a sectional interest, deeply worrying. Of course there was a lot of space for women in the big tent. Just not feminists. This was the terrible irony of that toe-curling image of the 1997 intake of power-dressed new Labour women MPs.
But it is a fact now universally acknowledged that Gordon Brown will not get his own mandate for No 10, and Labour cannot win a fourth general election, unless he wins the major share of the female vote. So the recent poll headlines declaring that women have lost faith in Blair and Labour are cause for alarm.
At the last general election, the gender gap appeared wider than ever. The switch-over from the deferential vote that typified older women to the differential vote of younger ones was confirmed. Women stayed markedly more loyal to Labour than men, but in the two years since, there has been a sharp decline. Forty per cent of women who voted in 2005 supported Labour. April's ICM poll for the Guardian showed that support among women had fallen to 29 per cent, and with Brown in charge it would fall further. Only 22 per cent of women would vote Labour if it meant getting Gordon Brown as prime minister. Young women in particular appear to be flirting with David Cameron's Conservatives.
However, closer scrutiny of the polls shows that this is not the whole story. What they actually reveal is indecision, based on ignorance. It is a trend that has been evident at least since November, when Ipsos MORI, working with the Fawcett Society, observed a collapse in women's support, not for Brown but for Cameron, even sharper and faster than among men: an approval rating down from +13 among women in the first quarter of last year to +1 by the end of June.
Labour's feminists (and let's hear a particularly big hand for Harriet Harman) have played a brilliant game with the new gender sensitivity of the opinion polls. They need our votes, girls; what do we want from them? Male politicians finally had a reason they could understand for paying attention to an agenda that had been hived off to the women's conference. It has become received wisdom that a particular kind of leader, a particular style of politics, is more likely to attract women. Tony Blair's success was taken as proof that women liked consensual, big-tent politics, and confirmation that the tribalism of the previous 20 years had been an even bigger turn-off for women than for men.
It is the importance of female voters that has encouraged Conservative Central Office to portray David Cameron as their man in the Marigolds, an honorary sister with a working wife. It has even become an established fact that, far from women not supporting female candidates (which itself used to be another established fact particularly popular with male-dominated selection committees), women actually prefer female candidates. Analysing the 2005 vote, the Electoral Commission found that, where there were female candidates (of any party), women were much more likely to contribute to the campaign. Yet new Labour has never talked about feminism. Among the articles of faith junked by the Blairites back in the Nineties was that women might be won on issues that reflected women's experiences. Ideas such as domestic violence or rape or childcare - issues that were negatives among Labour's new middle-ground constituency hovering over the centre of the liberal-conservative continuum - became as unfashionable as shoulder pads and underarm hair.
Part of the explanation for this might have been a misreading of analysis of election results in the Eighties and early Nineties, which showed that the tendency for women to vote Conservative was fading. The gender gap seemed to have closed. But when the academics - particularly Pippa Norris, now at Harvard - started grinding away at the figures from the 1997 and then the 2001 votes, they suggested that the convergence theory was not quite what it seemed. The research that became received wisdom in the 1990s was reappraised. Now, all sorts of ideas are being re-examined, not necessarily for the purest of motives. After all, murmur the Brownites, if women prefer consensus and abhor confrontation, Margaret Thatcher's ability to attract and retain women's support becomes inexplicable.
It is easy to overstate the scale of the differential vote (and risk undermining the case for rethinking politics). One of the leading authorities is Rosie Campbell of Birkbeck College. She states baldly in her latest book: "There is no longer a significant gender gap and any reference to a 'women's vote' would be spurious." Happily she does not stop there. Although women and men do indeed vote in roughly equal numbers for the main parties, the reasons why they vote for the party of their choice are different.
Two factors are at work. In both the United States and Europe, there is now a body of evidence that suggests that once women have access to education and economic independence they move perceptibly to the left. This has nothing to do with the sexiness of the politicians. In fact, it is quite distinct from any personal appeal. It is simply that women under the age of 45 con sistently say that they believe in the role of the state to support its citizens. This is the generational gender gap, much more marked in the two-party politics of the US, where since 1980 women's preference for the Democrats has been repeated at every election.
In multiparty Britain, the picture is hazier. But Campbell's analysis of the 2001 and 2005 votes detects the same trend. What she has also done is pursue the question "why?". Using focus groups, she has developed a picture of the different ways in which women and men talk about politics, picking up on the tendency among women to talk in personal rather than abstract terms.
She has confirmed how much women's votes are influenced by the priority politicians give to issues that concern them, issues that are ranked in a different order from men's concerns. Education and (among older women) the NHS are more important to women than to men, and women support higher taxes for higher welfare spending in far greater numbers. Counter-intuitively, there is little evidence of a gender gap on the environment: this, as Cameron has spotted, is a youth issue. Women are less concerned than men about the economy, Europe or immigration. But - and this could worry Labour strategists - women, while ranking the economy as a lower priority than men do, are more pessimistic and anxious about it. Rising house prices and the return of inflation both seem more significant concerns to women than to men.
But perhaps Campbell's most important discovery is the link between being a mother of school-age children and a tendency to vote Labour. Using the British Social Attitudes survey, she has found that a middle-class, well-educated, well-paid woman working in the public sector with children under 11 is 70 per cent more likely to vote Labour than a similar man. Yummy mummies for Gordon? Well, maybe. The research is still in progress: but it indicates that, far from being less significant, gender - in a world of relative economic security, independence and growing gender equality - actually becomes more important in shaping women's voting choices than in an age when social pressures and religion shaped women's lives.
This new reading of the polling data might look dangerous for Gordon Brown. In fact, it is a gift. If the secret of the successful politician is to describe politics so that the voters can recognise their own concerns in the politician's agenda, it should not take a creative genius to describe Brown as the man with the answers for British women.
Yet it's easy to see how the idea that Brown doesn't appeal to women got about. He might still creep into the 100 sexiest men in Britain (at 97) but no one can argue that his political style is more masculine, more shaped by that combination of arrogant confidence and physical and intellectual muscle that is thought to repel women, than any other leading politician of the television era. Forget the tear in the eye during the TV interview when he was tackled about the death of his baby daughter, Jennifer; in politics, he is a man who seems never to have encountered his feminine side, let alone kept in touch with it. There may be two powerful women among his closest advisers in Sue Nye and Shriti Vadera, but the impression is of a bloke surrounded by blokes who likes, on the very rare occasions when he isn't doing blokey economics, doing blokey things such as watching football with blokes.
This sense of a divorce from the everyday is reinforced by the discreet distance his wife, Sarah, usually keeps from the political fray. (This is a no-win for the Brown family: a higher profile would no doubt be called exploitation.) Even without the helpful pointer from his former permanent secretary Lord Turnbull, his public image is controlling, even dominating. Few who saw it can forget the press conference where Estelle Morris was expressly asked a question about Labour and women, and Brown barged in just as she opened her mouth to answer. Lobbyists recall the hours of argument before he accepted that family tax credit should go to the carer rather than the wage earner, or that lone mothers were entitled to support if they wanted to return to work. "He seems to have no comprehension," one remarked after a recent seminar on parenting, "of the sheer messiness of many women's lives."
Can't change now
Social justice, Brown's overriding concern, means economic justice. To him, gender equality (and racial equality), surely vital aspects of social justice, are a matter of economics. He talks about women only when he also talks about men; he uses gender-neutral language such as "parents" when he could be selling his policies to women; and he talks about childcare, it sometimes seems, not because he understands how it can transform lives, but because it allows women to go to work and earn their way out of poverty. It sometimes seems that in his instrumentalism, we are all white men.
But, say despairing female friends and admirers, he can't change his image now, least of all when what he most needs to do is to distance himself from celebrity politics. It would lack credibility and undermine his strengths. So here is a re assuring message: he doesn't need to.
Somehow it is good to learn that it may not have been Tony Blair's smile, or his easy charm, that won women over to Labour in 1997: it was something that was happening anyway. Because younger female voters tend to be swayed not by material concerns but by values (a finding reinforced this month by a study of favoured graduate professions that showed women choosing Oxfam or public service over the City or industry), they are predisposed to support a progressive party. It doesn't matter - or not that much - that Brown's political style is the antithesis of what is thought to appeal to women. It is not style that matters, it's substance. He has to do one big thing. Even if he can't use the F word itself, he must abandon Labour's reluctance to talk the language of feminism.
It is hardly rocket science to argue that a party that wants to build its support among women should promote its record on gender issues. The only mystery is why, when he has such a good story to tell about funding for health and education, Sure Start and childcare, Gordon Brown still doesn't recognise the importance of telling it.
Oh, and it would certainly help if he had a woman deputy.
l19.5% of MPs are women
44% of women voted Labour in 1997
95 is the current number of Labour women MPs (after 1997 election it was 101)
17% is the differential between men's and women's hourly pay
57% women's retirement income as proportion of men's
10% of FTSE-100 directorships are held by women
Research by Jonathan Pearson