However, although a lot of the evidence (pdf) is ambiguous, economic researchers have discovered some gender differences in decision-making, for example:
1. Groups in which women are the majority are more likely to be generous (pdf) to outsiders. However, all-women groups are less generous than majority-women groups.
2. Women are less good at negotiating than men, especially in situations where benchmarks are opaque - such as in businesses with unclear salary structures.
3. Women perform worse under pressure than men. For example, women tennis players are more likely than men to make unforced errors at crucial stages of a match.
4. Men are more over-confident than women. For example, men trade shares more often than women, but earn lower returns.
5. Women are more risk averse than men - perhaps because of socialization rather than innate differences - though not more averse to ambiguity.
Do these differences justify Harman’s claim? In themselves, not at all. These differences are true only of average tendencies. There are loads of women who are less risk-averse than men, for example. Charlotte is entirely right to say that it is silly collectivism to pretend otherwise.
Indeed, it could be that selection effects in politics are powerful than gender differences. For example, if political careers only appeal to people who are over-confident, the tendency for women to be less over-confident than men will be irrelevant in deciding who should be Labour leader, as women who aren’t over-confident won’t have entered politics in the first place.
There is, though, another way in which women leaders might improve decision making, even if they are identical to men in their cognitive skills or lack thereof. The very fact that women are different might help to reduce groupthink - the tendency for groups of decision-makers to become overly homogenous and so over-confident and insufficiently self-critical. Evidence for this lies in the fact that, sometimes (pdf), women board members can improve company performance.
What’s not so clear, though, is whether this effect is the result of their femininity or just their difference. If it’s the latter then there might be no more case for having a woman leader than having a black or gay one, as these too might reduce groupthink.