There's evidence that, entirely contrary to what people like Harriet Harman have been trying to tell us, it is men who are bearing the brunt of this recession in terms of job losses.

The jobless figures for May showed unemployment at 9.4%, a 25-year high. But while rates for men and women were roughly equal in 2007, 10.5% of men are now unemployed, compared with 8% of women. Four of every five jobs lost in the past two years had been held by men. The gender gap is the largest ever seen in US labour statistics, which go back to 1948.

“What’s happening in this recession is unprecedented," said Mark Perry, an economist at the University of Michigan. “It’s structurally different because the job losses are so concentrated among men."

Blue-collar jobs in manufacturing and construction are haemorrhaging while white-collar work in increasingly female-dominated, often publicly funded fields, such as education and health, are holding steady or growing.

As you can see a large part of this is down to occupational segregation. Further, it's not just the types of jobs, services against manufacturing, but it's also the sector, public or private. Public sector jobs are more secure, less likely to disappear in a recession.

Which leads us to a further conclusion, similarly entirely at odds with what Harriet Harman and her ilk try to tell us. That the gender pay gap (at least, not all of it) is not due to discrimination. It's due to entirely rational sorting and the choices made by individuals.

Just as more dangerous jobs pay a wage premium to compensate for the risks of injury so do or should those more insecure jobs pay a premium. Or if you prefer, workers will, if they are risk averse, choose a lower paid but more secure job. Which means that if women are preferentially employed in the lower risk public sector then womens' wages will be lower than mens'.

No discrimination required, just individuals deciding what they prefer by their own lights.