I have an article in this week’s TCS, “No longer any need to mind the gap”. Here’s the complete version:
Women earn approximately 17.2% per hour less than men, on average. The Fawcett Society’s “No Pay Day” claims that this means, from October 30th, all women are working for free. A speaker in a recent Union debate said that “women earn £569 per month less than men”, and that there is probably a pay gap at the University because more bedders are women and more professors are men. Well, yes.
The 1970 Equal Pay Act says that two workers doing the same jobs to the same standard should get paid the same. This is sensible. So why, forty years later, does the pay gap still exist? Is the remaining gap really the result of sexism?
I used to believe it was. But it turns out that if you control for things like part-time work, and men and women being more likely to do different jobs, the gap disappears.
For example, some jobs done more by men have disadvantages that are reflected by higher pay. Men are more likely to work outside in all weathers and work unsocial hours. “Women’s jobs” are less risky in two ways: men are much more likely to be made redundant, and suffer much higher rates of industrial injury. Women have shorter commuting times to work, and take more time off. Women report greater job satisfaction than men.
More women work part-time than men. It costs more to train two workers than one, so part-time workers cost an employer more per hour than full-time, and this is reflected in lower hourly pay. This shows up in the overall pay gap, but doesn’t indicate sexism.
More women than men do certain jobs, and vice versa. This is the result of different average preferences. For example, 36% of male managers work more than 48 hours a week, but only 18% of female managers do. Women with careers are 4.5 times more likely than men to say they preferred to work fewer than 40 hours per week. In one study, men tended to place more importance on “being successful in my line of work” and “inventing or creating something that will have an impact”, while women tended to place more importance “having strong friendships”, “living close to parents and relatives”, and “having a meaningful spiritual life.” But amongst men and women doing the same jobs, the gap can disappear, or even be negative. Female investment bankers and dieticians, for example, earn significantly more on average than male ones.
In many couples, the female partner often spends more time looking after the children, which would reduce her overall lifetime earnings. That is why there is no pay gap amongst the young. In the UK, the median pay gap between 22 and 29-year olds was less than 1% in 2007. A US government study found the gap between men and childless women between the ages of 27 and 33 was about 2%. Middle-aged women who remain single earn more than middle-aged single men. Lesbians and gays earn more than heterosexuals.
If you look at the figures more closely, you find not only is sexism not necessary to explain anything, but that there are some things which cannot be explained by sexism. On average, Bangladeshi women in the UK earn about 26.8% more than Bangladeshi men, and Black Caribbean women 1.5% more. This hardly indicates sexism.
I’m no apologist for sexism; it’s stupid and inefficient, and sexist employers who don’t hire the best person for the job are losing out themselves. And surely sexism does still exist in the workplace. But too often widespread sexism is inferred from simplistic econometric analysis with no other evidence. And, as I hope I’ve shown, this inference is misguided. A study by economist June O’Neill, former director of the Congressional Budget Office, found that women earn 98% of what men do when controlled for experience, education, and number of years on the job.
I’ve been talking a lot about averages. Of course, many women do jobs which are mostly done by men, and many women get paid more. Really, there is now so much variation in the lifestyles and economic behaviour of men and women that simple comparisons of average male and female pay etc are increasingly irrelevant.
The data do not indicate sexism, and those who claim they do are guilty of “cherry picking” data (a scientific cardinal sin), not comparing like with like, and selective reporting of the facts. They focus on the “headline” figure and don’t look any further.
