Sometimes, it's the little words that reveal so much. I was struck by Phil's response to Rashida Manjoo's claim that sexism is worse in the UK than elsewhere. He writes:

I didn't for one moment think that the UK's sexism problem was worse than other developed nations.

Hold on. What's that word "developed" doing here? Why judge the UK only by comparably rich nations? Is Phil presuming that sexism is worse in poorer nations?

If he is, he's right - up to a point. My chart plots gender inequality, as measured by the WEF's gender gap, against GDP per head. You can see that there's a positive correlation in the chart; it is 0.36 for all 134 countries, but 0.52 if we exclude middle east countries who tend to have high oil revenues and poor gender equality scores. Sure, some poor countries score well for gender equality - such as Mozambique or the Philippines - but generally, Phil's presumption is correct. Poor countries are generally sexist ones.

The evidence here, of course, isn't just cross-sectional. It also exists in time-series; western economies are much richer than they were 100 years ago, and gender equality has also improved in this time; women at least have the vote now.

Why is there this correlation? There are three possibilities:

- Gender equality actually promotes economic development. There's good evidence (pdf) that educating girls (pdf) has a high pay-off in poor countries, perhaps by enabling them to better control fertility (pdf) and child mortality. And one study has found that "countries with higher shares of women in parliament have had faster growing economies" - probably because large numbers of female parliamentarians are a sign of other enlightened policies.

- Poverty and gender inequality have a common, third, cause. If men are wedded to traditional ways of treating women, they are likely also to be disposed to other traditional ways of life which are hostile to development.

- Development creates equality. As Ben Friedman and Deirdre McCloskey have shown, richer people tend to be more civilized ones. Yes, western bankers and CEOs are bastards, but they're vastly better than the Lord's Resistance Army or Boko Haram.

I draw two inferences from this. One is to support a point made by Diane Coyle in GDP: A brief but affectionate history. She says that although GDP is not a meaure of welfare, it is "highly correlated with things that definitely do affect our well-being." It would be rash to assume that none of the correlation between gender equality and income was due to causality from income to equality. Maybe one reason we need growth, especially in poorer countries without oil, is to improve the odds for women.

Secondly, there's a question here. The evidence shows that prosperity and gender equality go together. So why shouldn't this be true for other forms of equality? Why are women so unusual?