If girls do better in country A than they do in country B, that seems to be a win. But when we look at how well boys do, we may find that boys in country B do

If girls do better in country A than they do in country B, that seems to be a win. But when we look at how well boys do, we may find that boys in country B do even worse than the girls, which means that country B scores better for some form of equality. If we score policies by this metric then we may well advise that country A do things just to knock down the boys' scores. This is probably the wrong lesson to draw. We don't improve life or results for girls just by making things worse for boys. At best, a disparity highlights a place to look. So the point isn't that it is bad to compare countries to each other. The point is actually that this comparison should consider girls-to-girls across countries, not just girls-to-boys inside each country.