I think we wrote past each other on this one. Where I see the parallel is in that we shouldn’t just look at the total gap, subtract known non-discriminatory reasons for the gap, and assume that the entirety of the remainder is due to discrimination.
For example, we have a “custody gap” of 85:15 instead of 50:50. I believe that there are societal biases against fathers as parents. We know why some of this gap exists, as discussed in my previous post, but I’m not going to say that the remaining difference is due *entirely *to anti-father discrimination. I believe the contribution of that discrimination is non-zero, and bounded on the high end by the difference, but I don’t know where it is.
We have a wage gap. I believe that there are societal biases that devalue work by women. I’ve seen them, and I referenced two studies upthread that show this. But as you point out, a very large chunk of this gap is due to factors that are not from direct wage discrimination. They do not account for the entire gap. But I don’t know how much of the remaining difference is due to that discrimination.
Regardless, I think that both these societal biases are bad. I don’t think I’m going out on a limb here.
Regarding the “dangerous work premium”, you might want to drop that one because it doesn’t pan out. A fully efficient market *should *equalize differentials, but real market inefficiencies eliminate the premium, and often it is negative! See DOI: 10.2307/2525246