> It probably relates because the marriages themselves don't matter, it is the family that forms and the impacts that has on individuals that do.
That's speculative, and not at all what the study found.
It could be that that's what's important; it could be, instead, that what is important is the social expectation of marriage combined with the opportunities available for it. It could be something else. It could be a combination of any of the three.
> Thus if you had some society which allowed only monogamy but also allowed concubines, you would likely see the same issues as caused by polygamy.
Since concubinage is generally exclusive in the same way as marriage is (and exclusive of marriage), you'd expect in pretty much any case it to have similar social effects to polygyny (if it isn't just considered a form of polygyny in which there is a wife with a formally preferred status, which is common in polygyny.)
> Or if you had a society that didn't do marriage at all, but the relationships that formed tended to be greater towards polygamy than monogamy.
OTOH, if it is the social expectation and opportunity aspect of marriage that is at issue, a society which lacked the formal marriage and social expectations tied to it might not have the problems of a formally polygynous society even if the informal relations formed were generally polgynous rather than monogamous.
And if, on yet another hand, there is a stabilizing effect of socially-recognized relationship, the absence of social recognition might produce problems similar to those in a formally polygynous society regardless of the form of the informal relationships that form.