I guess I'm not saying I think you're wrong overall, I'm just saying the effect size is wildly inflated by current sociocultural conditions.
And I would expect American data to be different to the global average, that was my point. I would expect a post-fiat society to be even more different but in the same direction.