I'm agnostic about the GMV hypothesis, but I will note that is generally taken very seriously in behavioral genetics. Interestingly, the example of the significant female advantage in life expectancy is a relatively recent phenomenon thought to be significantly influenced at present (at the population level) to differences in smoking rates. Pregnancy and childbirth morbidity and mortality used to neutralize the difference or even tip the scales in favor of men in some places in the past. Nature might not be fooled, but we can be fooled about nature, especially when complex behavior is involved.