In a community where everyone is genetically and epigenetically identical, variance attributable to additive genetic factors (heritability) is zero. In a commu

In a community where everyone is genetically and epigenetically identical, variance attributable to additive genetic factors (heritability) is zero. In a community where everyone is genetically diverse but their living and work situations are exactly identical (a very hard scenario to imagine), all variance is attributable to genetic factors. (The heritability is potentially 1, but since the definition of heritability is that it pertains to additive genetic factors, it may not be 1 in this case.) A few things flow from this. The first is that population-level estimates of heritability may change over time, even if only because environments change, sometimes wildly, over time. The second is that estimates of heritability for one population may not tell you about another population, and the population-level estimate itself is a blend of the community-level constituent values. (Even when you derive your estimates from twin studies, you can imagine how this is true.) The third is that even if you drive heritability to 1 because you have made everyone's environment uniformly awesome, this does not mean that you cannot make everyone's environment even more awesome and thereby increase productivity/output/whatever. Say that Trait X is one's "widget production ability." If Trait X's variability is now 100% genetic, it does not at all follow that Trait X's mean cannot be increased by changing the environment. (It merely follows that Trait X's variance, not its mean, is currently completely explained by genetic differences.) The fourth is that with substantial state intervention, you can probably manipulate the heritability of any human behavioral trait to whatever value you want it to be, given enough dedication of resources and willingness to flout norms.