Should we value human life at replacement cost?

Earlier this week I asked whether we should value human life at replacement cost.  Michael Vassar wrote me in response:

We should value human life at replacement cost, and we actually value generic human life less than that (negatively?), as demonstrated by the existance of late term abortion (there are almost certainly potential adoptees who would pay many women enough to motivate them to go through labor once late in a pregnancy in exchange for the child, especially considering the actual expenses many incur to adopt), but we should value a particular human life at the lower of total preference for the continuation of that life and replacement cost for that life.  For the latter, today replacement cost is infinite, though it needn’t be forever.  For the former, the preference varies dramatically depending on situation, but is frequently very high.  For simplicity’s sake, and because economical thinking is confusing or corrupting to people below a very high IQ threshold, we maintain a convenient fiction of infinite value.  Although we violate this fiction continuously, we do so in an extremely inconsistent manner because the cost of strict obedience or careful public analysis are greater than the costs of irrationality on particular instances.  The "ethical questions" raised by science are rather consistently actually situations where new scientific data challenges the viability of convenient fictions, but in the end it always turns out to be possible to ignore the new data and maintain these fictions, or to integrate the new data into life rather seamlessly without disruption.  However, the frequency with which these questions occur is increasing.