[lutorm]

Particularly noteworthy is the point that if a program sets an unusually high value on human life, it diverts resources from other programs also intended to protect human life

It's an interesting thought experiment, but of course that's not really relevant to the example at hand. If NASA insisted on 1e-10 astronaut risk, it would not divert any money from other programs that protect human life, because the NASA appropriations are not for that purpose. It would just mean that NASA got nothing done, and maybe eventually it would be shut down. But assuming that there's a rational and efficient allocation of resources for a clear goal seems just as much a fiction as other economic concepts like "efficient markets".