For evidence on the value people put on their own lives, see:
W. Kip Viscusi, "The Value of Life: Has Voodoo Economics Come to the Courts?" 3 J. Forensic Econ. 1, 4 (1990).
____Reforming Products Liability, Harvard University Press, Cambridge 1991.
____"Risk by Choice: Regulating Health and Safety in the Workplace, ch. 6 (1983).
R. Thaler & S. Rosen, "The Value of Saving a Life: Evidence from the Labor Market," in Household Production and Consumption 265 (Terleckyj ed. 1975).
In this chapter I discuss the value of life as measured by the value that individuals place on their own lives. An interesting and related issue is the price of life implied by the observed relationship between income and life expectancy.
On average, richer people live longer, for a wide variety of reasons. Not only does additional income result, on average, in better medical care and nutrition, it also lengthens lives in less obvious ways. Better maintained cars are less likely to have accidents. Well off people are less likely to find themselves having to go out in bad weather, or work long hours when they are tired, or take chances with food that probably isn't spoiled.
One implication is that policies which cost money also cost lives. Suppose that one result of increasingly stringent drug testing to avoid unexpected side effects is to increase the cost of drug development by (say) ten billion dollars a year. That is a real cost--resources are being spent on testing that would otherwise have been spent on other things that people value. The result is to make people on average poorer--hence, on average, likely to die earlier.
A variety of studies have attempted to estimate the relation between income and life expectancy--to determine how much of an increase in real income it takes, statistically speaking, to save one life. Results vary, with a range from about 1.8 to 12.4 million dollars. Suppose the correct figure is five million dollars per life. In that case, the hypothetical drug program has the indirect effect of increasing mortality by 200 lives a year. If the number of lives that the program saves by eliminating drugs with serious side effects is less than that, then, on average, making drugs safer makes people less safe.
For a much more extensive discussion of this issue, including cites to the empirical literature, see Health-Health Tradeoffs, a webbed working paper by Cass Sunstein.