There is evidence that fertility is positively correlated with infant
mortality, and that a child's chance of surviving to maturity increase
s with the level of nutrition, medical care, etc. received in the earl
y stages of life. By modelling parental decisions as a problem of choi
ce under uncertainty, the paper shows that fertility and infant mortal
ity are most likely to move in opposite directions if, as implicitly a
ssumed by existing economic theories, parents believe that there is no
thing they can do to improve the survival chances of their own childre
n. By contrast, if parents realize that those chances improve with the
amount they spend for the health, nutrition, etc. of each child that
they put into the world, then fertility and infant mortality may move
in the same direction. Under such an assumption, the model has the str
ong policy implication that directly death-reducing public expenditure
s are most effective, but stimulate population growth, at low levels o
f development. By contrast, at high levels of development, such expend
itures tend to crowd out parental expenditures, and are a factor in fe
rtility decline.