This study compares the cost effectiveness of various health inputs and government programs in reducing race-specific neonatal mortality or death in the first 27 days of life. Approximately two thirds of all infant deaths occur within this period. The programs and inputs at issue are teenage family planning use; the supplemental food program for women, infants, and children (WIC); use of community health centers and maternal and infant care projects; abortion; prenatal care; and neonatal intensive care. Using an economic model of the family as the analytic framework, effectiveness is determined by using ordinary least squares and two-stage least squares to estimate infant health production functions across large counties in the United States in 1977. Estimates of costs are from a number of published sources. We find the early initiation of prenatal care to be the most cost-effective means of reducing the neonatal mortality rate for blacks and whites. Moreover, blacks benefit more per dollar of input use than whites. Neonatal intensive care, although the most effective means of reducing neonatal mortality rates, is one of the least cost-effective strategies.