An estimated 2.5 to 3 billion people worldwide rely on biomass fuels to meet their household energy needs. Burning wood, animal dung, or crop wastes in simple stoves, thesehouseholds typically generate high levels of indoor air pollution that adversely affect health,especially of women and young children. Using cross-country data and household data from Pakistan, this thesis estimates theeffect of biomass fuel use on infant mortality and child mortality. The simple cross-countryresults show positive, significant associations between biomass fuel use and infant and childmortality. Perhaps because of considerable strain on the data, these associations do not stand upto a configuration using a cross-country panel with country-specific fixed effects. On the otherhand, household-level micro results show positive effects of traditional biomass fuel use onmortality, although the coefficients are not always well-determined.Taken as a whole, the results support the conclusion that dependence on traditionalbiomass fuels leads to higher risks of child mortality. The estimates imply that a 10 percentage-point decrease in the share of energy consumption due to biomass fuels for a country lowers itschild mortality rate by roughly 7 deaths per 1,000 live births. For a country such as Kenya, in which roughly three-fourths of total energy consumption is supplied by biomass fuels, the implication is that switching completely to non-biomass fuels would, ceteris paribus, lower itschild mortality rate by 38 percent and prevent 54,000 deaths to children before the age of five. On a global level, the health effects associated with traditional biomass fuels potentially rivalthose of the HIV/AIDS epidemic