Age-patterns of famine-related mortality increase: implications for long-term population growth.

In this paper, we reaffirm Watkins and Menken's (1985) conclusion that there is 'little likelihood that famines will be a major determinant of population growth in the future, any more than ... in the past'. We find that age and sex-specific patterns of famine mortality change that have markedly different proportional change in group-specific mortality can nevertheless lead to similar trajectories for population size. Subsequent growth is related more to relative reductions in survival by age than to relative increases in mortality. We also comment on the use of Dyson's conception index as a measure of the onset of famine.