Weather and Infant Mortality in Africa∗

To what extent have weather fluctuations in Africa affected infant mortality over the last fifty years? We investigate this question by combining individual level data, obtained from retrospective fertility surveys (DHS) for more than a million births in 28 African countries, with data for weather outcomes, obtained from re-analysis with climate models (ERA-40). The focus is on two mechanisms: malaria and malnutrition. We find robust statistical evidence of quantitatively significant effects. Infants born in areas with epidemic malaria that experience worse malarious conditions during the time in utero than the site-specific seasonal means face a higher risk of death, especially when malaria shocks hit low-exposure geographical areas, or hit mothers in the first trimester of pregnancy. Infants born in arid areas who experience droughts when in utero face a higher risk of death, especially if born in the so-called hungry season just after the start of the rains. We also uncover heterogeneities in the infant moratility effects of growing season rainfall and drought shocks, depending on household occupation or education. ∗We are grateful to seminar participants at the IIES, Amsterdam, UPF, UCLA, UBC, Chicago, LSE, Princeton, Berkeley, Oslo, LSHTM, Edinburgh, Nottingham, Houston, Stockholm School of Economics, Gothenburg, SEA Annual Meeting 2009, a CIFAR Meeting, and the 2010 EEA Annual Congress, Sandra Black, Robin Burgess, Angus Deaton, Colin Jones, Ben Smith, Peter Svedberg, and Jakob Svensson, for helpful comments; to Heiner Körnich and Lars Eklundh for asistance with data; to Pamela Campa for research assistance; and to Mistra and the ERC for financial support.