U.S. Food Aid and Infant Mortality Rates: An Instrumental Variables Approach

This thesis studies the effect of US wheat aid on infant mortality rates in developing countries. There is debate on the effectiveness of US food aid; some claim it disrupts local food production, while others discuss its role in prolonging conflict. This paper aims to address the intended impact of food aid, feeding people, rather than unintended impacts tackled by previous studies. Infant mortality rates serve as a measure of the health of pregnant woman and infants, who make up a vulnerable population that is susceptible to food crises. An instrumental variable approach is taken, which uses lagged US wheat production, a country’s tendency to receive any US food aid, a rainshock variable, population, and a measure of intrastate conflict, to determine the impact of wheat aid on infant mortality rates in recipient countries. As shown by the results, infant mortality rates decrease with more US wheat aid, which is conducive to the goals of food aid set out by USAID. Specifically, a 100% increase in US food aid, decreases infant mortality rates by 19.3 deaths per 1,000 live births. Furthermore, the effect of US wheat aid on infant mortality rates is strongest in countries that are more likely to receive aid compared to those with a below average propensity to receive US food aid.