Publisher/Institution: University of California, Riverside

Abstract: This study estimates the effects of a large scale trade policy shift on a development indicator like infant mortality, using the recent experience of African Growth and Opportunity Act (AGOA) affecting sub-Saharan Africa. The average difference in probability of death of children born to the same mother before and after AGOA in both AGOA affected and not-AGOA affected countries were estimated, using the retrospective Demographic and Health Surveys (DHS) from 30 sub-Saharan African countries. This helps in exploiting the within-mother variation, which is an improvement over the cross-country studies carried out till date. The identification assumption is that timing of AGOA is exogenous with respect to timing of birth across siblings, and mother fixed effects will difference out time invariant characteristics which may bias results. Findings suggest that infant mortality falls by about 7 to 13 infant deaths per 1000 which is as much as 9% to 16% of the sample mean. This result is robust to controlling for country specific linear trends and country level time varying indicators like GDP per-capita, average female literacy, commodity price index and political regime of the country. It was also found that uneducated women and rural women experience significant decreases in infant deaths. Moreover, the poor experience a large decline compared to non-poor implying trade does not reinforce inequality (in health outcomes). At the macro level, the effect seems to be taking place via increase in GDP per capita and increases in health expenditure per capita.