This paper extends Agenor and Aizenman (2004) in a number of ways. First, the paper considers an array of developing countries from different regions in Africa. Asia and America. This would help to see if the results are sensitive to region specific economic conditions. Second, the issues of joint endogeneity of all explanatory variables in a dynamic formulation and of potential biases induced by country-specific effects are directly addressed. This provides more consistent estimators. Third, a battery of sensitivity analysis is conducted to check the robustness of the results. These include the use of alternative determinants, estimation dropping one country at a time in order to identify an outlier country, country groupings, etc. finally, an extreme bound analysis is performed to test thereliability and robustness of the relationship between terms of trade shocks and the savings rate.