A replication of ‘The long-run impact of foreign aid in 36 African countries: Insights from multivariate time series analysis’ (Oxford Bulletin of Economics and Statistics, 2014)

Abstract

Macroeconomic data have been shown to vary substantially between sources, especially so for low-income countries. While the impact of data revisions on inference is well documented for cross-country studies, there is no systematic analysis of the robustness of results obtained from time series analysis. This is despite the fact that time series analysis is an integral part of the econometric toolkit of government analysts, and informs policy decisions in many areas of macroeconomics. This study fills this gap for the notoriously controversial aid-effectiveness debate using the statistical framework by Juselius et al. (2014, Oxf Bull Econ Stat): by adopting alternative sources of GDP data in 36 sub-Saharan African countries the author finds that results remain robust across datasets in two thirds of the countries, but sometimes drastically change in others. These findings suggest that robustness checks such as those carried out here should become standard procedure for macroeconomic analysis using single-country time series.