Version classiqueVersion mobile

The Effect of Foreign Aid on Sub-national Development

Dumebi Ochem


Texte intégral

1In this ePaper, I examine the sub-national effects of geo-referenced World Bank aid projects on economic development within African countries, measured as night-time luminosity. The unique data framework is based on a standardized grid cell structure, where the spatial unit of analysis is a single grid cell at a 0.5 x 0.5 decimal degree resolution. The advantage of this approach consists in its ability to mitigate endogeneity issues related to aggregation bias, reverse causality or sample selection.

2Using a fixed effects quantile regression approach, I estimate the non-linear effects of foreign aid at different levels of development across approximately 10,600 grid cells over the period of 1995 to 2012. I also adopt an instrumental variable approach as a supplementary identification strategy, in case there may still remain any unaccounted-for endogeneity. I instrument the foreign aid indicators with a binary variable indicating whether a country is below the IDA’s income threshold for eligibility to concessional aid, interacted with a grid cell’s probability of receiving aid over the whole sample period.

3Baseline results suggest a positive and statistically significant effect of foreign aid on night-time luminosity, albeit small in economic terms. Looking at the median effect, a one-unit increase in the number of projects would lead to a 2.2% increase in nightlights in following periods. In addition, the results imply that the marginal effect of aid is greatest within poorer regions, as the coefficients at the 95th percentile are consistently smaller. Moreover, the estimations using a spatial lag model show clear evidence of spill-over effects of aid from neighbouring grid cells, which may even dominate the within-cell treatment. For a given grid cell, the presence of an aid project in at least one of its neighbours leads to a 7.5% increase in night-time luminosity.

4The above findings are however fairly sensitive to different model specifications and variable transformations, namely the inclusion of year fixed effects and the use of the inverse hyperbolic sine transformation. In both cases, the coefficients are significantly smaller than those from the baseline model, which may raise concerns regarding previous studies that have overlooked these specifications.

5Furthermore, the 2SLS results diverge substantially from the baseline estimates, showing a negative effect of foreign aid on nightlights. This is likely a result of the severe weakness of the instrument, or a potential violation in the exclusion restriction. Hence, these findings may cast doubts on the reliability of instrumental variables to address the endogeneity of aid.

6To conclude, the above findings suggest that foreign aid is moderately effective in fostering within-country development at the grid cell level. However, further research in this context is necessary to obtain more conclusive results. For instance, it may be more advantageous to study the impact of aid on specific development outcomes, such as health and nutrition, educational attainment or female empowerment. Nevertheless, an important next step would be to move away from reduced-form analyses, and instead develop a theoretical framework to better explain the causality chain, as well as any indirect mechanism, between aid and economic development.


Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search