Version classiqueVersion mobile

The Effect of Foreign Aid on Sub-national Development

Dumebi Ochem

1. Introduction

Texte intégral

1The question of whether official development assistance is effective in promoting economic development is one that has been heavily discussed over the past decades. This debate is particularly relevant in the context of Africa, which is one of the largest recipients of development aid. According to Aiddata (2017), the World Bank disbursed over 75 billion US dollars towards 1,907 geo-referenced aid projects in African countries between 1995 and 2014, with the explicit mission to “end extreme poverty by 2030”.

2However, no consensus has been reached on whether foreign aid is actually effective, due to endogeneity concerns in the allocation of aid. This impediment has led researchers to pay more attention to the effects of aid at the sub-national level, which still represents a clear knowledge gap in the literature. In addition, existing studies generally assume the effects of foreign aid to be linear, when in reality it is very likely that these effects would vary at different levels of development within countries.

3In this paper, I analyse the heterogeneous effects of geo-referenced World Bank aid projects on economic development within African countries, at the disaggregated sub-national level. The data framework is based on a standardized grid cell structure, where the unit of analysis is a single grid cell at a 0.5 x 0.5 decimal degree resolution. This unique approach effectively addresses endogeneity issues related to reverse causality or sample selection, as aid donors are unlikely to allocate funds according to socio-economic conditions within arbitrarily constructed grid cells.

4Using a fixed effects quantile regression approach, I estimate the impact of foreign aid at different levels of development across 10,674 grid cells, over the period of 1992 to 2014. As a supplementary strategy, I also adopt an instrumental variable approach, following the methodology developed by Dreher & Lohmann (2015), which exploits a plausible quasi-experiment created by an eligibility threshold set by the International Development Association (IDA).

5The baseline results suggest a positive and statistically significant effect of foreign aid on development, measured as night-time luminosity. For instance, looking at the 75th percentile, the presence of an aid project would lead to a 15.9% increase in night-time luminosity in subsequent periods. In addition, this marginal effect is largest within relatively poorer grid cells, suggesting that World Bank aid is in fact achieving its primary objective. Moreover, the estimations show clear evidence of spill-over effects of aid from neighbouring grid cells, which may even dominate the within-cell treatment.

6These findings are however quite sensitive to different specifications and variable transformations, specifically the inclusion of year fixed effects and the use of the inverse hyperbolic sine transformation. Indeed, the coefficients become significantly smaller in both cases. The 2SLS results also diverge from the baseline estimates, showing a negative effect of foreign aid on night-time lights. Given all the findings above, I conclude that foreign aid is generally effective in fostering development at the grid cell level, but the economic impact may be smaller than initially predicted.

7The remainder of the paper is structured as follows. Section 2 provides an overview of the literature regarding the effectiveness of aid and the main empirical limitations. Section 3 describes the sample data and its construction. Section 4 discusses the identification strategy. Section 5 presents the empirical results and additional robustness checks. Section 6 concludes.


Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search