Foreign aid effectiveness in African economies: Evidence from a panel threshold framework

Abstract

The aid-growth literature has been explored using a wide range of econometric methodologies. The evidence of the effectiveness of aid to promote economic growth is mixed, suggesting that the link between aid and growth is complex and may not be well identified by traditional methods. We take another perspective and frame the aid-growth literature within a nonlinear panel threshold framework applied to a panel of selected African economies for the period 1980 to 2007. We also compare our results with the linear and nonlinear-polynomial specification and address potential endogeneity using instrumental variables and dynamic panel estimations. In the linear setting, we find no clear evidence of a positive effect of aid on growth. In the nonlinear setting, we explore four threshold variables capturing various macroeconomic policies. For each threshold variable, we estimate a polynomial model by interacting aid with the considered threshold variable and a threshold model by splitting the sample according to some endogenous thresholds. Although we find no evidence of polynomial effect in the case of aid as a threshold variable, there is weak evidence of a threshold effect with a diminishing return of aid. In a lower regime of past aid receipts less than of 1.5 per cent of GDP, aid appears to have a strong positive and significant effect on growth while the effect is insignificant when aid exceeds that level. We also find that under good policy environment characterized by relatively low inflation, high trade openness and low budget deficit, the effect of aid is greater. These findings suggest that both donors and African aid recipient countries should continue their efforts in strengthening the macroeconomic management of aid.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 2 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.