Aid Effectiveness: New Instrument, New Results?

Despite a voluminous literature on the topic, the question of whether aid leads to growth is still controversial. To establish the causality from aid to growth, researchers have relied on various instrumental variables. These must be exogenous to growth and explain aid flows well. We propose a new instrument based on aid quantities as pre-dicted by the starting date and length of development cooperation partnerships, and argue that this is a significant improvement relative to past approaches. We find a positive effect of aid on GDP, with an elasticity of GDP with respect to aid of 0.10. This effect is statisti-cally significant and relatively constant across different specifications. The fact that the effect is so small- ten times smaller than previous estimates- might help explain why the results in the literature have been so mixed