A diversified aid network could improve growth by reducing volatility. Alternatively, it could harm growth by encouraging waste and corruption. In this paper we test the effect of aid diversification on growth, growth acceleration, and growth duration. Using a large data set with a maximum of 126 countries over the period 1965–2010 and estimating three types of models (panel vector auto regression, binary dependent variable, and duration), we find the following. First, a diversified aid network Granger causes growth. Second, the “growth acceleration episodes” identified following the definition of Hausmann et al. (Journal of Economic Growth, 2005, 10, 303–329) do not seem to be affected by aid diversification. Third, the “growth spells” identified using Berg et al.’s definition appear to be prone to premature termination as a result of aid concentration. Our results appear to be robust to a wide array of tests and alternative measures of aid diversification.

中文翻译：

---

外援网络多元化及其对增长、增长加速和增长期的影响

多元化的援助网络可以通过减少波动来促进增长。或者，它可能通过鼓励浪费和腐败来损害增长。在本文中，我们测试了援助多样化对增长、增长加速和增长持续时间的影响。使用 1965-2010 年期间最多 126 个国家的大型数据集并估计三种类型的模型（面板向量自动回归、二元因变量和持续时间），我们发现以下内容。首先，多元化的援助网络格兰杰带来增长。其次，按照 Hausmann 等人的定义确定的“增长加速事件”。(经济增长杂志, 2005, 10, 303–329) 似乎不受援助多样化的影响。第三，使用伯格等人的定义确定的“增长期”似乎由于援助集中而过早终止。我们的结果似乎对广泛的测试和援助多样化的替代措施是稳健的。