Aid, agriculture and poverty in developing countries

The authors make two contributions to the debate on aid-effectiveness, illustrating that for impact on poverty what matters is not just the level but also the composition and stability of aid. One specific implication of this for aid policy is that aid most effectively reduces poverty if it supports public (and other) expenditures which are supportive of agricultural development. Regression analysis confirms that these are not only direct expenditure on agriculture, but also on education and infrastructure, and military expenditure has a negative impact. Three factors appear to be particularly conducive to the development of stable pro-poor expenditure patterns (and in particular pro-agriculture expenditure patterns). These are expenditure strategies which protect the poor against risk, the development of stable relations between governments and aid donors, and long-term political commitment to pro-poor strategies by government. The argument is pursued partly by panel-data econometric analysis of developing countries as a whole, and partly by case studies of sustained and non-sustained green revolutions in heavily aid-dependent countries in Africa.