Aid Flows
The Dutch disease in Africa

One of the main purposes of development aid has always been the creation of a stronger export sector, in the hope that the increased export revenues so generated would eventually do away with the need for aid as a source of foreign exchange. Export performance in Africa has been almost uniformly disappointing, however, although this continent has always been a major recipient of aid. High real wages and steady upward pressure on the real exchange rate (defined as the relative price of home or non-traded goods to traded goods) appear to have thwarted export growth. These trends are documented in the World Bank's 1984 study of Sub-Saharan Africa. In Discussion Paper No. 88, Research Fellow Sweder van Wijnbergen points out the possibility of a causal link between development aid and poor export performance: Africa, he says, may suffer from the 'Dutch Disease'.

Van Wijnbergen notes that aid, like many otherwise effective medicines, has unwanted side-effects. Policy-makers realise that aid will exert upward pressure on the real exchange rate, lead to increased labour costs in those sectors of the economy producing traded goods, and reduce external competitiveness. A substantial increase in the volume of aid inflows may therefore lead to reduced export performance. If this is to be avoided, policy- makers must introduce measures explicitly designed to reduce the conflict between export promotion and aid, argues van Wijnbergen.

Van Wijnbergen outlines the circumstances under which this conflict is particularly severe and the need for policy adjustment most urgent. Increases in aid will permanently reduce productivity in export sectors when externalities such as 'learning-by-doing' are present in the traded-goods sector. When there is learning-by-doing, productivity in a sector depends on cumulative output over time. Aid, by reducing output in the export sector, causes the industries in this sector to suffer an absence of 'learning' and a loss of productivity. Policies explicitly designed to promote exports are called for in this case, and this argument is strengthened by capital-market imperfections.

Under such circumstances, increased aid will magnify the cost of trade policies which are 'inward-oriented' and biased against exports. There is a very real possibility of a vicious circle, according to van Wijnbergen. Distortionary trade policies which are biased against exports lead to poor trade performance and lower real income, which in turn makes the country involved a more likely aid recipient. Increased aid will then lead to further deterioration in trade performance and increase the dynamic costs of anti-export policies, while at the same time allowing their continuation. This analysis, van Wijnbergen argues, supports the case for aid conditionality: aid should be conditional on the elimination of any bias against international trade that is present in domestic tax or trade policies.
Van Wijnbergen concludes by presenting empirical evidence for six African countries of the relationship between increases in real aid flows and appreciation of the real exchange rate. It is this relationship that underlies the possible conflict between aid and export performance. Moreover, in the theoretical analysis, upward pressure on real product wages in the traded-goods sector was identified as an important element of the mechanism through which aid leads to real appreciation. The econometric evidence presented by van Wijnbergen strongly confirms this relationship between higher aid flows and increased real product wages in the traded-goods sector.


Aid, Export Promotion and the Real Exchange Rate: An African Dilemma?
Sweder van Wijnbergen

Discussion Paper No. 88, December 1985 (IT)