Report on Africa aid should not be seen as IMF view, Letter to the Editor By Damian Ondo Mañe, Executive Director, and Peter Gakunu, Alternate Executive Director, IMF

July 8, 2005

Letter to the Editor
By Damian Ondo Mañe, Executive Director, and
Peter Gakunu, Alternate Executive Director
International Monetary Fund
Financial Times
July 8, 2005


As representatives of 44 African countries at the International Monetary Fund, we would like to express our concerns and misgivings about the article on research conducted by IMF economists on aid and growth ("Aid will not lift growth in Africa, IMF warns", June 29).

The headline portrays the research as the Fund's official view. This is not the case, because IMF working papers present the views of the authors and do not necessarily represent those of the IMF or IMF policy. Indeed, the IMF executive board, which is responsible for the Fund's policy direction, has not articulated a clear position on this issue - and on various occasions has acknowledged the need to scale up aid to enable low-income countries to reach the Millennium Development Goals.

The studies show methodological and factual weaknesses. They fail to establish unambiguously the direction of influence between aid and growth, as well as the robustness of econometric results. That does not mean there is no relationship between the two. The variables used in their analyses explain each other rather than explaining growth. In analysing the influence of aid and policy on growth, they treat each as if it had separate effects on growth. Donor aid is contingent on good policies, so each influences the other. The analysis of the impact of aid and fiscal deficits on growth ignores the relationship between aid and government expenditure since some aid (such as budget support) is contingent upon government expenditures.

According to the authors, there is a spurious relationship between aid and growth due to "noise" in the data. This ignores the fact that the presence of such noise does not rule out a predictable relationship between aid and growth. There are analytical techniques to deal with non-stationary data that the authors could have used if they had intended to arrive at conclusive results.

One does not have to rely on econometric estimation alone to demonstrate that aid has not worked in some cases, while it has in others. This is more so when such elusive qualitative variables as country circumstances, good policies and good institutions are involved. The two economists' conclusions are refuted by the evidence of African countries that have emerged from conflict to achieve persistent growth over the past decade because of good policies and good institutions supported by aid.

The authors allege that aid leads to aid dependency.. This conclusion does not always hold - it For this to take place presupposes that recipient authorities are adopting inappropriate policies, a fact not supported by Fund surveillance reports on African countries over the past decade. It is general knowledge that there have been strong improvements in the macroeconomic management and governance climate of African countries. In our view, aid does support growth in low-income countries, and we hope this letter goes some way to clarify this issue.


Public Affairs    Media Relations
E-mail: E-mail:
Fax: 202-623-6278 Phone: 202-623-7100