Aid to conflict‐affected countries: lessons for donors

Click to increase image sizeClick to decrease image size

Notes

Tony Addison and Mark McGillivray are members of the World Institute for Development Economics Research, of the United Nations University (UNU‐WIDER), Helsinki, Finland. This paper is linked to the post‐Conflict Reconstruction and Aid Effectiveness research projects conducted at UNU‐WIDER. An earlier version of this paper was presented at the Security and Development Workshop of the Global Development Network Annual Development Forum, New Delhi, India, February 2004.

They include Svensson, 1999; Hansen and Tarp, 2000a, 2000b; Lensink and Morrissey, 2000; Collier and Dehn, 2001; Dalgaard and Hansen, 2001; Gounder, (2001, 2002; Guillaumont and Chauvet, 2001; Hudson and Mosley, 2001; Lensink and White, 2001; Lu and Ram, 2001; Dalgaard et al., 2002; Collier and Hoeffler, 2002; Chauvet and Guillaumont, 2002; Gomanee et al., 2002, 2003; Mavrotas, 2002; Easterly et al., 2003; and Roodman, 2003. See Robinson and Tarp, 2000; Beynon, 2001, 2002; and McGillivray, 2003a, 2003b for reviews of much of this literature.

Roodman takes Collier and Dehn (2001), Guillaumont and Chauvet (2001), and Burnside and Dollar (1997, 2000) to task, arguing that their econometric results are not robust. Collier and Hoeffler (2002) have not yet been subjected to such a critique. Aid does not appear alone as an explanatory variable in their model, and it would be useful to test the robustness of their estimates by including this term. As they stand, though, the Collier–Hoeffler results do seem quite robust, and the model from which they have been obtained has quite a sound theoretical base. This is not to say that more empirical validation of the impact of aid in post‐conflict scenarios is not required.

This number has been calculated using data obtained from OECD (2003b) and the country classifications in Collier and Hoeffler (2002).

This is not to imply that the general trend in aid away from project support towards budgetary support is unwelcome. It is to say, though, that such an approach, which requires well functioning institutions and (possibly) good policy regimes, is inappropriate for some countries.

Here we use the concept of tradability in its macroeconomic sense, not in the sense of whether the good is sold in a market or not.

See, for example, Ndikumana (2001) on Burundi and Rwanda.

Additional information

Notes on contributors

Tony Addison

Tony Addison and Mark McGillivray are members of the World Institute for Development Economics Research, of the United Nations University (UNU‐WIDER), Helsinki, Finland. This paper is linked to the post‐Conflict Reconstruction and Aid Effectiveness research projects conducted at UNU‐WIDER. An earlier version of this paper was presented at the Security and Development Workshop of the Global Development Network Annual Development Forum, New Delhi, India, February 2004.