New Perspectives on Foreign Aid and Economic Development

There exists a reasonably large literature seeking to explain the amounts of aid allocated to developing countries. Various models and econometric modeling procedures have been used. The “recipient need” and “donor interest” modeling approach has been dominant, especially since the publication of the well-known and influential paper of Maizels and Nissanke (1984). ¹ This approach involves separate estimation of two regression equations: one containing indicators of recipient need and the other solely containing indicators of donor interest. The recipient need model assumes that donors are motivated purely for humanitarian reasons and hence seek to allocate aid on the basis of relative development need. The donor interest model is premised on the assumption that donors are motivated purely by commercial, political, and strategic self-interests. Each model is fitted to sample data and conclusions are based primarily on the respective explanatory power of each model. Most studies reject recipient need as a determinant of aid allocation. McKinlay and Little (1979), for example, concluded that there are “no grounds for asserting that humanitarian criteria have any significant direct influence” (243) on aid allocation.

Recipient need/donor interest studies of aid allocation have estimated their econometric models using cross-country data. A number of relatively recent studies have used individual, country-specific time series data. Gang and Khan (1990), Gounder (1999), and Gounder and Sen (1999), for example, have looked at the cases of India, Papua New Guinea and Indonesia, respectively. ² The use of these data as opposed to cross-country data is appealing both on econometric and informational grounds. Econometrically, it avoids the typically restrictive assumption that the regression coefficients are fixed across all countries in the sample employed and more

-11-