Donors allocate huge sums for assistance intended for poverty reduction to countries that arguably do not need it. For example, the United States spends more in Jordan—with no extreme poverty—than it does in the six countries with the highest poverty rates in the world combined, including the Democratic Republic of the Congo and Somalia. Prior studies have suggested that if global aid were allocated according to need, it could roughly double its impact on poverty reduction.
Country need can be broken down into the magnitude of the development challenge and the domestic resources available. This study examines these two components of development need to determine which countries have the greatest need for foreign assistance.
I produce a set of data-driven observations documenting the magnitude of the challenge and the resources available. I first explore which countries have the highest rates of absolute and multidimensional poverty. I then examine which countries have the ability to redistribute by comparing their poverty gaps with the potential domestic resources available. This produces the marginal tax rate (MTR) required for a country to finance its own poverty eradication.
There is a clear trend in the magnitude of the challenge: low-income countries (LICs) have the highest poverty rates by far; lower middle-income countries (LMICs) are more mixed; and upper middle-income countries (UMICs) have comparatively little extreme poverty, except in a small handful of countries. The implied MTRs required to close a country's extreme poverty gap are excessively high for LICs and most LMICs but easily manageable for most UMICs. This strongly suggests that both the magnitude of the challenge is much greater and the resources available much lower in LICs and LMICs.
The policy implication is that LICs and LMICs should be strongly favoured in terms of aid allocation. The two different components of country need point in the same direction and both suggest a strong focus on allocating assistance towards the poorest countries.