What is the threshold beyond which aid loses its effectiveness? How do recipient resource allocation strategies affect development? This paper for the Center for Global Development focuses on the problem of proliferation of aid projects and absorption. Using a sophisticated mathematical modelling process, the author analyses the relationship between total aid and recipient activity, and the distribution of projects by size. An absorption threshold is reached when development maximisation calls for the recipient to withdraw from donor-backed projects, but the recipient does not, for the sake of private benefit (‘throughput’).
There has been a lengthy debate among academics and policy makers about the impact of aid quantity. Recently this debate has been widened by considerations of aid quality, which can be defined as the capacity per dollar for aid to increase development and reduce poverty. Increasing aid quality entails a number of themes including untying, selectivity, harmonisation, alignment, coordination, and reducing proliferation (of many small projects). The Paris Declaration 2005 defines 12 indicators of donor aid quality, and commits donors to meeting them. The paper focuses on the issue of project proliferation, which imposes administrative burdens on recipient governments. It models aid delivery as a set of production activities with two inputs, donor aid and recipient resource, and two outputs, development and throughput.
Throughput is defined as the private benefits of implementing aid, which can accrue to both recipients and donors. For example, for donors the career benefits of being associated with successful projects are throughput, while per diems, overseas training, and top-up salaries are examples for recipients. Application of the model suggests that:
- If the recipient is purely development-focused, and cares not at all about throughput, increasing aid always increases development.
- But if the recipient cares only about throughput, increasing aid can reduce development. In particular, if the recipient cares only about throughput and the donor moves aid from a project with lower marginal productivity to one with higher (because of scale economies), development goes up. Otherwise it decreases.
- There are significant sunk costs for recipients, which are the costs required to run the project. If recipients cover sunk costs, projects go forward. If they don’t, donors shut them down.
- A threshold is revealed when development maximisation calls for the recipient to withdraw from donor-backed projects, but the recipient does not for the sake of throughput.
The threshold for aid effectiveness introduces a notion of absorptive capacity. If donors break out of current proliferation patterns, they may be able to push the threshold back if there are scale economies to be exploited in some sectors. The policy implication of this is to emphasise sectors where scale economies are more likely, for example infrastructure. Other policy implications are:
- When donors are contemplating large aid increases, they need to be sensitive to indications that relatively developmental recipients are resisting proliferation.
- In countries with limited budgets, monitoring and accountability requirements can push the aid delivery process against structural limits and undermine effectiveness. Donors should consider funding fewer, larger aid interventions in these circumstances.
- More research is needed on issues relating to sunk costs and understanding why recipients refuse to invest resources in aid projects.
