Due to limited resources, the cost-effectiveness of aid is an important issue for donors and policymakers. So what models are used to measure the effectiveness of aid?
One important example comes from the health sector. The World Health Organization Choosing Interventions that are Cost Effective (WHO CHOICE) is an initiative which started in 1998 with the aim of developing clear and simple cost-effectiveness analyses (CEA). The models analyse and compare the cost effectiveness of different health interventions (both individually and together in packages). The results are measured in terms of the average number of DALY (disability adjusted life years) per funds spent, which is the number of healthy years of life gained, per dollars spent (DALYs/US$1). A similar measure is used by the Disease Control Priorities Project, led by the World Bank. Figure 1 shows the results of one such study on the effectiveness of different strategies to address HIV/AIDS. Education seems to be the most cost effective treatment. A second example is found in the field of malaria; and figure 2 shows that in this case the cost effectiveness of the different approaches is roughly uniform.
So what are the possible drawbacks of using such models? First of all they may overlook other priorities, such as cultural and social concerns (for example they may not consider priority for the sick, reducing social inequalities in health or the well- being of future generations), while trying to measure outcomes in terms of a single indicator. In addition, the definition of costs and budgets is a complex issue which may need to involve the evaluation of nonmonetary benefits as well as monetary costs. For example, an intervention that is considered justified by cost effectiveness may be impossible to deliver if the costs are monetary and involve the public budget while the benefits are non monetary and diffused over the population. Economic theory would suggest removing the current budget constraint by raising more revenue (usually taxes) until the marginal social cost of the intervention plus the cost of obtaining the revenue equals the marginal social benefit. But this may not be possible for political reasons or because the economic cost of raising extra taxes is prohibitive. Finally, there is a problem of lack of generability to different contexts due, mostly, to data unavailability in many developing countries regarding the health sector (little knowledge about the labour force for example); this problem results in a higher uncertainty in the generalization of these models. Therefore these models may fail to direct special needs of the population, for example in the proximity of a humanitarian crises Still, there is little doubt that more attention will be given to such models in the future as a way to go beyond case studies and scale up the results of aid packages in the long run. But more data and more information are essential in this process.