How should aid be provided? Aid appears to be at its best when it is designated to finance beneficial projects. If aid finances a school or a health clinic then surely it is useful. Unfortunately, this is often an illusion for the simple reason that such projects may well have been undertaken by the government even in the absence of aid. The aid may ostensibly be assigned to schools and health clinics, but actually finance quite different activities that otherwise the government would not have been able to afford.

Once this problem with project aid became well understood, the European Commission and many other donors faced reality and decided to provide much of their aid directly as support for government budgets. The government received the money and was free to spend it however it chose. An advantage of ‘budget support’ was that, while not changing how aid was actually used, it substantially reduced the administrative costs of financing projects. However, it did not address the deeper reason why donors had often preferred projects: the governments of poor countries were sometimes either unwilling or unable to serve the interests of their citizens.

In some of these countries, democratic accountability of government to citizens is a sham: where government officials are unwilling to further the interests of their citizens there is little to stop them. Donors are realizing that in such contexts budget support is unwarranted and that even project aid does not overcome the problem. One alternative is to bypass central government and provide finance directly to those local governments which donors judge to be better aligned with the interests of their populations.

In other low-income countries the government does aspire to serve the interests of its citizens, and indeed understands those interests better than the donors. These are potentially the right environments for budget support. However, even where governments are willing to serve the interests of their citizens they may not have the administrative capacity to do so. Often, the civil service has deteriorated to the point at which the objectives of good political leaders are frustrated by the corruption of the officials who are needed to implement them. How should donors handle such situations? Currently, they usually turn a blind eye to deficiencies in the mechanics of budget processes – how public money is actually spent as it works its way through the bureaucracy. They simply hope for the best. What else might they do?

What is needed is an assessment system, which certifies whether a budget system has the administrative integrity necessary for donors to be confident that direct support to the budget would not be diverted into corruption and hidden uses such as illicit military spending. To avoid political pressures that might influence professional judgment, the assessment should be independent of any donor. The obvious agency to undertake it would be the International Monetary Fund, which already reports on budget processes. Currently, the Fund does not take the extra step to certify whether a system is suitable for budget support. If it was reluctant to do so, the function could be contracted out to one of the international accounting firms.

When I have discussed the idea of certification with donors the usual response is ‘but how many of these countries have budget systems that would meet the standard for certification?’ The implication is that donors recognize that, currently, budget support is usually being channelled into budget systems that cannot be trusted. However, in this case the right action is not to bury the problem but to deal with it.

In poor countries the reaction to my idea has been far more positive: honest officials know that their systems currently face problems and judge that the challenge of certification would provide the impetus for internal reform. Indeed, in other contexts the European Union has often created thresholds as a device for encouraging improved performance. What would happen where a country currently in receipt of budget support failed to gain certification? Clearly, some transitional arrangements would be needed: the government might, for example, be given three years to raise standards to the necessary level. During this period the donors would provide substantial technical assistance on how to improve the budget system, and at its end the independent assessment would be repeated.

Such a certification would bring one huge advantage. Where a budget system was certified there would be no reason to provide aid in any other way. The American Administration would have the evidence to persuade Congress to rethink its opposition to budget support.

If you found this article interesting please consider donating. Thank you very much for your support!