Aid delivery methods have had many “names” over the past 50 years ranging from projects, to programmes, to integrated development projects, to sector approaches, to sector budget support or direct budget support. In this plethora of “titles”, donors are not agreeing amongst themselves. In addition, countries and practitioners are sometimes confused, if not lost. They have to redefine the meaning of the words used for every assignment. In this situation, how can one begin to discuss content rather than form? In some situations, donors even argue against the words used in their own financing agreement, spinning a definition of their support that is not included in the title (such as a financing agreement giving entitlement to budget support, but being considered as programme support). Clearly, this is not only confusing for the practitioners but also misleading for countries in terms of aid flow predictability, control and reporting. Different aid delivery methods require different management, i.e. aid management, increasing the transaction costs. It does not support a basis for donor coordination and harmonisation and certainly does not allow easy and practical aggregation of data.
A simple classification framework may thus be useful for aid delivery methods. This can be constructed to allow for the classification of:
– aid flows into on or off budget according to their intended purposes, and
– on-budget financial flows according to the “path taken by the money” to enter the economy.
The first objective can be responded to: domestic revenue shortfalls do not allow the delivery of the basic services required for socio-economic development (achieving the Millennium Development Goals) and for the economy to generate sufficient broad-based wealth to raise revenues. Governments are committed to such developments whether to their citizens or through ratifying international agreements. Therefore aid is provided to support service delivery until governments’ own revenues have increased sufficiently and budget processes and stakeholders recognised the importance of the objectives to allocate necessary resources. The responsibility for setting standards on service delivery is a shared one between countries, citizens and donors.
The underlying principle of aid is thus to work within a country to support the delivery of services by directly providing these services (substitution) or by reinforcing the means to deliver the services (investments, reforms, technical cooperation, etc) through government’s systems and procedures.
Thus independent of the aid delivery method chosen, the resources channelled that meet this intent, should be “on-budget”. They can be broadly categorized as a form of support to a government’s budget, as the prime instrument to implement its policies, either through cash or through goods, works and services being purchased. This prevents the governments having to mobilize these resources through domestic revenue collection or debt. Yet accounting for them through the budget system is still required to ensure that the government is accountable to its citizens, taxpayers and aid beneficiaries (three groups which largely over-lap each other). This is independent of the form the aid takes, how the aid is managed and who delivers the services.
Ultimately, donors may want to work with governments to support and/or influence their policy-making processes. This puts the emphasis on long-term objectives and achievements that are best measured by citizens’ satisfaction (hence the link to the political process) rather than by simple quantitative measures of outputs.
Any other aid delivery method that does not seek to substitute to or to enhance a government’s service delivery and activities should be considered “off budget”, independent of the provider and of how aid is managed. The particular case of aid going through NGOs or other project management forms to deliver services at the local authority level should also be reflected in the central government’s budget as grant-in-aid and in the local authority’s one too (different levels of accountability).
Having clarified what constitutes “on” and “off budget”, in real terms, it becomes possible to classify the aid received through different aid delivery methods. Two categories can be broadly defined: Direct Budget Support and Indirect Budget Support. Such a classification though formal, would help the dialogue amongst donors and with partner countries, especially when appraising the potential impact of aid on the country’ institutions and the objectives that can be reasonably pursued. Even if the focus changes to be centred on the final beneficiary rather than the institutional arrangement (as in the case of cash transfers in Baluchistan and North West Frontier Province in Paskistan), one could still make the case that it constitutes a different form of procurement for delivering services.
The test for the classification should be one condition:
Does the financial flow enter the economy through the government’s financial systems (the Treasury or the financial systems of a line ministry, for example, the Ministry of Health) or does it enter it through contracts for works, goods and services procured to implement the support (paid for directly by the aid agency)?
Any support that is non-financial will necessarily be Indirect Budget Support. In any situation where the support is financial but does not go through the government’s financial systems, it is termed indirect budget support. If an aid method qualifies as budget support, but is not reflected in the budget and its accounting and reporting instruments, for lack of an adequate budget process or lack of donor information, the definition remains the same, but the practice is found not to be in compliance with the commitments given in the Paris Declaration and the Accra Agenda. This places a particular onus on donors to provide adequate information on their support (including accounting standards, classification, timeliness and periodicity).
When financial support is provided through the government’s financial systems, it is considered to be Direct Budget Support, irrespective of the conditions laid down in the financing agreement, which may place more or less emphasis on a particular sector.
 This also implies that creating pressures on governments to address service delivery in new areas or for new beneficiary groups imply that donors take a responsibility in providing sufficient resources for a sufficient transition period.
 Flows entering through the Single Treasury Account are direct budget support. However, flows entering through the government financial systems, e.g. a ministry’s account, are also direct budget support. The fungibility argument can be applied to all forms of budget support as they finance services or activities of the government; the question raised is about the predictability of the flow and the discretion with which a budget holder can use the funds.