General budget support – a means to what end?

Development Policy05 Feb 2008Oliver Morrissey

A role for the (inter)national private sector in development and the importance of results-oriented accountability.
These two issues are problematic and their juxtaposition even more so. Supporting private sector development in recipient countries is a legitimate objective of development agencies. However, beyond aligning donor activities with recipient development strategies (and influencing such strategies), there is little guidance on how best to achieve this. Donor (i.e. taxpayer) support for the activities of private sector agents, including NGOs, has no solid basis as such agencies are not publicly accountable (in either the donor or recipient countries).

Results-oriented accountability, which is not consistently imposed on private agents, is at best limited and at worst inappropriate in application to recipient government activities. The reason is simple: rarely if ever do donors have mechanisms to attribute specific outputs (results) to specific inputs (aid) in their own projects, and any such link is even more difficult to establish for government activities. One could reasonably consider results as depending on inputs (aid plus government resources) plus a stochastic variation. Any results-oriented approach that fails to allow explicitly for the latter is inherently deficient. Too often, donors attribute any weak correspondence between inputs and results to government failure, ignoring donor and private sector failures and neglecting the stochastic element.

General budget support

The inherent perception of GBS seems to be prejudiced as the evaluation criterion is phrased in terms of a ‘ positive effect of GBS on poverty reduction and economic growth [which therefore remained little more than a theoretical assumption]’. It is more appropriate to view GBS is a mechanism or modality for delivering aid to the government rather than an instrument for achieving development outcomes. In general, GBS functions best as a mechanism for supporting government expenditures, if: i) donors and recipient governments have agreed the pattern of allocation of spending across headings (spending preferences are aligned), and ii) donors are coordinated in offering GBS. If these conditions are met, GBS can facilitate more effective budget planning by providing a predictable source of aid revenue with low transaction costs (governments do not have to negotiate separately with many donors on each sector).

The IOB appears to treat GBS as a poverty reduction instrument, which it was never intended to be. GBS is nothing more than a mechanism for delivering more effective donor support to (accepted) government expenditure plans. As the commentary on the following quotes demonstrates, the Dutch aid approach seems to have missed these crucial points:

‘ In practice GBS often proved to be a ‘balancing item’. If so, donors should accept some blame as they should not have allowed this to be the case. If donors collectively permitted ‘ incidental budget support to reach almost one-third of all GBS expenses between 1998 and 2006’, then they should accept the blame for failing to implement the GBS principle effectively.

The IOB’s conclusion ‘ that GBS is de facto a political instrument’ can only be construed as a mistaken donor inference. Designed and implemented properly, GBS should support accepted plans. By implication, therefore, it is not amenable to manipulation (i.e. cannot be a political instrument). If recipients did not stick to the agreed allocation of expenditures on which GBS should be predicated it would be legitimate to require recipients to justify any deviation and renegotiate entitlement to GBS.

‘ The evaluation shows that those involved in decision making regularly disagreed about whether or not GBS should be given’. By implication, the officials involved had no agreed understanding of what they were trying to achieve, and probably made things worse rather than better.

The IOB warns that ‘ GBS leads to increased donor dependency: what will happen to the salaries now paid to teachers and nurses if donors decide to cut down on GBS?’ Again, this suggests a misunderstanding. Any aid, GBS or otherwise, can engender dependency and any unforeseen reductions will undermine budget planning (the costs of aid volatility). In fact, as GBS should only be proffered to countries when expenditure management is ‘specified’ (e.g. in an MTEF). Such recipients are in principle the best equipped to respond to any aid reductions provided these are announced in advance (so that they can be factored in to planning).

‘ The complicated system that the Dutch government uses for [GBS] is not accessible for either other donors or the recipient countries’. This admission ensures that GBS as deployed by the Dutch can never be effective, and probably undermines any (co-ordination) effectiveness of Dutch aid – GBS is by its nature most effective when donors club together, in recognition that the recipient has appropriate public expenditure monitoring mechanisms in place.

The IOB states that ‘sectoral support, like general budget support, does not focus on the root causes of the problems of low-income countries’. Statements such as this (and many similar throughout the IOB summary) reveal limitations in the underlying conception of how aid works more than any fundamental weaknesses in the poor aid-recipient countries. The impression given, however deserved, is that all the donor needs to do is decide to give GBS (which should be based on a full assessment of relevance for each country) and then development impacts will follow. Nothing could be further from the truth – GBS is nothing more than an effective means of providing aid support to government spending. There is no assurance that the spending is itself effective in delivering outcomes (this is actually the remaining concern for development – how to ensure the effect of spending).

The commentary on education illustrates this criticism, e.g. the view that the focus ‘ on primary education … has been at the expense of improved access to and the quality of adult education…’ misses the point. Ensuring adequate provision of primary education is a necessary condition for any educational attainment. Donors should actually ensure that the level of support for other levels of education is not reduced in absolute terms during efforts to expand support for primary education. In fact, successful expansion of primary education provision will generate a demand for increased provision of secondary education.

The statements on rural development and agriculture, notably that ‘ no explicit policy change was formulated’, imply that no thought was given to redressing the imbalance against support for the rural sector. On this point all donors are culpable. While agriculture remains the major economic sector in low-income countries it will continue to require investment and aid support.


Unfortunately, due to the age of this contribution and several migrations to online content management systems, the footnotes in the text may have been lost. The footnotes below are listed in its original order of appearance in text.
    1. Quotes from the IOB evaluation summary are shown in italics.