CHAPTER 4. Dealing with Aid Uncertainty and Volatility

Kevin Fletcher, Sanjeev Gupta, Duncan Last, Gerd Schwartz, Shamsuddin Tareq, Richard Allen, and Isabell Adenauer
Published Date:
April 2008
  • ShareShare
Show Summary Details

Aid volatility complicates fiscal policy implementation. Aid flows are more volatile than revenues (Bulíř and Hamann, 2007) and significantly more volatile than remittances (Gupta, Pattillo, and Wagh, 2007). Moreover, such volatility has increased over time. The relative volatility of aid, even for HIPCs, with respect to revenue (when variables are expressed as a share of GDP) has increased to 62 in 2000–03 compared to 25 in 1995–98 (Bulíř and Hamann, 2007) (see Appendix 1). This problem is likely to worsen, for two reasons. First, even if the volatility of aid does not change, a larger aid volume implies that a larger portion of the budgetary spending would be aid financed and thus subject to volatility. Second, the volatility of aid itself may increase because of shifts in the composition of aid away from project aid and toward budget support and program loans. A rising share of budget support can aggravate aid volatility because of the inability of donors to make long-term commitments for budget support.

Aid recipient countries can take several steps to mitigate the effects of aid volatility and uncertainty:

  • Conduct stress tests on baseline projections to assess the impact of aid volatility. Subjecting MTFs (and DSAs) to periodic stress tests can help identify short-term financing risks. The impact of aid shortfalls on the budget should be assessed regularly through such tests.

  • Build up reserve buffers to sustain spending if shortfalls materialize. Countries can also self-insure against aid volatility by accumulating reserves that can be drawn down in the event of a temporary aid shortfall. The size of the buffer should be determined on a case-by-case basis but could vary from 50 to 100 percent of annual aid-financed spending (Eifert and Gelb, 2005). Such a buffer would supplement other reserves that countries might accumulate to provide cover for imports or short-term debt, and enable countries to smooth expenditures without recourse to costly bridge financing from their domestic banking systems in the event of an aid shortfall.18 However, building up reserve buffers also requires that countries have in place appropriate strategies to invest and manage the reserves efficiently during aid windfalls.

  • Identify priority spending to be safeguarded from cuts in the event of an aid shortfall. This would ensure that critical programs are not starved of funds in the face of aid volatility. This policy is easier to adopt when such spending is defined in the Poverty Reduction Strategy Papers (PRSPs) or similar country documents, and when the PFM systems are capable of monitoring budgetary allocations to specific spending programs and their outturns.

  • Build elements of flexibility into spending programs. Designing spending programs so that they can be scaled up or down in response to fluctuating aid disbursements can mitigate the adverse effects of aid volatility. For example, wage spending could be made more flexible by using temporary and flexible employment contracts; contracting out services could be another option.

  • Include appropriate adjustors in IMF-supported programs to fully or partially accommodate aid volatility. Recent studies have shown that low-income countries respond to aid volatility by adjusting domestic financing and expenditure, but that this response appears to be asymmetric—in particular, aid shortfalls lead to cuts in domestic investment spending while governments do not increase such spending in response to aid windfalls.19 Accordingly, adjustors in IMF-supported programs should be designed to avoid a cutback of critical spending, such as domestic investment, when aid falls short of program projections. The degree to which shortfalls should be financed and windfalls saved depends on country-specific considerations, such as macroeconomic stability, absorptive capacity, and debt sustainability.

Long-term donor commitments can reduce aid volatility. The international community has pledged to provide more predictable and multiyear commitments on aid flows (Paris High Level Forum, 2005). Implementing the agreed-upon steps would help reduce the uncertainty and volatility of aid. Some bilateral donors are moving toward longer-term aid commitments. For example, the United Kingdom has provided a 10-year commitment on development assistance to Rwanda and Ethiopia. Similarly, the United Kingdom Department for International Development (DfID) has agreed to provide six-year program support for the health sector in Malawi under the Sector-Wide Approach. In addition, initiatives such as the International Financing Facility for Immunization, which is designed to provide “front-loaded, reliable funding over a number of years,” can reduce aid volatility.

Certain types of arrangements facilitate long-term donor commitments. Simple, formalized partnership agreements between aid recipient countries and donors help to deliver multiyear aid commitments in a systematic fashion. Recent examples of such successful collaboration are donor groups in Burkina Faso, Ghana, Mozambique, and Tanzania. Typically, such donor groups serve to jointly monitor both aid commitments and disbursements regularly, within a predefined performance framework. This has improved aid predictability and fiscal programming in these countries.

Countries should strive to ultimately avoid long-term reliance on aid. Besides strengthening the domestic revenue base, other elements of this strategy include strengthening countries’ debt management capacity and fiscal institutions, including PFM systems (see Chapter 5). This would ensure that resources are used both efficiently and effectively.

Bulíř and Hamann (2006) show that negative aid shocks coincide with negative income shocks, and propose building a reserve cushion to insure against aid shortfalls.

Celasun and Walliser (2006). IMF (2007b) discusses these findings in more detail. Our analysis, however, indicates that health and education spending is relatively unaffected by aid volatility.

    Other Resources Citing This Publication