The impact of foreign aid on the fiscal behaviour of the Ugandan government

UNU-WIDER / Dec 2013

Key findings

Foreign aid donors hope that their efforts will contribute to the development of the target country and provide the necessary resources to fill gaps in the sectors seen as relevant. However, because foreign aid will often be distributed through the state, understanding its effects on the fiscal behaviour of governments is a necessary condition for its effective and successful deployment. The new incentives and conditions created by the addition of aid to the action of the state will disrupt how it disposes of the fiscal tools of tax revenues, expenditure and public debt in uncertain ways. Although there is an expectation that aid will increase spending, raise taxation and decrease borrowing, this is clearly too general a statement to always hold true. The existing body of research shows that country-based evidence is the only way to reliably explore these dynamics as experiences between countries vary due to their different institutional foundations. On average, between 1990 and 2006, Uganda received foreign aid worth 11 per cent of its GDP. This long-term interaction of aid with government spending, tax revenues and public borrowing, appears to have led to a situation where development assistance has become incorporated into the country’s fiscal calculations on par with those other variables. Indeed, given that governments may count on foreign aid when planning their policies it can be argued that the effectiveness of foreign aid is best ensured by transparent, reliable and predictable aid contributions. This seems certainly to be true in the case of Uganda’s long experience with development assistance.

© Dana Smillie / World Bank

Different African experiences

The dynamic of aid and fiscal policy has been explored in several other African countries. In Ghana, aid has been found to increase the tax base, facilitating higher revenues and thus allowing the government to increase spending without needing to borrow more. On the other hand, in Zambia international aid has been reported to discourage efforts at tax collection and to increase government borrowing. This has not been the case in past studies of Malawi and Uganda where aid was found to raise government expenditure and revenues with negligible effects on borrowing. A similarly negligible effect of aid on tax revenues was found for Kenya, where aid does seem to have increased public spending. Despite some methodological concerns, in a study of aid to Ethiopia it was found that aid was not fungible, increasing development expenditure specifically intended by donors but not general recurrent government spending. Furthermore, aid was not found to have an effect on tax efforts, although borrowing increased due to unexpected decreases in revenue.

Revisiting Uganda’s experience with aid

Over the last twenty years Uganda has had extensive experience with international aid becoming an important contributor to GDP, mitigating the negative effects of low tax revenues. The ratio of aid-to-GDP increased from 1 per cent in 1980 to an average of 11 per cent between 1990-2006, peaking at 19 per cent in 1992. Revisiting the country’s experience permits an expansion on previous studies. The coverage is allowed to extend beyond the previously explored 1972-79 period and cover the full length of time over 1972-2008. This is insightful as the 1970s were fraught with political and economic instability and relatively low levels of aid, which may bias results. Methodologically, it is also possible to make use of techniques that help disentangle short- and long-term dynamics. The long-term relationships between aid, government revenues, expenditure and borrowing can be estimated separately from the short-term effect that shocks to any of these variables may have on the others.

Aid and long-term fiscal relationships in Uganda

In the long term, aid can be shown to have a stable relationship with tax revenues, government expenditure and borrowing. This appears to be unaffected by the changes that took place at the end of the turbulence that plagued Uganda during the 1970s. Over the last thirty years, the Ugandan government seems to have incorporated aid into its fiscal planning; ostensibly planning to fund its expenditure via tax revenue, aid and domestic borrowing, in that order. This internalization of aid by the Ugandan government rejects the opposite case, known as ‘aid exogeneity’. 

President of Uganda, Yoweri Museveni welcomes World Bank Group President Jim Yong Kim and United Nations Secretary-General, Ban Ki-moon in Kampala, Uganda. May 24, 2013.  Photo © Dominic Chavez/World Bank

Other findings suggest that although Uganda’s government has a budget constraint, of which aid is a component, it does not use aid to balance its budget and it is dependent on further borrowing to meet its obligations. It is often argued that aid flows can lead to ‘revenue displacement’, the phenomenon whereby aid deters from efforts at tax collection. The Ugandan case does not seem to lend credence to this. On the contrary, aid appears to have increased efforts at tax collection. Often, this is due to the conditions attached by aid donors that require efforts to be conducted to ensure the sustainability of tax revenues. Although aid is estimated to stimulate government spending, the movements are not equivalent or one-to-one as the theory of ‘aid additionality’ would suggest. Furthermore, aid appears to decrease costly longer-term borrowing. This is consistent with the view that domestic borrowing is often triggered by aid’s unpredictability and volatility. However, the evidence does not support a one-to-one ‘perfect’ substitution between the two either.

Short-term shocks and the drivers of Uganda’s fiscal system

Shocks to aid, spending and government borrowing seem to have permanent effects on the long-term fiscal system of Uganda, causing disruptions with each other that persist over subsequent periods. However, shocks to tax revenue are transitive and stable without permanent effects. This is consistent with the fact that Uganda does not follow a balanced budget path. Instead of aid, tax revenues and government spending adjusting to each other’s movements to bring about a balanced budget, budget spending plans in Uganda adjust to tax revenues.

Lessons from Uganda: aid should be transparent, reliable and predictable

ImplicationsUltimately, the lesson from Uganda’s experience is that in case of long-term dependence on foreign aid development assistance, aid-recipient governments may begin to incorporate this factor into their fiscal calculations on par with spending, revenues and borrowing. This adds another aspect of uncertainty to fiscal planning in a developmental environment where an unbalanced debt path and low tax revenues may already limit the agency of the state. This should emphasize the need for donors to ensure  transparency, reliability, and predictability of aid disbursements.

This Research Brief is based on WIDER Working Paper 2013/101 ‘A timeseries analysis of the impact of foreign aid on central government’s fiscal budget in Uganda’ by Thomas Bwire, Oliver Morrissey, and Tim Lloy