Research Brief
The Supply Side of Aid

In the wake of the financial crisis of 2008 and 2009 interest in the supply side of the donor equation has been rekindled. During a period where the economies of many major donors are unstable, and in some cases, shrinking, it is important to understand the effect this will have on their supply of foreign aid. In the recent WIDER Working Paper ‘Aid Supplies Over Time: Accounting for Heterogeneity, Trends and Dynamics’ Sam Jones looks at how domestic economic factors have influenced foreign aid supplies in the past and addresses some of the main gaps in the literature.

Long-term and Short-term Trends

Jones points out that there are both long and short-term determinants of aid supply. In the long term, the level of a country’s aid supply is likely to be largely dependent on domestic factors which are slow to change. Studies in this area have shown that the relevant long-term determinants include a country’s historical relations and cultural preferences, as well and economic factors such as domestic income and the government’s fiscal position.

The important role long-term factors play in determining aid supply is illustrated by the targets many donors commit themselves to. The current target for donors to set aid levels at 0.7 per cent of GDP dates back to a 1969 recommendation by the World Bank’s Pearson Commission, which set 1980 as the year by which this should be achieved. This figure has since been set as a target by the General Assembly and at the UN’s 2002 International Conference on Financing for Development. Recently at the 2005 Gleneagles G8 and Millennium +5 summits donors have made public pledges to meet this target. Despite these commitments, donors tend to treat aid targets as something to be gradually moved towards rather than something that needs to be achieved immediately. Also, the fact that aid has stayed at similar levels as in the past shows that long-term factor’s play a clear role in determining aid supply.

The literature on aid supply that looks at short term variations has largely focused on responses to macroeconomic shocks. This area of research has seen renewed interest since the onset of the global financial crisis, with studies attempting to predict what effect it will have on aid disbursement. However Jones points out that there are a number of other factors which may influence aid supply in the short term, and that idiosyncratic local factors may well play a large role. For instance, a 2010 study by Fleck and Kilby found that America’s War on Terror has driven a large increase in foreign aid disbursements. Similarly, while the end of the Cold War led to an overall reduction in aid supply, this effect varied significantly depending on the political implications for individual countries. Jones also points out that the Paris Declaration’s drive towards more efficient aid may mean that, to a certain extent, supply will increasingly be determined by a recipient’s ability to spend aid effectively rather than by the domestic situation of donors.

Variations between Countries and over Time

It seems clear, given the multiple potential determinants of aid supply, that there will be considerable variation in aid behaviour both between countries and within countries over time. This variation is not well-reflected in most of the literature on the subject which tends to use fixed effects models which control for, and thus hide, heterogeneity. For example some countries, such as Norway, have tended to consistently increase their aid budget over time, whereas as others, such as Finland and Italy, switch between periods of cutting and increasing budgets. Similarly in the wake of the financial crisis, many smaller donors cut their aid budgets, while larger donors actually tended towards increasing them, meaning that overall bilateral aid budgets actually showed an increase in 2009.

Jones goes on to point out that variations in aid supply over time also need to be treated carefully. Aid budgets any given year are largely dependent on decisions made in the past, in other words there is likely to be a lag between an event that is the cause of a change in aid budget, and the point at which the aid budget actually changes. Thus, it is important to account for a time lag between cause and effect when assessing aid supply, Jones suggests that previous studies have largely failed to do this.

Model and Results

Jones builds on this existing literature with the aim of designing a conceptual framework that accounts for both long and short-term trends in aid supply and uses empirical methods which allow for variations both between countries and over time. His model is based on three assumptions. First, over the long term donors seek to meet a particular target level. Second, domestic or global events may cause donors to deviate from that target. Third, after a deviation occurs donors will adjust towards meeting the target. The model allows him to achieve the following results:

1. Bilateral aid supplies broadly follow long-run trends determined by fixed and slow-moving factors.
2. Variation between countries over time is substantial. This suggests when possible estimates of future aid supply should be made on a country-specific basis.
3. Supply behaviours continue to evolve over time.
4. Previous studies have been too confident in their ability to predict aid supplies. The forecast drop in aid supply due to the financial crisis has not occurred either on average or on aggregate.

Jones points out that his findings are fairly cautious in nature and that the inherent heterogeneity in the historical data casts doubt on our ability to use the past to predict the future. However, despite this some positives can be taken from these results. Firstly, this kind of analysis allows us to make comparisons between donors and thus argue against poor or erratic donor behaviour. Secondly, the difficulty in predicting how domestic events will affect aid flows may indicate that aid is becoming less responsive to such short-term factors; this has the potential of making aid flows more predictable at the aggregate level. Jones finishes by suggesting that as more data becomes available, more elaborate models need to be designed; in particular future research should aim to explain the behaviour of both donors and recipients.