A quarterly magazine of the IMF

September 2005, Volume 42, Number 3

An analysis of five African countries that received big increases in aid

The international community is currently seeking to scale up official development assistance, provide further debt relief, and explore several innovative mechanisms for development financing to help low-income countries achieve the Millennium Development Goals (MDGs). But a surge in aid inflows—possibly to the tune of several percentage points of a recipient’s GDP—presents considerable macroeconomic challenges for the recipient country. How should countries adapt their monetary and fiscal policies? Will inflation result? Will the large inflows boost the exchange rate and make exports less competitive, and should this be resisted?

We examined the record of five African countries—Ethiopia, Ghana, Mozambique, Tanzania, and Uganda—that have recently experienced a surge in aid inflows, often from already high levels (see Chart 1). These countries benefited from the general rise in aid over the past decade and in particular from the Heavily Indebted Poor Countries Initiative, which reduced debt servicing costs and thus increased net aid flows. All five countries enjoy relatively strong institutions, so that policymaking is not dominated by macroeconomic disarray, misgovernance, or postconflict reconstruction. Therefore, their experiences offer useful lessons in scaling up aid to well-performing poor countries.

A framework for policy analysis

To examine the macroeconomic management of aid inflows, we found it useful to highlight the interaction of fiscal policy with monetary and exchange rate policy. Too often, fiscal policy concentrates on directing aid to worthy projects, while concerns about competitiveness and inflation drive monetary and exchange rate policies that may negate the desired impact of the aid.

We concentrated on the typical case in which aid dollars go to the government, which immediately sells them to the central bank and then uses local currency to increase spending on domestic goods. Of course, aid could come in kind, or equally, the government could spend the aid directly on imports, but these cases are analytically less interesting and pose fewer policy challenges—with a limited impact on macroeconomic variables. Our framework is underpinned by two distinct but related concepts: absorption and spending.

• Absorption is defined as the widening of the current account deficit (excluding aid) due to incremental aid. It measures the extent to which aid engenders a real transfer of resources through higher imports, or through a reduction in the domestic resources devoted to producing exports. Absorption depends on both exchange rate policy and on policies that influence the demand for imports. The central bank controls the exchange rate through its sales of foreign exchange, while monetary policy can be used to control aggregate demand—and the demand for imports.
  
  
• Spending is defined as the widening of the fiscal deficit (excluding aid) due to incremental aid.

The combination of absorption and spending defines the macroeconomic response to a scaling up of aid.

How the five performed

Broadly speaking, there are four possible responses to an aid surge in the short- to medium-term, and we looked at how the five sample countries performed according to these categories.

Absorb and spend. This first choice is the textbook response to aid. The government spends the aid on domestic goods, widening the fiscal deficit. The central bank sells foreign exchange, removing from circulation the local currency spent by the government. This foreign exchange pays for a widening of the current account deficit, as the aid is absorbed by the economy. There is a reallocation of productive resources from the traded goods sector to government spending. A key point here is that some real exchange rate appreciation may be necessary, and indeed appropriate, to affect this reallocation of resources. This is because if aid inflows are used to increase net imports, some combination of rising aggregate demand and real exchange rate appreciation is necessary to induce the increase.

Absorbing and spending is, in general, the first-best response to aid. Absorption ensures that there is a real transfer of resources to the recipient country, while government spending draws resources away from the traded goods sector. Other responses may be justified under particular circumstances for a short time, but in the long run the only sensible alternative to absorbing and spending is to forgo aid altogether.

Despite its theoretical attractiveness, an absorb-and-spend strategy was found to be surprisingly rare in our sample countries. No country fully absorbed and spent the incremental aid it received during its aid-surge episode (see table). In four of the five countries, less than one-third of the aid increment was absorbed.

Neither absorb nor spend. In this case the government does not spend the aid and keeps the new foreign exchange in the central bank. Governments may choose this option because they want to build up international reserves from a low level or smooth volatile aid flows. In two of the sample countries—Ethiopia and Ghana—absorption and spending were both very low (see Chart 2). Both countries entered the aid-surge period with a low level of reserves—at 2.2 months of imports in Ethiopia and 1.3 months of imports in Ghana—and used the increments in aid to increase import coverage.

In Ethiopia’s case, some reserve accumulation was also necessitated by the desire to protect the birr’s peg to the dollar. Prior to the aid surge, during the country’s border conflict with Eritrea, reserves had been declining precipitously as the central bank defended the currency in the face of a large fiscal deficit. The aid surge made it possible to arrest and even reverse the decline in international reserves, while maintaining the peg.

In Ghana’s case, the accumulation of reserves over the aid-surge period is partly attributable to the desire to smooth extremely high aid volatility. Aid surged in 2001 and collapsed in 2002. When aid surged again in 2003, the authorities were keen to create a buffer against the painful fiscal adjustments necessitated by the previous year’s aid collapse, and hence channeled much of the aid increment into reserves.

The reserve buildup meant that aid was effectively not available to finance increased domestic spending. However, in neither country was the aid surge accompanied by a significant widening of the fiscal deficit. Hence, the aid was not spent and there was no impact on domestic liquidity. In Ethiopia, inflation was low when the aid surge began, and remained within a 2–5 percent range throughout the period. In Ghana, the growth rate of reserve money declined significantly over its two distinct aid-surge periods. Inflation stood at over 40 percent a year when the first aid surge began, but had fallen to about 15 percent by end-2003.

Absorb but do not spend. This response substitutes aid for domestic financing of the government deficit. The government keeps a lid on expenditures, and the money supply shrinks as the central bank draws liquidity out of the domestic economy through foreign exchange sales. Where the initial level of domestically financed deficit spending is too high, this can help stabilize the economy. Although none of the countries in our sample exhibited this response over the aid-surge period as a whole, some countries used this strategy for particular years. For example, absorption exceeded spending in Ethiopia in 2001. This held domestic liquidity in check and probably averted a take off in inflation.

This approach to aid can also be used to reduce the level of domestic public debt and crowd in the private sector. The central bank uses the proceeds from its sales of foreign exchange to buy back government debt. With less public debt, more savings can be channeled into private investment. Such behavior was not common in our sample; although in Ghana, which had relatively high debt and domestic financing of the budget before the aid surge, the program targets were broadly consistent with such a response.

Spend but do not absorb. To spend but not to absorb aid is both problematic and common, often reflecting inadequate coordination of monetary and fiscal policy. The government increases expenditures, but keeps aid dollars locked up in the central bank as reserves. This response is similar to a fiscal stimulus in the absence of aid because the increase in government spending must be financed by government borrowing from the domestic private sector or by printing money. There is no real resource transfer in this case because net imports do not increase. Given a situation in which the fiscal authorities spend the aid and the central bank is unwilling to sell the foreign exchange and thereby allow aid absorption, the question arises as to how to adapt exchange rate and monetary policies. In particular, how should the increase in the money supply associated with aid-related expenditures on domestic goods be handled?

One strategy is to allow the money supply to increase. This approach risks high inflation and a depreciating nominal exchange rate. Although inferior to an absorb-and-spend response, this has the advantage that it leaves open the possibility of absorption in the future, but only if foreign exchange reserves are eventually sold to defend the nominal exchange rate. A second strategy is to curtail inflation and real appreciation through treasury bill sterilization. The sale of treasury bills reduces the money supply and raises interest rates. There is no real resource transfer from donors because the aid is not absorbed. All that happens is that resources are reallocated domestically from the private to the public sector via an increase in interest rates.

Our remaining sample countries—Uganda, Mozambique, and Tanzania—followed a combination of these last two strategies. In all three, concerns about the negative impact of a real appreciation on competitiveness dictated the pattern of aid absorption and the monetary response. Net foreign exchange sales by the central bank were contained to a level consistent with a depreciating nominal exchange rate. This held absorption in check. Because the respective governments simultaneously increased domestic expenditures, there was an injection of liquidity that led to inflationary pressures and various episodes of attempted sterilization through treasury bill sales.

In Uganda and Tanzania, the authorities were largely successful in keeping inflation under control, with underlying inflation never exceeding 10 percent during the aid-surge period. However, this was achieved at the expense of squeezing private sector investment through the sale of government paper during some periods. Episodes of treasury bill sterilization were typically accompanied by substantial increases in interest rates.

In Tanzania, the focus on sterilizing through treasury bill sales was subsequently abandoned in favor of allowing the money supply to increase. For a year or two, little inflation, real appreciation, or absorption resulted. Eventually, as inflation increased and excess liquidity built up in the banking system, the authorities turned to selling foreign exchange to sterilize the monetary injection associated with aid-related spending. This resulted in a sharp rise in absorption and some real exchange rate appreciation. In effect, the authorities delayed the movement toward a spend-and-absorb strategy.

In Mozambique, domestic expenditures rose much more rapidly than in the other two countries, partly in response to the devastating floods of early 2000. Government expenditures in the aid-surge period were 6.7 percentage points of GDP higher than in the pre-surge period. Accommodating the fiscal expansion required considerable monetary loosening. Reserve money growth shot up from about 7 percent a year in the pre-aid surge period to 53 percent in 2001. Inflation followed suit, peaking at well over 20 percent in early 2002. The high level of inflation contributed to nominal depreciation and currency substitution throughout the period.

In all three countries, unlike in Ethiopia and Ghana, the level of import coverage afforded by gross reserves was quite high, with reserves accumulating steadily. A more suitable response to increased aid inflows might therefore have combined a widening fiscal deficit with concurrent sterilization through foreign exchange sales, as encouraged by the countries’ IMF-supported programs. In Uganda and Tanzania, such a strategy would have relaxed the need for open market operations and the consequent steep rise in the treasury bill rate and interest payments. In Mozambique, it would have reduced inflation through two channels: through the resulting decline in the growth of base money, and through arresting the sharp nominal depreciation.

Was Dutch disease a factor?

A permanent increase in the level of aid to a country should lead to some degree of equilibrium real appreciation of the exchange rate. It is this appreciation that draws domestic resources from the production of traded goods to the production of schools, clinics, roads, and whatever other investment and social spending is deemed crucial to development (a phenomenon known as Dutch disease). Meanwhile, the aid dollars finance the rise in net imports that comprise the counterpart to the reallocation of resources. In essence, this is the spend-and-absorb case. The real appreciation cannot, therefore, be viewed simply as an unfortunate by-product of aid; it is integral to the process by which aid is supposed to work.

However, aid recipients often resist absorbing aid precisely because they are worried about Dutch disease-type effects on their international competitiveness. None of the countries in our sample fully absorbed incremental aid. In at least three of the countries—Uganda, Mozambique, and Tanzania—concerns about competitiveness loomed large in their decision to restrict the sale of foreign exchange and thereby limit absorption. Consequently, it is unsurprising that none of the countries experienced significant real appreciation, despite the surge in aid inflows (see Chart 3).

Lessons for others

Aid raises the stakes in the development challenge, promising great benefits if well used, but also risking an exchange rate appreciation that could impair the prospects for export-led growth. Forgoing aid on the basis of Dutch disease–type effects requires a compelling case that the latter outweighs the former, and this is a determination that is best made on a country-by-country basis. In any event, aid recipients can seek to minimize Dutch disease by improving the quality of aid usage. For example, aid allocated to productivity-increasing projects would tend to dampen or even reverse the impact of an appreciation of the real exchange rate. Another remedy is to use aid to directly purchase tradable goods, thereby reducing the impact on the exchange rate. In many cases these purchases may yield substantial benefits, as with antiretroviral drugs or machinery for building infrastructure.

In the long run, the only two sensible responses to a permanent increase in aid are an absorb-and-spend strategy or a don’t-absorb-and-don’t-spend strategy, with the latter being equivalent to refusing aid. To absorb and not spend can be very helpful in stabilizing the economy, as it permits a reduction in domestic debt and/or inflation without the fiscal contraction that this would normally require. It cannot be a feasible long-run strategy, however, as domestic debt stocks cannot fall indefinitely, and also because donors typically would not accept that aid proceeds be saved by the recipient government. To spend and not absorb is also not a desirable long-run strategy because it implies permanently rising domestic debt stocks or a permanent, large jump in the inflation tax.

From the start, aid recipients must be careful to coordinate fiscal policy with monetary and exchange rate policies. Budgetary authorities and the central bank need to agree on the short-term strategy for aid flows. A country may choose to absorb without spending, in order to stabilize the economy or to reduce domestic debt. It may choose to build reserves temporarily, perhaps as a buffer against aid volatility. Critical to either of these choices is the need for a prudent fiscal policy to complement the absorption strategy. But the response that epitomizes a lack of policy coordination is to spend without absorbing the aid. This combines a zero real resource transfer with the risk of crowding out the private sector. A persistent response along these lines should raise concerns for the recipient country and international partners about the strategy for utilizing aid.

The strategic question of how to best absorb and spend aid in the long run is one about which the IMF can provide only supportive guidance. However, the long run famously never arrives. In the meantime, the authorities in aid-recipient countries must balance a complex set of objectives involving fiscal policy and exchange rate and reserve management. But one message is simple: a given aid dollar can be used to build reserves or to increase the fiscal deficit—but not both. The cases reviewed in this article suggest that trying to do both is more common than it should be.