Foreign aid is facing difficult times. Even some aid practitioners have called its effectiveness into question. While aid has had success in humanitarian relief, family planning, and reducing infant mortality, its record in promoting economic growth has been mixed. Economic growth is not the sole objective of U.S. foreign aid, and it may be the least important goal for policymakers concerned with security, short-term solvency, human rights, or democracy. But the effects of aid on growth can be measured empirically, and growth is a necessary condition for meeting most of the broad objectives of aid.

While aid has succeeded in promoting growth in some countries, in many others it has failed or even been counterproductive. A number of countries, many in sub-Saharan Africa, are poorer than when they began receiving aid several decades ago. Donors have often subsidized unsound economic policies. In such situations, foreign aid has perpetuated poor policies and weak economic performance. The solution is not to end or even reduce aid flows, but for donors to allocate resources more selectively.

In the 1960s and 1970s, aid was driven primarily by the security concerns of the Cold War, with an underlying focus on reducing poverty. During the debt crisis of the 1980s, the rationale for aid shifted to limiting the liability that developing country debts increasingly placed on the international economic system. Concerns about reducing poverty also resurfaced, but a new emphasis on promoting democracy and human rights complicated those sentiments. In the 1990s these rationales are joined by new priorities such as supporting export markets in developing countries. It is not clear that aid can achieve all these goals simultaneously; at times they may work at cross-purposes.

It is clear, however, that the manner in which the United States conceptualizes and evaluates aid has changed dramatically. Decades ago donors saw aid as a transfer of resources from rich to poor countries. Today they see it more as a means to improve the use of domestic resources in recipient countries. The focus of aid has changed from development-related projects in health, education, and agriculture to encouraging the adoption of growth-oriented fiscal, trade, and monetary policies.

While convincing governments to implement development projects, particularly when donors provide the bulk of the resources, is not difficult, convincing them to implement a new macroeconomic framework requiring painful fiscal adjustments is no small task. The politically unpopular costs of these policies are visible up-front. But the benefits of sustained growth tend to become visible only in the longer term. Donors' skepticism about recipients' fiscal discipline has led to a growing reliance on conditional aid, and the conditions placed on multilateral development bank loans increased markedly during the 1980s. But these conditions can be difficult to implement. While compliance with some conditions, such as the stabilization of inflation or the devaluation of the exchange rate, is easily measured, it is much more difficult to assess and enforce others, such as progress in reforming the tax system or civil service.

Several studies in the 1990s have found no positive relationship between conditioned financial aid and economic growth. Some assessments even find a negative relationship, particularly in low-income countries. Yet these studies acknowledge that the broader policy orientation that foreign aid seeks to promote -- open economic policies with prudent fiscal management -- is producing strong results worldwide. While some countries with little or no external aid have had strong growth performance due to market-oriented policies, others, despite increasing levels of conditional aid, continue unsound economic policies.


Average aid-to-GNP ratios suggest that countries that receive greater amounts of aid do not grow faster than those that receive less. For example, African nations have aid-to-GNP ratios more than ten times that of either Latin America or East Asia, but still suffer inferior economic performance. Within Africa, countries with poor economic policies have received more aid per capita than those with responsible policies. While some studies of World Bank lending find a negative correlation between aid flows and growth, other studies, using broader measures of aid flows, find that aid on average has no effect on growth.

There are multiple explanations for these findings. One is that aid flows remove a hard budget constraint, postponing the need for domestic consensus on economic adjustment. Such a consensus is important in developing public support for reform, as well as a commitment to prudent economic policies among policymakers, which is often referred to as "ownership." A recent World Bank review found that 44 percent of adjustment loans associated with a high degree of ownership had satisfactory economic performance, and 28 percent had unsatisfactory performance and a low degree of ownership.

World Bank economists Michael Bruno and William Easterly recently came to a similar conclusion, albeit in a different fashion. Among a sample of 55 developing countries, those where inflation reached crisis levels of more than 40 percent initially suffered a drastic drop in growth, but recovered quickly after prices stabilized. These countries achieved and maintained growth rates higher than before their inflationary crisis. In contrast, countries that muddled along at slightly lower inflation maintained lower growth rates.

The political side of this story is that the inflation crises gave governments the latitude to implement reforms -- trade liberalization, pension reform, the introduction of independent central banks -- that are essential to sustained growth. Lower-inflation countries did not feel domestic or external pressure to change their policies. In the high-inflation countries, aid diminished before the onset of the financial crisis. In the low-inflation countries, grants increased sharply over time. And of the 19 low-inflation countries, 11 were in the franc zone, where inflation has been kept artificially low by a unified exchange rate that is pegged to the French franc and subsidized by substantial amounts of bilateral aid. The franc zone countries have had slow growth and high poverty.

Equally important is how governments use aid. A recent study of 97 countries conducted by economist Peter Boone found that nearly all foreign aid goes to government consumption and has little impact on investment or growth. Aid's record is much better when appropriate policies are in place. One World Bank study found that aid increased growth in countries with open trade, low inflation, and a fiscal surplus. But growth performance rarely determines how aid is allocated. Donors' strategic interests are usually more decisive than any interest in rewarding responsible policies.

Suggesting that more countries be allowed to slip into crisis, however, is not an acceptable policy recommendation. Crises cause widespread suffering and are not always resolved peacefully. Moreover, aid or the prospect of aid can play an important role in the consolidation of a reform-minded government. In Poland in 1989, external support was a benefit that the reform team could deliver and that was vital to its political ascent. Still, too much aid flows to countries that are unable or unwilling to implement necessary reforms.


Aware of these dangers, lenders in the 1980s began to offer conditional loans. Conditionality implies a contract between a borrower and a lender, usually a multilateral institution like the World Bank, that provides credit at below-market rates. The "grant" element of these loans varies according to the country's risk. While a particular interest rate might not be considered a grant for a low-risk country like Korea, it would be concessional for a high-risk country like Zambia, which would pay much higher market rates. The International Development Association, the World Bank's soft-loan window, provides interest-free credit for the poorest countries.

Conditions for aid are usually applied ex ante, that is, borrowing countries must meet specific requirements to be eligible for a loan and then meet other conditions to receive subsequent loan payments. Weak borrower commitment compounded by poor donor confidence has led to a mushrooming of conditions. A high number of technical conditions can be difficult to comply with, particularly if there are unexpected political developments or external economic shocks. And the more numerous the conditions, the more likely that governments will avoid full compliance. In addition, when all policy measures are dictated from abroad, an important policy learning process is lost.

Donors have strong incentives to continue lending even when conditions have not been met. Ultimately, the World Bank must make loans to remain in operation. The average loan officer has more incentives to disburse approved loans than to enforce strict compliance with conditions. By the World Bank's own calculations, only 60 percent of agreed conditions are typically implemented during the first loan period, yet the second phase of a loan is almost always approved. The World Bank's wide exposure to a number of highly indebted countries in the 1980s gave it a stronger stake in preventing those countries' insolvency than in promoting policy reforms.

Many observers, and even some World Bank reports, have called for a reevaluation of conditional aid, moving toward more selectivity, fewer conditions, increased domestic involvement in aid programs, and stricter enforcement when conditions are not met. For conditions to be credible, they must be enforced across the board, which is far from the current practice.

A more fundamental shift in lending practices would change the criteria for conditional aid. Instead of applying detailed conditions at the beginning, donors and borrowers should agree on a general policy package. The borrower's implementation of reforms would determine the country's eligibility for future loans. This approach would avoid disagreements over specific measures, focus attention on the overall policy, and encourage "ownership" of policy reforms.

While a shift toward more selective lending based on countries' track records is intuitively appealing, it would entail substantial risk since it could result in the withdrawal of aid from many poor countries, particularly in Africa. Despite the potential social costs, a withdrawal of financial flows could generate a shift toward genuine policy reform in some countries. But there are likely to be instances where governments do not adopt reforms and economic performance deteriorates further. The social costs to these countries could be mitigated by continued flows of humanitarian aid and unconditional technical assistance to government or non-government actors.

Aid is most beneficial when it influences policy at the level of ideas. When reformist ideas are adopted by a critical mass within society, they tend to be implemented as policy. The experience of many East Asian countries in the 1970s and Latin American countries in the 1980s and 1990s supports this proposition. Donors who demonstrate a willingness to stop loans when poor policies are being pursued would put all borrowers on notice.

It is important to consider whether countries' economies perform poorly because of their initial conditions, such as per capita income and natural resources, or because they pursue inappropriate policies. According to the so-called convergence theory of economic growth, poor countries grow faster than wealthier ones due to a higher rate of return on capital investment. However, since the early 1970s some poor countries have experienced negative per capita growth. The disheartening performance of many poor countries, particularly in sub-Saharan Africa, has led some economists to give credence to theories of a "convergence club." To be a member, countries must have enough human capital to take advantage of modern technology. Initial conditions, this theory suggests, predetermine poor countries' performance -- a pessimistic scenario for both poor countries and aid.

In a 1995 Brookings Institution paper, Harvard economists Jeffrey Sachs and Andrew Warner dispute this theory, arguing that policies matter more than initial conditions. Countries that fail to secure private property rights, that follow autarkic trade policies, and that maintain inconvertible currencies are unlikely to see convergence regardless of their production technology or initial endowment of human capital. In a study of 117 countries, Sachs and Warner found that market-based policies brought an overwhelming tendency toward convergence, even among countries that started with extremely low levels of human capital and per capita income. It was difficult to identify a single case in which a poor country protected private property rights and maintained economic openness yet failed to achieve economic growth. Few countries that pursued closed policies grew at equivalent rates and in a sustainable manner.

These results suggest that responsible policies are the most important factors behind performance, and that the greatest contribution aid can make to a country's well-being is to support the adoption of such policies. In most poorly performing low-income countries, ineffective policies have been in place for decades. In such countries, where political or institutional conditions stymie reform, external support can encourage change. Initially this support should not be in the form of substantial financial flows, but rather that of technical assistance, ideas, and the promise of financial assistance to come.


The evidence on aid and growth calls for much greater selectivity in allocating bilateral and multilateral aid. It will be difficult to meet any other objective -- stability, democratic government, respect for human rights -- without a sustainable macroeconomic framework that makes inroads toward reducing poverty. Available resources must be used more effectively to assist countries willing to follow sound economic policies.

A drastic curtailment of overall aid resources is undesirable. In many cases the promise of future assistance helps forge a consensus on policy reform. It is important for donor countries to reward poor performers when they adopt appropriate policies. Those rewards should include not only financial flows but also debt forgiveness, open markets for trade, and private capital inflows. Most increases in foreign capital take place in fast-growing countries well into their reform process, not among those just launching reforms. Aid resources can encourage reforming governments to stick to their objectives and help cover transition costs along the way.

A strategy of aid selectivity will entail its own short-term costs for donors. Cutting off poor performers is likely to result in some defaults on debt payments, losses that for the most part will have to be absorbed by multilateral institutions. And while cutting off aid may result in the formation of a reform-minded consensus in some countries, in others it may exacerbate social tensions beyond the crisis point, perhaps necessitating humanitarian intervention.

At present, the United States' capacity to administer aid is threatened. Under the recent balanced-budget deal, real international affairs spending will probably be reduced from $19.5 billion to $16.5 billion by 2002. Given these projections, official development assistance (ODA), which is almost half the international budget, at $9 billion, would drop to less than $8 billion. This reduction would contribute to an ongoing decline in global ODA donations and reduce the United States' contribution to less than 15 percent of all global aid, guaranteeing a decline in U.S. influence in multilateral institutions. While America was once the world leader in foreign assistance, it is now falling to fourth place behind Japan, Germany, and France. The United States devotes a smaller percentage of GDP to development assistance than any other member of the Organization for Economic Cooperation and Development (OECD). It is unlikely to convince other countries to adopt a new approach to aid when its own contribution is falling to historic lows.

Rather than cutting ODA, the United States should increase it by $3 billion to $12 billion per year, which would hardly bring it to an all-time high. Because other types of international spending would fall by $1 billion under the president's budget proposal, a $3 billion increase would result in a $2 billion overall rise in the international affairs budget. The United States' ODA contribution would return to 0.15 percent of GDP, its level during the early 1990s. While the American contribution would remain below the OECD average of 0.35 percent of GDP, a lower than average aid-to-GDP ratio for the United States should be accepted, due to the U.S. military's unrivaled contribution to international stability. But maintaining global aid flows at adequate levels requires an increase in the American contribution. This funding would ensure that there are sufficient resources to help reforming countries as much as countries like Korea and Taiwan were helped in the past. It would also assist the United States in convincing other donors and multilateral institutions to adopt a more selective approach to aid -- a strategy that will not succeed if the United States acts alone.

The recent consolidation of the U.S. Agency for International Development with the State Department will remove a political stumbling block to adequate financing of foreign aid. But it does not resolve the central issue of aid effectiveness, and therefore the concerns of aid's most vociferous critics. Leading the effort to make foreign aid more effective in promoting growth and reducing poverty in developing countries would be important in itself, and could go a long way toward reducing skepticism about an important foreign policy tool.

You are reading a free article.

Subscribe to Foreign Affairs to get unlimited access.

  • Paywall-free reading of new articles and a century of archives
  • Unlock access to iOS/Android apps to save editions for offline reading
  • Six issues a year in print, online, and audio editions
Subscribe Now
  • Carol Graham is a Visiting Fellow and Michael O'Hanlon a Research Associate in the Foreign Policy Studies Program at the Brookings Institution. Graham is also a Special Adviser to the Executive Vice President of the Inter-American Development Bank. They are co-authors of the April 1997 Brookings study A Half Penny on the Federal Dollar: The Future of Development Aid.
  • More By Carol Graham
  • More By Michael O’Hanlon