Co-Sponsors: Center for Democracy and Free Markets of the Council on Foreign Relations; and the Institute for Global Democracy of the Henry L. Stimson Center

Speaker: Thomas Carothers, Vice President for Studies, Carnegie Endowment for International Peace

Commentators: David Dollar, Research Manager, Development Economics Research Group, World Bank; and Timothy Geithner, Senior Fellow, Council on Foreign Relations

Moderator: Morton Halperin, Senior Fellow, Council on Foreign Relations


Mr. Thomas Carothers spoke on the topic of whether donor organizations should condition development assistance on the existence of certain fundamental democratic institutions. (At the Institute for Global Democracy's inaugural luncheon on May 7, Dr. Larry Diamond had proposed that all official development assistance "be conditioned on the existence and effective operation of institutions to control corruption and to promote transparency and the rule of law," assuming that without these institutions economic development is not possible. For details of Dr. Diamond's proposal, see the summary of the May 7 discussion that is posted on the Institute's Web site.)

Mr. Carothers began his presentation by locating Dr. Diamond's proposal in the history of donor efforts to condition assistance. The first generation of assistance was characterized by the imposition of economic conditions for economic purposes (such as the macroeconomic requirements that the IMF still regularly employs in its stabilization programs). In contrast, the second generation entailed the use of political conditions for the purpose of promoting political reform (e.g., cutting off aid if an election is conducted fraudulently). According to Mr. Carothers, Dr. Diamond's proposal should thus be seen as a synthesis of these two generations of conditionality in that it seeks to impose political conditions in the service of economic development, arguing that basic democratic institutions such as free elections and an independent judiciary are part of the foundation of any society's sustained economic progress.

Mr. Carothers commended Dr. Diamond's proposal for its sweep, boldness and sharp challenge to existing policy. However, he said that he could not endorse the proposal for three reasons.

First, Mr. Carothers asserted that history does not demonstrate that good governance is a necessary precondition for economic development. In lieu of empirical evidence, he stated, there has arisen in recent years a "credo" that affirms but does not prove this causal link. While he acknowledged that there is some positive evidence for this relationship, he cited equally negative evidence whereby economic development took place despite a lack of transparency, accountability, free press or rule of law. In this latter regard, he made particular note of the phenomenal economic growth of Taiwan and South Korea under authoritarian regimes. Mr. Carothers asked how, for example, donors could preach to developing countries that the rule of law is a pre-requirement for attracting foreign investment when China (a country clearly without the rule of law) is the world's leading recipient of such investment. He was careful to point out that in questioning the credo, he was neither endorsing the so-called Asian model of development nor of course arguing that good governance and the rule of law are undesirable for developing countries. He was simply warning that the credo has mushroomed into an ideological proposition that is too simplistic and thus potentially harmful to developing countries.

Second, Mr. Carothers argued that democratic conditionality is a flawed idea because bilateral donors-especially the United States Government-would never apply such a policy in a fair and universal manner. They would not do so, he explained, because strategic, economic and other interests compete with the pure development interest in determining the level of foreign aid to a country. For example, he could not imagine that the United States would ever cut off aid to countries such as Egypt and Armenia because they have failed to implement political reforms. Thus the United States would apply conditionality only to weak states in which it had very few interests. And such an inconsistent policy, he concluded, would not impress the international community.

Third, Mr. Carothers asserted that even if it were applied universally, democratic conditionality would not be effective in promoting political reform because it does not suit the complexity of the problem. Citing the literature on economic conditionality (including the work of David Dollar and Paul Collier), he said that the consensus among these scholars is that economic conditionality has been ineffective in producing economic reform. In the view of Dr. Collier, the policy of economic conditionality is based on a misunderstanding of how economic policy is reformed on a sustainable basis. In turn, Mr. Carothers believes that the failed record of economic conditionality provides four cautionary lessons for the application of political conditionality:

· Too blunt a tool for donors. Like economic reform, the development of democratic governance is a long-term process composed of many complex parts. For instance, fostering the rule of law alone involves the reform of judges, prosecutors, police, bar associations, legal education, criminal procedure laws, etc. Progress in any of these areas will at best occur in fits and starts. In such a complex situation, how would donors accurately monitor progress or regress in order to apply the broad instrument of conditionality?

· Too small a carrot for recipients. If a government wants to reform, it will not do so simply because it wants to receive foreign assistance. Similarly, if a government does not want to reform, it will not change its mind simply because more assistance is offered. By their very nature, political reforms transform the relations of power within a society. Simply put, political leaders are not going to relinquish core elements of their authority for the sake of receiving external assistance.

· Too risky a choice for donors. In most poor countries, donors would choose not to cut off aid to promote political reform because to do so would threaten the country with economic disaster. For example, donors severed aid to Zambia in 1996-97 because the president excluded his main political opponent from running in the presidential election. However, in spite of the flawed re-election of the president, donors soon returned to Zambia lest the economy-which had actually been making progress in some key sectors-suffer.

· Too little ownership for recipients. Lasting political or economic reform must be based on a strong social consensus. But conditionality creates the opposite: indigenous resentment that reforms are being imposed by powerful outsiders. Take the recent example of U.S. actions in Yugoslavia. If the overriding goal of U.S. foreign policy is to encourage the Yugoslav people to themselves come to terms with their country's war record of the past decade, then forcing Slobodan Milosevic's extradition to the Hague ill served this purpose.

In conclusion, Mr. Carothers stated that in spite of his opposition to the concept of democratic conditionality, he of course believes that donors should promote democracy and good governance as a major goal of foreign aid. They should do so both because they hold democratic governance as a core political value and because democracy will contribute to a society's long-term economic development. However, he said, donors should use assistance only for "tactical" purposes. That is, they should cut off programs that are failing due to the absence of political will or they should not assist governments that commit egregious violations of human rights. In so doing, their motivation should be to avoid either the wasting of money or the legitimating of undeserved leaders. But they should not assume that such actions would in turn change the behavior of those governments. And he urged donor countries to promote democratic governance by means of multilateral diplomacy so as to institutionalize the norm on a universal basis. The complexity of the problem demands a multiplicity of diplomatic and programmatic approaches. In contrast, he warned, a policy of democratic conditionality would lead to disappointment on the part of donors and ill will on the part of recipients.


As the first commentator, Dr. Dollar introduced his remarks by stating his general agreement with Mr. Carothers' presentation. He proceeded to make three main points.

First, drawing on his research conducted at the World Bank, Dr. Dollar asserted that the effectiveness of foreign assistance in reducing poverty largely depends on the soundness of the economic policies of recipient countries. In his view, good economic policy includes strong macroeconomic management, a relatively open trade regime, and secure property rights. He expressed a slight disagreement with Mr. Carothers' first point: that there is insufficient empirical evidence demonstrating that good governance is a prerequisite of economic development. Dr. Dollar's view is that the specific aspects of good governance and the rule of law that protect property rights are absolutely essential to economic development. Otherwise, he said, individuals and small firms would not risk investing their money.

Dr. Dollar concurred with Mr. Carothers' view that the consensus of researchers is that donors have very little influence in initiating policy reform. However, he said, once economic reforms emerge, foreign aid can play a very useful supporting role, accelerating further reform and poverty reduction. In this context, Dr. Dollar stated that in the last 15 years there have been only four major successes in reducing poverty: China, Vietnam, India and Uganda. In each of these cases, economic policy was first reformed, and then donors supported the reforms with their assistance. (In spite of their authoritarian governments, he noted, China and Vietnam score above the median in terms of foreign-investor confidence about the security of their money.) Dr. Dollar said that the clear implication of this research for donor policy is that development assistance should be targeted at poor countries with good economic policies.

Second, Dr. Dollar acknowledged that in the long run democratic governance can contribute to economic development. In fact, he said, the causality runs in both directions: democracy can foster responsible economic policy through its mechanisms of accountability; and good economic policy can create the social basis for democracy through the increase and distribution of wealth. However, stressing that the long run can be very long indeed, he noted that in the short run democracy and good economic policy are not related. Citing political and economic indices, he argued that for the poorer half of countries in the world, there is no statistical correlation between democracy and good economic policy. Good policy can occur under either democracy or authoritarianism; ditto for bad policy. Having already identified Vietnam and China as authoritarian regimes with good economic policy, he now invoked Haiti, Ukraine and Nepal as prime examples of more democratic regimes with bad economic policy.

Therefore, Dr. Dollar argued that if donors are truly interested in reducing poverty, they should support countries like Vietnam, China and Uganda. And, he added, they should not have any expectation that assistance to the current governments in Haiti, Ukraine and Nepal will result in the reduction of poverty.

Third, speaking as a private individual, Dr. Dollar said that he supports the use of foreign assistance to promote democracy. There is no reason, he believes, that donors cannot have more than one objective. But the different objectives require different methods. In terms of promoting democracy, Dr. Dollar agreed with Mr. Carothers that the application of tough conditionality would not likely be successful in producing the desired outcomes. Instead, he proposed, one of the best ways to support democratic revolutions such as in Nigeria would be to counsel the new governments to undertake economic reform in tandem with political reform. For without economic reform, he reiterated, there will be no improvement in the material lives of the people; and without economic development, democracy will forever rest on an unstable foundation.

As the second commentator, Mr. Geithner expressed general agreement with Mr. Carothers and full agreement with Dr. Dollar. Having said that, he contributed the following points to the discussion of the topic.

Mr. Geithner began by asserting that the main goal of development assistance should not be the mere augmentation of a country's financial resources but rather the improvement of its economic policies. Echoing Dr. Dollar's comments, he said that without policy reform, additional resources would be wasted; but also that reformed policies require external financial support in order to be fully implemented. This continuing challenge to promote policy reform derives from the failure of donors to do so in the past.

According to Mr. Geithner, the donor community possesses (apart from the provision of technical assistance) two general tools with which to promote better economic policy: one negative and one positive. The negative tool is "selectivity." By this, he meant that donors often decide to select out of the pool of potential recipients those governments that are deemed to be excessively weak, corrupt or immoral. For instance, he said, it was perfectly reasonable for the international community not to assist Burma or North Korea except in the area of humanitarian relief. Such cases of exclusion, Mr. Geithner advised, are best handled multilaterally. He noted, however, that the ability of the United States to organize such multilateral coalitions is limited. Moreover, even if coalitions were successfully organized, the potential for selectivity to actually change the behavior of these governments is also limited.

The positive tool that Mr. Geithner identified is "conditionality"-that is, the provision of assistance only if certain policy reforms are first adopted. These reforms include what he called the "trinity" of good macroeconomic management-sound fiscal, monetary and exchange-rate policies-as well as adequate spending to address primary social needs such as education and health. Mr. Geithner said that there is a consensus within the development community concerning these preconditions of economic development.

In Mr. Geithner's view, those governmental institutions that support the private sector and the operation of the free market are equally important as preconditions of economic development. However, he noted that there is very little consensus within the donor community about these institutions. The lack of consensus is due partly to the fact that these institutions are "vague, amorphous and hard to influence from the outside" and partly to the fact that there is a large diversity among these types of institutions. Notwithstanding the absence of a consensus, he said it was legitimate for donors to insist on the existence of indigenous "governance" institutions that could prevent the theft of assistance funds, ensure budgetary transparency, secure property rights, and support a healthy financial sector (including an adequate regime for addressing insolvencies). The United States, he said, pressed such "positive conditionality" during the financial crisis in Indonesia, believing that such measures were relevant to Indonesia's long-term economic performance. But, he said, many donors viewed the U.S. effort as misguided and overly ambitious.

Mr. Geithner also said that some of the difficulty in achieving a donor consensus on these governance-related preconditions was due to the proliferation of such demands within the U.S. aid debate. As with other areas of U.S. foreign policy, he said, the erosion of the traditional center of the political spectrum has magnified the positions of the far left and far right. In his view, these extreme positions, although well-intentioned, take the issue of conditionality far afield from its central purpose of poverty reduction. He proposed that in the future the United States should restrain its tendency to use foreign assistance for messianic purposes. It should do so, he argued, because Americans themselves cannot agree on the content and method of conditionality in terms of political reform, and because the United States does not have enough moral authority or a large enough assistance budget to lead the donors toward a consensus.

In closing, Mr. Geithner stressed that apart from these problems, the United States still has a decisive role to play in moments of great opportunity, whether in regard to the use of negative selectivity to deny assistance to an outcast government or in regard to positive conditionality to support a reformist government. But to play this role successfully, he said, the United States needs humility, realism and a disciplined focus on the reduction of poverty.

-- Written by David Yang and Virgil Esguerra