Prospects for Reform

The good news is that a number of important developing countries have accomplished considerable reforms in the past two decades. In 1980, about 60 percent of the world's extreme poor—those living on less than one US dollar per day—lived in just two countries: China and India. At that time, neither country seemed a particularly likely candidate for reform. Both had rather poor property rights and government efficiency according to the measures used in cross-country studies, and both were extremely closed to the world market. Over the past two decades, however, China has introduced truly revolutionary reforms, restoring property rights over land, opening the economy to foreign trade and investment, and gradually making the legal and regulatory changes that have permitted the domestic private sector to become the main engine of growth. Reforms in India have not been quite as dramatic, but have still been very successful at reducing the government's heavy-handed management of the economy and dismantling the protectionist trade regime. Among low-income countries, there have been a number of other notable reformers as well; Uganda is a good example in Africa, and Vietnam in Southeast Asia.

The general point about all of these low-income reformers is that outside donors were not particularly important at the start of these reform efforts. These movements are home-grown and each has an interesting and distinct political-economy story behind it. Once these reforms began, however, foreign assistance played an important supporting role in each case. Institutional reform involves much social and political experimentation. The way that China has gradually strengthened private property rights is an excellent example, as is the way India reformed its energy sector. Foreign assistance can help governments and communities examine options, implement innovations, and evaluate them. To do this effectively, donor agencies need to have good technical staff, worldwide experience, and an open mind about what might work in different circumstances.

The World Bank is often criticized for giving the same advice everywhere, but this simply is not true. World Bank reports on different countries show that the World Bank typically makes quite different recommendations in different countries. The criticism that comes from government officials in the developing world is a different and more telling one: that the World Bank tends to make a single strong recommendation on each issue, instead of helping clients analyze the pros and cons of different options so that communities can make up their own minds about what to do. We do not know much about institutional change, so it is more useful to promote community learning than to push particular institutional models.

For example, the Education, Health, and Nutrition Program—known by its Spanish acronym, PROGRESA—is a successful program of cash transfers that encourages poor families to keep their children in school that was developed and evaluated in Mexico without any donor support. A number of donors now have helped communities in Central American countries to implement similar programs. In each case, communities need to tailor the program to their particular situation. Systematic re-evaluation is important because the same idea will not necessarily work everywhere. But this is a good example of how donors can promote learning across countries and support institutional change by presenting a variety of reform options for developing countries to follow.

Money Matters

While supporting country and community learning is probably the most useful role for aid, and the one that will have the largest impact, there is a still a role for large-scale financial aid. Studies have shown that there is little relationship between aid amounts and growth rates in developing countries, but there is a rather strong relationship between growth and the interaction of aid and economic policies. This finding, as well as microeconomic evidence about individual projects, suggests that the growth effect of aid is greater in countries with reasonably good institutions and policies. The success of the Marshall Plan is a classic historical example. More recently, states such as Uganda show that the combination of substantial reform and large-scale aid goes together with rapid growth and poverty reduction. In a poor institutional environment, however, large-scale aid seems to have little lasting economic impact and may even make things worse by sustaining a bad government.

What follows from this is that aid is going to have more impact on poverty reduction if it is targeted to countries that are poor and have favorable institutions and policies. This philosophy underlies a number of new initiatives in foreign aid—European countries, including the United Kingdom and the Netherlands, have reformed and expanded their aid program along these lines. The new US Millennium Challenge Account is based on these principles as well.

Using aid to support learning and being selective in the allocation of large-scale financial resources are linked. When donors tried to push large amounts of money into weak institutional environments, they naturally wanted to have large numbers of conditions dictating how institutions and policies would change. But this neither promoted effective learning nor led to good use of money. The new model argues for much less conditionality—encouraging countries and communities to figure out what works for them—but retaining some form of selectivity in the allocation of financial resources.

Keeping in mind the persistence of institutions and the difficulty of changing them, one should have modest hopes for what foreign aid can accomplish. But as long as there are countries and communities around the world struggling to change, the international community must support them. Afghanistan today is a good example. The country is trying to develop new institutions at the national and local levels, and the world has a big stake in helping it succeed. The international community does not know for sure what will work, but outsiders dictating a new set of institutions will almost certainly fail. On the other hand, donor agencies can help both national and local governments learn about options, implement policies, evaluate results, and re-design if necessary. As a solid institutional framework develops, there will be increasing scope for large-scale funding of roads, schools, and other social infrastructure. The effort may fail. No doubt the lack of good institutions in Afghanistan reflects extensive historical and political factors that will be hard to overcome. It is important to go in with eyes wide open; trying to reform aid based on what we know is preferable to giving up on aid and closing our eyes to the massive poverty that remains throughout the developing world.