A Farewell to Alms

By ARVIND SUBRAMANIAN
August 22, 2007 - WSJ

When celebrities such as Angelina Jolie or Bono highlight human tragedy to show that something can be done to alleviate it, the heart melts and the purse strings loosen. But the stars, alas, aren't up on the economic literature. Research is increasingly questioning the benefits of foreign aid.

For a long time, persistent underdevelopment in aid-receiving countries fostered doubts about whether the do-good impulse was doing permanent good. Validation seemed to arrive with two World Bank researchers, Craig Burnside and David Dollar, who purportedly showed in the late 1990s that aid helped boost long run economic growth. It did so not everywhere and all the time, but only where recipient countries followed good policies and had reasonable institutional environments for these policies to be effective.

Governments, nongovernmental organizations, donors, the press and civil society embraced this work with the hungry enthusiasm of the long-deprived. The research had the great virtues of plausibility and expediency -- a finding that aid is unconditionally good would have strained credulity. And by linking aid effectiveness to policies, the research gave intellectual justification to donors' practice of imposing "conditionality" on recipient governments. Tough love had found its intellectual savior.

Unfortunately, the Burnside-Dollar findings did not hold up to further scrutiny. Aid, after all, simply expands resources available to countries to build schools, hospitals and roads, and to pay teachers. These investments in human capital and infrastructure surely boost growth and improve living standards, the thinking went, even if there is some wastage of resources along the way through corruption or mismanagement. But as researchers pored over the data, it became increasingly difficult to maintain that there was any systematic relationship between aid and long-run economic growth.

The problem is that development and long-run growth are less about resources than about the environment for generating and sustaining private sector investment. Two key aspects of this environment are decent public institutions or governance -- the essential "software" for running a market economy, for creating rule of law and protecting property rights -- and incentives that encourage the private sector to export, especially manufactured products.

Aid, especially in large amounts, can damage governance and make an economy uncompetitive. Like revenues from natural resources, it is manna from heaven for governments. When governments receive large oil revenues or aid, they have less incentive to be accountable to their citizens, and governance suffers. In theory, donors impose an alternate form of accountability. In practice, donors' motivations are sometimes a mixture of the murky (think of the U.S. and Pakistan post 9/11 or the West and Zaire's Mobutu several decades ago) and the mindless (in 2000-02, the Tanzanian government reportedly had to write a few thousand reports to donors every quarter). Even where motivations are honorable, recipients have infinite ways of circumventing donor conditions.

Aid can also have adverse effects on an economy's competitiveness. When foreign resources come pouring in and are spent domestically, wages tend to rise, especially for those in scarce supply such as managers, supervisors and entrepreneurs. Factories that export will find themselves becoming uncompetitive and go out of business.

In research with Raghuram Rajan, we find that in countries that received more aid, exportable industries systematically underperformed. And exporting manufactured goods has been the mode of escape from underdevelopment in many of the East Asian successes. Is it a coincidence that, with rare exceptions (Mauritius), there are no booming clothing industries -- the launching pad for some of the East Asian miracles -- in aid-addled Africa? This despite the fact that clothing is only minimally demanding of infrastructure and entrepreneurship, and despite the very favorable access that Africa has always had in Western markets for exports of clothing products.

The new research findings have provoked genuine soul-searching among aid practitioners on the need to do things differently. But a new line, expressed most recently by the New York Times' Nicholas Kristof, goes something like this: So what if aid cannot do permanent good? The question is, can it do some good? Or, even if aid cannot promote livelihoods, can it save lives?

If it can, aid for a set of well-defined objectives, such as improving health and education, should continue to flow, and even increase substantially. There is ample evidence that foreign assistance helps fight disease in poor countries, documented most vividly in the Center for Global Development's "Millions Saved: Proven Successes in Global Health." So, why not do more of the same?

Because the fact that aid can save lives does not mean that aid might not have some of the adverse long-run effects relating to damaging governance and making the economy uncompetitive. Better health could be accompanied by slower growth, and hence reduced prospects for long-run prosperity. Even if the trade-off is worth making, it needs to be acknowledged. Aid advocates evade this by thinking and acting as if the long-run problems caused by aid can be fixed independently. That rarely happens.

Aid advocacy leads to perhaps an even more serious problem. There is a limited stock of good will and good intentions in the rich world and the question becomes whether this stock is best harnessed by mobilizing more aid, or by pursuing alternative actions that could have a bigger impact.

Consider a few: mobilizing more money to provide incentives for greater research and development devoted to addressing poor country health and agriculture problems (the green revolution in Asia was made possible by research on high-yielding varieties of wheat, and Africa hasn't had a similar revolution of its own); making regulatory changes in industrial countries that can reduce corruption (for example, more rigorous enforcement of bribery and corruption by rich country officials and corporations) in poor countries, which could have a huge impact on economic performance; or allowing more immigration from the poorest countries, which would directly benefit the poor.

These solutions are seldom pursued with the zeal that they deserve, in part because they are more difficult to support politically, and in part because that zeal which is essential to overcome the difficulties gets diverted toward, well, to calling for more aid.

Giving aid is like looking for the lost key under the lamppost because that is the easiest thing to do. But it is not obviously the most effective way that outsiders can help. When Ms. Jolie appears on the screen calling for more aid, she not only distracts our attention toward her obviously good looks, she may also be distracting our attention away from the search for more effective solutions to helping the poorest around the world.

Mr. Subramanian is a senior fellow at the Center for Global Development and at the Peterson Institute for International Economics.