Agence France-Presse

wHEN President Bush promised to channel billions of dollars in aid — but only to countries that were committed to "ruling justly, investing in their people and encouraging economic freedom" — it was an extraordinary example of how an idea hatched by scholars could wend its way into mainstream politics.

The plan, which Mr. Bush announced as the Millennium Challenge Account at the 2002 United Nations International Conference on Financing for Development in Monterrey, Mexico, was largely based on the work of two economists, Craig Burnside and David Dollar. After completing an elaborate statistical analysis, they concluded that yes, foreign aid did promote economic growth, as long as the recipient government had solid fiscal, monetary and trade policies in place; their article, "Aid, Policies and Growth," was published in The American Economic Review in 2000.

But new research that uses the same methodology, though more up-to-date and comprehensive data, says those findings don't hold up. "We no longer find that aid promotes growth in good policy environments," write three researchers: William Easterly, a former principal economist at the World Bank who is now an economics professor at New York University; Ross Levine, a finance professor at the University of Minnesota; and David Roodman, a research associate at the Center for Global Development in Washington.

In a forthcoming article, also for The American Economic Review, these authors do not claim to prove that aid is ineffective, but they do argue that "policymakers should be less sanguine about concluding that foreign aid will boost growth in countries with good policies."

Because the new research has not yet been published, it is too soon to say just how influential it will be. But the conflicting studies highlight the often fraught and puzzling relationship between the world of professors and the world of politics. Why can one study trigger the flow of millions or even billions of dollars while another, equally valid work sits on the shelf? Often, the answer is that that is the way policymakers want it.

Mr. Burnside and Mr. Dollar's work caused a sensation when their paper first began circulating, in 1997. Ever since President Harry Truman, at his inauguration in 1949, announced a "bold new program" to alleviate world poverty, economists and others had been trying to figure out how to use aid to turn poor countries around. Here, finally, seemed to be the answer.

Soon, the British and the Canadian development agencies began saying that the idea that foreign aid reduced poverty in countries that were well governed was backed up by solid research. Most important, the Burnside-Dollar study became the cornerstone for a 1998 World Bank report that White House officials cite as the "key study" supporting the Millennium Challenge. Steven Radelet, who helped design the Millennium Challenge program as a deputy assistant secretary of the Treasury from 2000 to 2002, said that Mr. Burnside and Mr. Dollar's work was unusually influential, partly because it "resonated with an idea that a lot of development practitioners had come to believe, independent of the research."

Research that confirms what policymakers already believe is, not surprisingly, going to have greater impact than research that does not. As the economist Henry J. Aaron noted 25 years ago in his study of the Great Society programs, "Research reflects prevailing moods at least as much as it influences them."

Mr. Levine said of the Burnside and Dollar study: "It gave a reason for foreign aid and a strategy for giving it out. That's great for one paper. People grabbed this like I've never seen any other academic article grabbed before. It became the pretext for why the World Bank exists, and for the Millennium fund. One of the reasons for the World Bank is to funnel money to less developed countries and have them grow faster as a means for alleviating poverty. So it's nice to have evidence that suggests the money is linked to growth and hence the alleviation of poverty. Prior to the Burnside-Dollar finding, it was difficult to make that claim."

As Mr. Easterly chronicles in his book "The Elusive Quest for Growth: Economists' Adventures and Misadventures in the Tropics" (M.I.T. Press, 2001), for decades economists have been on an "audacious quest" for the magic formula that would make poor countries rich. Yet each time they thought they had found the elixir, he writes, they were disappointed.

Whether it was more investment, better technology, lower birth rates, broader education or debt relief, each once-fashionable theory of how to end third world poverty has ultimately fallen flat. Sub-Saharan Africa, the area most tended to by Western economists since their search began, has barely grown at all. Although some directed forms of aid have been highly effective — like health programs providing oral rehydration therapy or vaccinations for children — the key to the overall development puzzle has remained a mystery.

"When you're talking about development, you're talking about wholesale transformation of a society," Mr. Roodman said. "You're trying to recreate the history of the Western world over the last 200 years, to put other countries through that process. No one knows how to do that."

Even if someone did, it would be tough to prove on paper. Cross-country statistical studies may be inaccurate because, as Mr. Dollar pointed out, "there are a relatively small number of countries in the world" and so many potential variables. For one thing, aid is not always aimed at promoting growth. During the cold war, for example, much of American foreign aid was funneled toward American allies in the fight against Communism, regardless of its likely effect on development. Since the collapse of the Soviet Union, the amount of American foreign aid has declined, and its focus has shifted.

As Mr. Radelet, now a senior fellow at the Center for Global Development, noted: "Our two largest aid recipients are Israel and Egypt. The main reason for that aid is to support the Camp David Accords, not for economic development."

He continued, "In the last two years we've provided huge amounts of foreign aid to Pakistan, Jordan, Egypt and Turkey because they are frontline states in the war on terrorism."

Even when aid does seek to promote development, though, how recipients spend it is only part of the story. Mr. Easterly stressed problems on the donor end: "You have bloated, inefficient bureaucracies that don't really know what they're doing as far as achieving results," he said in an interview. He compared agencies like the United States Agency for International Development and the World Bank to "companies that get no customer feedback." The customers are the poor, and the poor often have little say in their own governments, let alone in international organizations, he said, adding, "It's as if Hollywood was making movies but never had any information about ticket sales."

So people look at the latest research — but only to a certain extent.

Mr. Burnside, who is now an economics professor at the University of Virginia, called the Easterly study "a significant challenge to our results" because it was the only one that used a very similar methodology. Still, he added, "the notion that among poor countries aid should be allocated to those with better institutions and policies has gained currency not because of any single statistical study, but rather because several different types of information all point in the same direction." (The two authors, who were both economists at the World Bank when they published their article, are working on a new paper responding to their critics.)

The Bush administration, for its part, does not seem particularly concerned about the latest findings. A senior White House official familiar with the new paper dismissed the dispute as "a statistical issue" that does not alter the soundness of the government's approach.

Few economists would disagree that policymakers should consider how governments are likely to use aid money. Dani Rodrik, professor of international political economy at the John F. Kennedy School of Government at Harvard, called the basic philosophy of the Millennium Challenge Account sound. "We have learned that attaching policy conditions to aid is futile when the recipient governments are uninterested in generating economic growth and poverty alleviation," he said.

But he added that there was still no magic formula for growth. China, India and Vietnam, for example, have all been economically successful in the past two decades, he said, but would not have passed the criteria set out by the Millennium Challenge Account. Trying a one-size-fits-all foreign aid policy "presumes not only that we have a fairly good idea what growth-promoting policies are, but also that these policies are the same everywhere," Mr. Rodrik said. "Neither of these presumptions is supported by evidence."

Copyright 2003 The New York Times Company. Reprinted From The New York Times, Arts & Ideas, of July 26, 2003.