Don't tell anyone, but a dirty little secret within the foreign aid world is that aid often doesn't work very well.

Now that truth has been aired (and sometimes exaggerated) in a provocative new book by William Easterly, "The White Man's Burden." Mr. Easterly, a former World Bank official who is now an economics professor at New York University, has tossed a hand grenade at the world's bleeding hearts — and, worst of all, he makes some valid points.

Let me say right off that stingy Republicans should not read this book. It might inflame their worst suspicions.

But the rest of us should read it, because there is a growing constituency for fighting global poverty, and we need to figure out how to make that money more effective.

I disagree with many of Professor Easterly's arguments, but he's right about one central reality: helping people can be much harder than it looks. When people are chronically hungry, for example, shipping in food can actually make things worse, because the imported food lowers prices and thus discourages farmers from planting in the next season. (That's why the United Nations, when spending aid money, tries to buy food in the region rather than import it.)

On one of my last trips to Darfur, I had dinner at a restaurant in Nyala called K2. Out back were 18 big white S.U.V.'s belonging to the U.N. and aid groups; that amounted to nearly $1 million worth of vehicles, in a country where people are starving.

The aid workers are struggling heroically in a dangerous and difficult place, and I don't begrudge them reliable vehicles. But something seems wrong when international agencies are more successful at maintaining S.U.V.'s than clinics. (One reason is that budgeting is often done annually, and one of the ways to spend a grant in a single year is to buy a vehicle.)

It's well-known that the countries that have succeeded best in lifting people out of poverty (China, Singapore, Malaysia) have received minimal aid, while many that have been flooded with aid (Niger, Togo, Zambia) have ended up poorer. Thus many economists accept that aid doesn't generally help poor countries grow, but argue that it does stimulate growth in poor countries with good governance. That was the conclusion of a study in 2000 by Craig Burnside and David Dollar.

Professor Easterly repeated that study, using a larger pool of data, and — alas — found no improvement even in countries with good governance.

Saddest of all, Raghuram Rajan and Arvind Subramanian of the International Monetary Fund have found that "aid inflows have systematic adverse effects on a country's competitiveness." One problem is that aid pushes up the local exchange rate, discouraging local manufacturing. Mr. Subramanian also argues that aid income can create the same kinds of problems as oil income — that famous "oil curse" — by breeding dependency and undermining local institutions.

All these findings can ... [ Read More (0.2k in body) ]