With more than a billion people now living on less than a dollar a day, and with eight million dying each year because they are simply too poor to live, most would agree that the problem of global poverty is our greatest moral challenge.The large and pressing practical question is how best to address that challenge.Although millions of dollars flow to poor countries, the results are oftendisappointing.In Making Aid Work, Abhijit Banerjee--an "aid optimist"--argues thataid has much to contribute, but the lack of analysis about which programs reallywork causes considerable waste and inefficiency, which in turn fuels unwarrantedpessimism about the role of aid in fostering economic development.Banerjeechallenges aid donors to do better. Building on the model used to evaluate new drugsbefore they come on the market, he argues that donors should assess programs withfield experiments using randomized trials. In fact, he writes, given the number ofsuch experiments already undertaken, current levels of development assistance couldfocus entirely on programs with proven records of success in experimentalconditions.Responding to his challenge, leaders in the field--including NicholasStern, Raymond Offenheiser, Alice Amsden, Ruth Levine, Angus Deaton, andothers--question whether randomized trials are the most appropriate way to evaluatesuccess for all programs. They raise broader questions as well, about the importanceof aid for economic development and about the kinds of interventions (micro ormacro, political or economic) that will lead to real improvements in the lives ofpoor people around the world. With one in every six people now living in extremepoverty, getting it right is crucial.