
He does not argue that aid can never be effective. He argues that the way we have gone about it has often been ineffective and proposes some ways to help it be more effective. He cites Botswana as an example of aid "supporting reform and good government," though detailing how abysmally it has failed to do the same elsewhere; he cites pages and pages of World Bank, WHO, and IMF programs that were successful.
He does not claim that we should give up trying to help people. He pleads that we change the question from How can I use foreign aid to solve every problem in the world to "What can foreign aid do for poor people" and then do it. "Aid agencies cannot end world poverty, but they can do many useful things to meet the desparate needs of the poor and give them new opportunities."
He does not provide a blueprint for how to solve every problem with aid. He does suggest several dimensions where significant progress could be made by experimentation in making aid more effective. They include:
Accountability and Specialization "The utopian agenda has led to collective responsibility for multiple goals for each agency, one of the worst incentive systems invented since mankind started walking upright.... Have individual accountability for individual tasks. Let aid agencies sepcialize in the sectors and countries they are best at helping. Then hold aid agencies accountable for their results by having truly independent evaluation of their efforts." "Not overall sweeping evaluations of a whole nationwide development program, but specific and continuous evaluation of particular interventions."
Feedback and Meaningful Participation from Recipients. "The needs of the rich get met because the rich give feedback to political and economic Searchers, and they can hold the Searchers accountable for following through with specific actions. The needs of the poor don't get met because the poor have little money or political power with which to make their needs known and they cannot hold anyone accountable to meet those needs." The jargon is already there, but he takes down Poverty Reduction Strategy Papers as one set of planners talking to another set of planners, often written by the recipient government at the dictation of the donors, sometimes even undermining democratic processes and selections. "Visibility gives more power to Searchers while invisibility shifts power to Planners. The problem with aid is that the poor are mostly invisible." "Participation should mean more buying and voting power in the hands of the poor in aid."

On Peacekeeping: "Peacekeeping could be good, but just who is willing to be accountable for its success or failure. ... Interventionism suffers from the patronizing assumption that only the West can keep the locals from killing eah other. ... Peace usually succeeds war because of a decisive victory by one side, not because of negotiated settlements by outsiders. ... UN interventions produced a stable peace only a quarter of the time. With no intervention, a stable peace resulted nearly half the time." He is overall quite negative on military interventionism.
One thing that I have admired about Easterly is that he is willing to admit that he was wrong. He does so at least twice in the book, once discussing market reform "shock therapy" and once in discussing structural adjustment. I asked him at his blog whether some of his own work fell under a criticism he had just made, and he admitted that one of the papers I mentioned fell into the same category. He promised a bit more of a response which I'm still waiting for, but there's something comforting in talking with someone you know is willing to say "I was wrong." In the book he opined that the military would be the least responsive to criticism, but on his blog has noted with surprise that the military responded the most cordially and promisingly of any group he has criticized.
One of my few criticisms is that he simultaneously calls for more observability while deriding current aid projects for overly focusing on observability without differentiating what should be observable. That is, he wants to trim down the number of goals any agency is responsible for, so it is more observable who does what, but he laments incentives that support building facilities or roads because they are more visible than textbooks or maintenance.
After presenting a fairly convincing case that Planning really doesn't work, I also noted with some irony that there was a heavy element of planning involved in the homegrown development cases he lauds at the end of the book. He acknowledges some of this, that "the success stories follow a variety of formulas," but prefers to call South Korea and Japan's actions intervening in economies rather than planning and to note that China and Singapore are "quite far from a laissez-faire model." "What we do know ... is that the West played small part in them." This leaves open a door for domestic planning to be good and foreign planning to be bad, a door I think he could do a much better job turning into a window or a peephole.
"Success attracts paternity claims." So every theory of development claims success for the various successes. When Sachs visited Cornell, for instance, he laughed at Easterly's assertion that aid hasn't done much and cited the total figure of aid given to China and India. Easterly notes that those large sums only amount to $0.001 per day for each Chinese person and $0.005 for each Indian person. Compare that to the much larger sums spent in Africa (which Sachs calls a mere $0.50 per person) and you wonder how so little could do so much while so much does so little.
Some very good 'side note' sections: The importance of social networks (p. 82-87), property rights (p. 90-99), the difficulties of making democracy work and its evolution (p. 116-129), comparing the cost effectiveness of health initiaives (HIV prevention, AIDS treatment, and others; p. 249-258). He also has some excellent snapshots of World Bank publications from the 1950s or 1980s to today, all saying the exact same thing: things are bad, but there are promising signs out there. I learned that it is much easier for me to be sanguine when reading about the colonial disasters of England, France, and Portugal than to read about the (usually CIA-backed) disasters the US sponsored (270-338).
Notable Quotables after the break: