The Bottom Billion by Paul Collier

I'm long overdue in blogging this book, as I read it a couple weeks ago, so more of it than I'd like has probably already slipped from my mind. The book is about the author's studies of development strategies for countries whose populations are among the "bottom billion" in countries that are in various traps that cause them to stagnate. He goes through these traps in convincing detail. Notable among these traps are bad governance, conflict, civil war, natural resources, bad neighbors, and being landlocked. Some of these are rather obvious, but others require a little thought before seeing why they cause a problem. For example, natural resources might at first seem to be a good thing to have, but in many countries, they have thwarted widespread development, cause power to become concentrated, and inspire coup after coup after coup. Being landlocked prevents trade, which, in the long run, is a key to most societies' development (just look at what has happened throughout China's history for lots of examples of "trade = growth, no trade = stagnation").

Of particular interest to me in this book is that Collier (along with others) has done a number of empirical studies about different types of aid, different types of military interventions, etc. As you might imagine, these haven't been terribly popular with this or that aid agency whose pet theories are disputed by his research. But, I had the same feeling reading this book that I did reading Gawande's observations on medicine: replacing pet theories with empirical studies will dramatically improve results. If I have time today, I'll expound on that point. Regardless, for those of you who want to save the poor of the world as your life's work, read this book. It will challenge a lot of notions, both traditional right wing notions (such as "aid doesn't work") and traditional left wing notions (such as "capitalism is bad").

As for recipes for success, he has quite a few, but most of them seem to me more or less pipe-dream variations of "get everybody to agree on <whatever>". There are probably a few smaller ideas that can be easily implemented -- for example, aid to post-war societies shouldn't be "a bunch of aid in the first two years, tapering off quickly thereafter", but instead "little aid in the first two years, as the societies need to develop institutions to handle the aid, then plenty of aid for a significant time thereafter". But a lot of the bigger ones, notably international charters (good luck) on military intervention and trade (Doha, anyone?) seem much harder.