Edward Miguel

This book is about ten years old, so most of the research done has been concluded and/or superseded. However I am happy that I read it at this juncture in time after moving to Kenya, because it has many historical references about this country.

The book makes a good case for the thesis that development is possible wherever governance is generally good and free of corruption. It cites many examples about the effect of corruption in Asia and Africa and gives examples of places where development and change worked despite the scourges of poverty and war (Vietnam and Botswana). There are also many examples of failure, like Iraq. The authors also give some evidence that economic hardships may be one of the underlying reasons behind civil wars, witch killings and even genocide.

I found the history, the stories and the economic experiments fascinating. This book, however would be more useful as reading material, as it often cites numbers and statistics that I would like to see and ponder black and white. These are hard to grasp fully when listened to.… (mais)

As Raymond Fisman and Edward Miguel explain at the beginning of their book, there are two main currents of thinking among those who opine on the wisdom of foreign aid: the "poverty trap" view, which holds that aid must be injected to end a vicious cycle in which inability to save leads to disaster in lean years, and the view that more such aid is simply sending good money after bad, straight into the hands of corrupt officials to be funneled away or otherwise wasted. Fisman and Miguel aim to look at corruption and violence in developing countries to determine how prevalent such evils are, how they are caused, and how they can be prevented--and, therefore, what the best way, non-ideologically-speaking, of raising up poor nations might be.

The funny thing about corruption is that it tends to exist out of sight--at least, out of sight of official statistics and public measurements. No one reports the bribes he takes on his income tax returns. So Fisman and Miguel have to come up with creative means of measuring corruption of various types, and this is the most fun part of their book. Economic Gangsters is completely accessible to the general reader, with virtually no economic jargon or concepts more difficult than "incentives matter," but it perfectly captures the exciting, puzzle-solving nature of this kind of academic research.

Fisman and Miguel's biggest, and most important, suggestion is the basic one that foreign aid and other solutions to developing-nation poverty be studied and implemented in an evidence-based manner. Without experimental data it's very difficult to determine whether a particular program is actually effective or not (or cost-effective or not). Randomized trials, like those carried out for developing medicines, are rare in the field of poverty reduction. But sometimes they are carried out. For example, local democratic control of public works projects is often touted as an antidote to corruption and skimming of funds. But in Indonesia a test was conducted to compare road building under local control, the thread of a federal audit, and no corruption prevention. Local control did little better than the control group, while those projects that were audited involved significantly less stolen money.

The authors adhere to their intention to remain non-ideological, and their interest is clearly in going where the evidence leads them. Unfortunately, large-scale economic experiments are often impossible and unethical, so some things can never be tested. But those interested in solutions that actually work should use what information they can. Economic Gangsters provides some of that information, and an interesting look at how to find it. It also tells some great stories about the incentives economic gangsters respond to, the strange circumstances that sometimes create these incentives, and how governments and other groups can play with them to aim for better outcomes.
(More at http://www.bibliographing.com/2008/10/06/economic-gangsters-by-raymond-fisman-an... )… (mais)