From http://rodrik.typepad.com/dani_rodriks_weblog/2007/05/can_anyone_be_i.html:
"... I am not sure that it is good policy for the Bank to prioritize corruption--as a rule--over other problems that developing nations face. As I have stressed in my work with Ricardo Hausmann and Andres Velasco, the binding constraint on growth differs from country to country. In some cases (Zimbabwe?), governance problems are indeed the most serious binding constraint. In many others, the problems lie elsewhere--in low savings, poorly functioning financial markets, low entrepreneurship, poor infrastructure, and myriad other syndromes of underdevelopment.
Let me make a bolder claim. A development strategy that focused on anti-corruption in China would not have produced anything like the growth rate that this country has experienced since 1978, nor would it have resulted in 400 million plus fewer people in extreme poverty."
FROM http://knowledge.wharton.upenn.edu/india/article.cfm?articleid=4192:
"Somebody has to watch, and someone has to watch the watchers. But when does breathing down the workers' necks get in the way of the work? That is a problem for policy makers who want to deter corruption but don't want the remedy to be worse than the ailment. For insight into these tradeoffs, Shawn Cole of the Harvard Business School and two colleagues looked at Indian banks, which use aggressive monitoring and severe penalties to keep lending officers honest.
The research results are described in the paper, "Are the Monitors Over-Monitored? Evidence from Corruption and Lending in Indian Banks," co-authored by Abhijit Banerjee and Esther Duflo, who are at the Massachusetts Institute of Technology. "We find evidence that vigilance activities result in reduced lending. The amount of credit declines sharply at the affected bank branch, as well as neighboring branches," the researchers report. "This effect is economically and statistically significant, persisting up to two years." "
