Reply to comment

One comment on this very desirable idea. Recently, we had a similar discussion with some colleagues and someone wrote "I have generally found that evidence is used only where it can support the current or proposed policy of the government or institution, rather than the other way around..."

This is of course not to argue for ignorance; but to try to understand why is that we do not see more appetite for ex-ante assessment of projects’ impact in current documents, or why we see superficial and low quality projects’ appraisals.

I think we need to realize that by its nature the Bank is not only a knowledge institution, but a sort of cooperative for countries, with a basic function of responding to the financing needs of its members. There is also pressure from competitors around. I find this dichotomy at the heart of this discussion, as sometimes projects respond to a particular demand from the country and not directly to the best evidence available.

Perhaps one solution is to construct a project approval decision process in two layers, the technical first and then political one, without trying to artificially supplant one or the other. If this was possible we could be more sincere in terms of the evidence and avoid project documents that selectively choose it or just avoid any evidence. And on a second stage the board could frankly approve projects on the basis of strategic, political or practical considerations notwithstanding what existing evidence points at, with a written note saying so. This would at least increase transparency of the decision-making process and reveal other inputs and incentives that decision-makers have, without punishing but simply accepting this is a reality of our type of “cooperative” institution.